Infectious diseases

Other flukes

Although around a dozen other flukes occur in humans, only a few affect large numbers of people. One such parasite is Clonorchis sinensis, the Chinese or Oriental liver-fluke. Adult flukes of this species are typically 10-25 mm in length and occur in the bile duct (in the liver), where they feed on blood and can survive for around 25 years. Their life cycle is identical to that represented earlier in Figures 8.1a and 8.2, and when cercariae are released from a freshwater snail (the primary intermediate host) they penetrate into fish such as carp or minnow (the second intermediate host) and encyst as metacercariae. In the Far East, especially China and Thailand, raw or lightly-cooked, salted, pickled or smoked freshwater fish is a delicacy. But such culinary habits increase the risk of ingesting the living parasite, and infection by a few flukes can cause serious liver malfunction and, in cases of heavy infection, even death. In many areas the life cycle of Clonorchis sinensis is maintained in carnivores both wild and domestic, so cats and dogs can be infected and thus act as reservoir hosts. Interestingly, the snail hosts of Clonorchis are species of Oncomelania, the same vectors as for S. japonicum, suggesting that a better understanding of this snail might yield a way to interrupt the life cycle of several fluke parasites. Other liver-flukes, belonging to the genus Fasciolopsis, are large flukes (up to 75 mm in length and 25 mm in width) that can occupy the human bile duct. Sporocysts, rediae and cercariae occur in freshwater snails but, on release, the cercariae attach to water plants such as water chestnut, lotus and bamboo. If the plants are then ingested uncooked by the definitive host they excyst, and the cyst releases its metacercaria in the intestine. The metacercariae then migrate to the bile duct, where they mature. Fasciolopsis is endemic in South East Asia, including parts of India and the South West Pacific, where it causes profuse diarrhoea, duodenal blockage, generalised toxaemia and sometimes death. • What is the main difference between the life cycles of Fasciolopsis and Clonorchis? • No intermediate host is involved for the metacercaria stage in the Fasciolopsis life cycle, whereas Clonorchis uses snails and fish as intermediate hosts. As such, there are no reservoir hosts for Fasciolopsis. A closely related fluke to Fasciolopsis, Fasciola hepatica (Figure 8.15), is found worldwide and is a common liver-fluke of sheep in the wetter areas of the UK. The metacercaria often encysts on watercress leaves and every year a number of localised infections of humans occur as a result of eating wild, and insufficiently washed, watercress. Finally, the Oriental lung-fluke Paragonimus westermani is a small parasite measuring 7.5-12 mm in length, which occurs as an adult in the lungs of mammals in Asia. It is found as a parasite in humans in those areas where there is high consumption of salted, but uncooked crabs. Reservoir hosts include any carnivore that preys on crab or crayfish. • What part in the life cycle do you suspect the crab might play? • It is the second intermediate host, containing the metacercaria larvae, which would have reached there from the snail intermediate host. The metacercariae are often in cysts in the crab and a common route of infection occurs when crabs are prepared on a surface: cysts are released and contaminate salad vegetables that are also being prepared, which are then eaten uncooked. Paragonomiasis is a dangerous disease. Within the lungs, cysts and adhesions occur, resulting in long-lasting chronic pulmonary problems. Cysts sometimes end up in other tissues, including the brain, causing epilepsy and meningitis. Diagnosis of infection by the parasites described above is the same as for blood flukes, namely the presence of eggs in stool samples. Again, the drug of choice for treating all the above flukes is PZQ.

BSE

BSE was first detected in UK cattle in 1986, but it is now believed to have been infecting cattle as long ago as the 1970s and early 1980s. It can only be speculated why these early cases were missed. The disease was unknown at the time, and may have occurred in isolated cases, or the infected cattle may still have been incubating the disease at the time of slaughter. It was the practice at the time to recycle abattoir wastes into animal feeds. This practice may well seem abhorrent now, but it was previously considered by some to be a cost-effective way of using material unsuitable for human consumption. Originally, it was believed that material in cattle feed made from scrapie-infected sheep was the source of infection. This hypothesis has since been rejected, and the source of infection is now thought to have been a novel prion mutation in cattle, or sheep. This novel TSE agent is believed to have emerged in the early 1970s and gradually spread through cattle stock via recycled bovine material in their food. BSE was first recognised in dairy cows, and by 1992 around 1% of all adult cattle in the UK were infected (181 376 cases in total by 2002). The disease did affect cattle in other European countries, particularly Switzerland and Ireland, with 3679 cases across Europe by 2002, but the scale of these epidemics was nowhere near as large as that in the UK. A handful of cases were also reported in Canada. The UK government's response to the BSE crisis was slow because it was widely believed that BSE posed no risk to human health. Ministers, officials and the scientific advisory committees were all more concerned that the public would be alarmed by BSE and an economically damaging food scare would result. It wasn't until 1988 that BSE was made a notifiable animal disease and it was made illegal to feed ruminant-derived protein to ruminants in the UK. This removed the main source of infection and was successful in the long term. However, it wasn't until September 1990 that specific cattle tissues such as brain and spinal cord were banned for consumption by humans, livestock and poultry. Worryingly, it is estimated that before 1995, nearly one million BSE-infected cattle may have ended up in the human food chain. In 1993, at the peak of the epidemic, 1000 new cases of BSE were being reported every week, but by April 1996 the number of BSE cases reported had shrunk to less than 1200 a month. The control measures taken in the UK were severe and involved isolation of affected farms and slaughtering of any herd containing an infected animal (it is estimated that up to 4.4 million animals were slaughtered during the eradication program). The first diagnosis of what turned out to be vCJD was in the UK in 1994, although the disease was not officially recognised until 1996. At that point, the UK government admitted that BSE had in all likelihood been transmitted to humans as vCJD. Suspicions that BSE might cause vCJD in humans were first raised by the strong geographical and temporal association between these diseases, and subsequently by biological and genetic tests. • Nearly all cases of vCJD had come from the UK (cases of vCJD between 1996-2002: UK 129; France 6; Canada, Ireland, Italy and the USA one each (WHO, 2002). France imported beef products from the UK at the time, whilst the Canadian and USA cases had lived in the UK at the time of the BSE crisis. • Macaque monkeys inoculated with BSE develop pathological features that resemble vCJD, and mice inoculated with either BSE or vCJD show entirely similar brain lesions (WHO, 2002). • Inherited forms of BSE and CJD both show identical mutations in the PrPc gene, suggesting that the same misfolded protein could be disease-causing in both species. In January 1998, the BSE Inquiry was set up by the Government to analyse the events surrounding the emergence of BSE and vCJD. It concluded that government policy decisions such as the ruminant feed ban were appropriate, but had often been taken too late and not enforced rigorously enough. By the autumn of 2011, 175 people had died of vCJD (NCJDRSU, 2011). • If people were infected before 1994, roughly when might vCJD symptoms appear? • TSEs are slow infections, so a peak of disease might not be apparent until 2020-2040.

Drug resistance

Developing new pharmaceuticals to halt the spread of drug resistance in M. tuberculosis presents a major challenge to public health - not least because of the huge number of people worldwide who are infected with drug-susceptible bacteria that could evolve drug resistance. Approximately two billion people (one-third of the global population) are already infected with M. tuberculosis. The WHO calculates that in the 20 years to 2020, almost one billion people will be newly infected with M. tuberculosis, 200 million will get sick and 35 million will die from TB, unless control measures are significantly improved (WHO, 2010a). The scale of the TB epidemic and the need for new drugs to control it should not be underestimated. Many healthcare-associated infections (HCAIs) acquired by patients in hospitals, nursing homes and other healthcare facilities are due to drug-susceptible pathogens. Here we focus on the problem of drug-resistant strains. The European Centre for Disease Prevention and Control has estimated that over four million patients acquire HCAIs every year in the countries of the European Union, resulting in at least 37 000 deaths. The majority of these deaths are due to a bacteraemia or pneumonia due to the drug-resistant bacteria listed in Box 6.1, and to diarrhoea caused by Clostridium difficile. Drug-susceptible and drug-resistant Staphylococcus aureus and Enterobacter species also cause post-operative surgical site infections. Drug-resistant HCAIs have become a major topic of concern in European and North American hospitals, with every outbreak generating a report to a central registry, adverse publicity in the news media and claims for compensation. In Unit 1 of this block (Section 5.2.2) we reviewed the progress that has been made in recent years in reducing MRSA and C. difficile infections in National Health Service (NHS) hospitals in England (see Figure 1.13 in that section). By contrast, relatively little is known about the prevalence of drug-resistant HCAIs in low- and middle-income countries (LMICs), but some clues about the extent of the problem are beginning to emerge. A report by Rosenthal et al. (2010) found striking evidence of resistance in device-associated bacterial HCAIs in patients in the intensive care units of 172 hospitals in Central and South America, Asia and Africa and one in Lithuania - all in countries with membership of the International Infection Control Consortium (INICC). Device-associated infections enter the patient via an indwelling respiration tube, urinary catheter, intravenous blood line, or other invasive device. Table 6.1 compares the frequency of bacterial resistance in device-associated infections in intensive care units (ICUs) in INICC hospitals with the rate in ICUs in the USA. The extent to which drug-resistant device-associated HCAIs occur in intensive care units in the USA is cause enough for concern, but it is clear from Table 6.1 that bacterial drug resistance in HCAIs is an even greater problem in poorer countries. As researchers begin to turn their attention to healthcare facilities in less developed parts of the world, a picture of accelerating rates of drug resistance is emerging. It is fuelled in part by technical difficulties in the early diagnosis of drug resistance in patients who may have multiple infections and malnutrition, and the frequent shortage of expensive antibacterial drugs with which to treat them effectively. A further problem is the widespread availability of antimicrobial drugs sold 'over the counter' or bought via the internet without prescription in almost every country, including some in Europe and parts of North America. Antibiotics are among the most commonly purchased drugs and include those against tuberculosis (Morgan et al., 2011). The effect of self-treatment with privately purchased drugs on the emergence of antibiotic resistance cannot be quantified, but is assumed to be significant. An additional cause of antibiotic resistance is the widespread sale of counterfeit drugs in LMICs, containing insufficient, incorrect or no active ingredients (WHO, 2010b). Finally, there is growing concern about the speed at which infected individuals can transport drug-resistant pathogens to distant locations, for example when refugees, tourists, civilian or military personnel travel between countries, or as a consequence of medical repatriation (Rogers et al. 2011).

Estimation from epidemiological data

From earlier sections in this unit you know that for infections in an endemic steady state in large, homogeneously mixing populations, the basic reproduction number, R0, can be estimated, provided that either the average proportion susceptible, S, or the average age at infection, A, is known, along with the life expectancy, L. The life expectancy (and the age structure of the population) can often be obtained from published data for the population, or estimated from separate surveys. The proportion susceptible and the average age at infection, however, must be estimated from epidemiological data. This can be done using data from a serological survey, as demonstrated in the next section. 3.1 Estimation from serological survey data Table 4.1 shows data from a serological survey of hepatitis A virus infection in Bulgaria. As in many surveys, only a small proportion of the whole population was surveyed. Provided that the sample is representative, however, you can draw inferences about the population as a whole from this sample. Table 4.1 Numbers tested and positive for hepatitis A antibodies, Bulgaria Age group/years Tested Positive 1-10 135 46 11-20 160 66 21-30 133 97 31-40 116 94 41-50 101 96 51-60 101 96 61-70 61 59 Hepatitis A virus is transmitted by the faecal-oral route, through direct contact or the ingestion of contaminated food. The data are aggregated in seven 10-year age groups. Table 4.1 gives the numbers tested in each age group, and the numbers positive, that is, with IgG antibodies to hepatitis A virus. (Note that in some infections, such as influenza, the presence of antibodies may not indicate immunity.) For hepatitis A virus, a positive test result indicates immunity. Thus the proportions positive in each age group provide an estimate of the proportion that is immune, as you will see in the next section. 3.1.1 Calculating proportion susceptibility and R0 Susceptibility within each age group In order to estimate S, the overall proportion of the population that is susceptible to hepatitis A virus infection, the first step is to calculate the proportion susceptible within each age group. To do this, first calculate the proportion that is immune. For example, the proportion of 1-10-year-olds who are immune is: It then follows that the proportion of susceptibles in the 1-10 age group is: 1 - 0.341 = 0.659 Table 4.2 shows the proportions immune and susceptible within each 10-year age group in this survey. Table 4.2 Proportions immune and susceptible to hepatitis A virus by age, Bulgaria Age group/years Proportion immune Proportion susceptible 1-10 0.341 0.659 11-20 0.413 0.587 21-30 0.729 0.271 31-40 0.810 0.190 41-50 0.950 0.050 51-60 0.950 0.050 61-70 0.967 0.033 Note that the data for infants (i.e. those less than one year of age) are missing; you should therefore assume that all infants are protected by maternal antibodies and hence the proportion susceptible among them is zero. This assumption makes the calculations easier but is probably incorrect; however, it has little bearing on the final result since infants account for only a small proportion of the population. • Check that you can derive the values in Table 4.2 from those in Table 4.1. What do you notice about the age trend in the proportion immune? How might you explain this? What assumption have you made? • The immune proportion increases with age. The reason for this is that older people have had more time to become infected than younger people. Note that this interpretation relies rather heavily on the assumption that hepatitis A virus infection is in an endemic steady state in Bulgaria. An alternative explanation is that the force of hepatitis A infection has declined over time, and that the high proportions immune at older ages reflect past exposures to hepatitis at a time of high incidence. Susceptibility within the whole population Table 4.2 gives the proportions susceptible within each age band. In order to obtain S, the overall proportion of the population susceptible, you need to calculate a weighted average of these age-specific proportions, with weights proportional to the size of each age group within the population. This statistical measure is needed because surveys are not random samples of the population: you cannot be sure that the proportion tested in each age group reflects the proportion of the population in that age group. If you do this, you can calculate that the proportion susceptible in this population is 0.259. (See Box 4.3 for details of the calculation, if you are interested.) Calculating the basic reproduction number Given an estimate of the proportion susceptible, S, you can now calculate the basic reproduction number for hepatitis A virus infection in this population. • Calculate the value of R0 for hepatitis A in Bulgaria. What assumptions have you made? • Thus, if the population were totally susceptible, one infectious case would infect, on average, about 4 others. The main assumptions behind this calculation are that the population mixes homogeneously, and that the infection is in a steady state. Box 4.3 (Optional) Calculation of the proportion susceptible, S We shall assume that the age distribution is broadly rectangular with life expectancy of 71 years, and that the population size is constant. With these assumptions: • the proportion of the population in each 10-year age band is 10/71 • the proportion of infants is 1/71, although remember our earlier assumption that all infants are protected by maternal antibodies and thus have a proportion susceptibility of zero. The proportion susceptibility for the whole population is therefore the sum of the proportion susceptible for each age group multiplied by the proportion of the total population that each age group represents: This can be simplified as: 3.2 Estimation from case reports The average age at infection, A, and hence R0, can in some cases also be estimated from case reports, provided that the proportion of cases that are reported does not vary with age (so that, for example, a case in an adult is as likely to get reported as a case in a child). This is a reasonable assumption for measles, but not for infections for which the proportion of infections that produce symptoms (and hence are likely to be reported) varies with age. For example, infection by hepatitis A virus is much more likely to be symptomatic (and hence reported) in adults than in children. To illustrate this idea, Table 4.3 shows the total number of notifications of measles by broad age group in England and Wales between 1956 and 1965, namely prior to the introduction of vaccination. Table 4.3 Measles notifications by age, England and Wales, 1956-1965 Age group/years Notifications 0 221 267 1 424 616 2 544 251 3 580 855 4 595 364 5-9 1 868 340 10-14 109 937 15-24 26 790 25+ 15 855 All 4 387 275 Provided that the proportion of measles infections that are notified does not vary with age, the proportions of notifications in each age group represent the probability of acquiring infection within that age group. For example, the proportion of all the infections that occurred within the 5-9-year age group is: or about 42.6%. These infections occurred on average at 7.5 years of age, the midpoint of the 5-9-year age group. Table 4.4 shows the distribution of age at infection, together with the age group midpoints. The midpoint of the 25+ age group is calculated assuming that everybody lives to 75 and then dies. Table 4.4 Distribution of the age at measles infection Age group Midpoint/years Proportion of infections 0 0.5 0.0504 1 1.5 0.0968 2 2.5 0.1241 3 3.5 0.1324 4 4.5 0.1357 5-9 7.5 0.4259 10-14 12.5 0.0251 15-24 20.0 0.0061 25+ 50.0 0.0036 • What is the peak age at measles infection? (Hint: you need to take account of the age group width.) • Although about 42.6% of infections (rounded up from 0.4259 in Table 4.4) occur in the 5-9-year age group, this spans 5 years so the average proportion per year is 8.52%, although it is probably higher at age 5 than at age 9 (since fewer children are immune at age 5 than at age 9). So the peak age at infection is probably 4 or 5 years. The average age at infection, A, is then calculated as the weighted average of the midpoints, the weights being the proportions of infections occurring in each age group. For these data, the average age at infection is 5.4 years. (The details of the calculation are shown in Box 4.4, if you are interested.) • Calculate the value of R0 in this population using a rectangular age distribution with L = 75 years. What assumptions have you made? • So, if the population were entirely susceptible, a single measles case would infect, on average, about 14 others. This calculation assumes homogeneous mixing, and that the infection is in an endemic steady state. Box 4.4 (Optional) Calculation of the average age at infection The calculation of the average age of infection, A, in this example is based on three assumptions: • in this population, everyone will eventually get infected with measles • the age distribution in this population is rectangular, everyone dying at 75 years • the proportion of cases of measles reported does not vary with age. With these assumptions, A is calculated by multiplying the midpoints of each age group by the proportion of infections for each age group and then summing all the values: Study note So far in this unit you have dealt primarily with infections in an endemic steady state. This state is characterised by regular fluctuations around a long-term average. This average endemic level was used to derive simple expressions for key epidemiological parameters, which can be estimated from surveillance data. In the next two sections attention will be focused on the fluctuations around the long-term average level.

Keratitis

Inflammation of the cornea, usually caused by microbial infection of the eye; over 50 bacteria, yeasts, moulds and the protist Acanthamoeba are known to cause keratitis, which can occur after eye injuries and in contact-lens wearers.

Dysentery

Inflammation of the gut that produces profuse bloody diarrhoea that may contain pus and mucus, and be accompanied by abdominal cramps.

Herd immunity level

The proportion of a population that is immune.

Odds ratio

The odds for one category of individuals (e.g. a treated or an exposed group) divided by the odds for a second category (e.g. a control or an unexposed group).

The Red Queen hypothesis

The lecture on antibiotic resistance by Neil Woodford of the UK's Health Protection Agency introduced the mechanisms that enable bacteria to evolve resistance to antibiotics, or acquire resistance gene elements transferred from other bacteria. • The 'Red Queen' hypothesis, to explain why the evolution of bacterial resistance to every new antibiotic is (in his words) 'inevitable'. Very rapid bacterial replication underlies their ability to evolve resistance under the selection pressure exerted by chemical treatments, and they can do this much faster than the time it takes scientists to develop new chemical agents. Like the Red Queen in Lewis Carroll's Through the Looking Glass (published in 1871) 'it takes all the running we can do to keep in the same place'. The speed of bacterial replication is illustrated by Escherichia coli, which can replicate by binary fission in about 30 minutes under ideal conditions. Viruses, some parasites and the invertebrate vectors of infectious organisms also 'outrun' the pace of human ingenuity at devising new chemical agents. For example, drug resistance in Schistosoma parasites (see Figure 6.1) is a major limiting factor in controlling a disease that affects more than 230 million people worldwide (WHO, 2012a). The mechanisms that enable antimicrobial resistance to emerge are primarily driven by suboptimal chemical treatment with antibiotics, antivirals, antifungals or anti-parasite drugs, or externally applied toxic agents, and their indiscriminate use. This can result in the survival of the most resistant variants in the pathogen or vector population, while reducing the numbers of competing variants and creating ideal conditions for the more resistant individuals to replicate. Over time, resistant organisms form an increasing proportion of the population. Resistant vectors can migrate to new habitats and resistant pathogens are readily passed on to susceptible contacts. The cycle of indiscriminate suboptimal use of antimicrobial or anti-vector chemicals selects variants with ever greater resistance every time it is repeated, until the agent has little or no effect against its target organisms. Three examples illustrate the threats posed by resistant organisms: multidrug-resistant tuberculosis (MDR-TB), drug resistance in pathogens causing healthcare-associated infections (HCAIs), and chemical resistance among disease vectors. A rapidly increasing global health problem is posed by multidrug-resistant (MDR) strains of Mycobacterium tuberculosis. MDR mycobacteria are resistant to at least the two most widely used drugs, isoniazid and rifampicin. The WHO's 2010 global status report on Multidrug and Extensively Drug-Resistant TB estimates that there were around 440 000 new cases of MDR-TB in 2008, causing 150 000 deaths worldwide (WHO, 2010a). Almost 50% of all cases occurred in two countries - China and India - but there were significant burdens of MDR-TB in at least 27 countries, mainly in Eastern Europe. The extent to which suboptimal TB treatment drives the evolution of bacterial resistance can be judged from the experience in China. The first ever Chinese national survey, cited in the aforementioned WHO report, found that 5.7% of all new cases of TB were MDR, amounting to an incidence of 100 000 new MDR-TB cases in China annually (WHO, 2010a). Significantly, 25.6% of all previously treated patients were found to have MDR-TB. • What features of the treatment regimen for tuberculosis contribute to the underlying causes of drug resistance? Answer The Tuberculosis Case Study (Section 6.6) referred to the large numbers of tablets that TB patients have to take at regular intervals every day for at least six months, which is reflected in Figure 6.3. Such a treatment regimen can prove hard to maintain, particularly in patients who have difficulty in accessing medication regularly, or in sustaining their commitment to treatment due to major problems in their social circumstances or lifestyles. Incomplete or intermittent treatment drives the evolution of drug resistance. The evolution of drug resistance does not stand still. Figure 6.4 shows the distribution of countries that had reported at least one extensively drug-resistant (XDR) TB case to the WHO by the end of 2010. The mycobacteria infecting such patients are not susceptible to isoniazid or rifampicin, or any of the fluoroquinolone antibiotics or injectable anti-TB drugs. In 2009, the first totally drug-resistant (TDR-TB) cases were reported in Iran (Velayati et al., 2009) and small numbers have since been detected elsewhere, including in India. None of the 12 known anti-TB drugs have any efficacy in TDR-TB patients.

Evolution of transmission mechanisms

Throughout this block you have learned about the importance of the transmission of pathogens. In this unit, you have also considered the importance of transmission with regard to virulence. It is therefore appropriate to spend some time discussing evolutionary aspects of transmission, focusing on two transmission types: host-to-host contact (specifically, via sexual contact), and vector-borne diseases. 4.1 Sexually transmitted infections Sexually transmitted infections (STIs) are caused by pathogens whose principal means of transmission from one host to another is sexual intercourse by the host. This is a somewhat loose definition; some STIs can be transmitted in other ways (e.g. HIV by sharing hypodermic needles or blood transfusion). In addition to horizontal transmission during sex, many STIs are also transmitted vertically, either to the foetus in the uterus, or during birth, as the baby passes through the infected vagina. While more than 20 pathogens may be transmitted during sex in humans, we are only beginning to appreciate the diversity of STIs among other host species. 4.1.1 Diversity of human STIs Examples of the diversity of human STIs are given in Table 5.2. They typically cause chronic symptoms, for reasons that are explained later, and, with the exception of HIV, cause relatively low mortality. In 2010, 34 million people were infected with HIV, with 3 million new cases occurring in that year (WHO, 2011). Syphilis spirochaete bacterium Treponema pallidum estimated 24 million new cases worldwide in 1999 can cause abortion, premature delivery, stillbirth Gonorrhoea coccal bacterium Neisseria gonorrhoeae estimated 62 million new cases worldwide in 1999 common cause of pelvic inflammatory disease with subsequent risk of infertility; vertical transmission to infants causes blindness Genital herpes herpes simplex virus 2 (HSV2) approximately one-fifth of US adult population infected in late 20th century closely related HSV1 causes oral herpes (cold sores) AIDS HIV estimated that about 5 million new cases occur worldwide every year see above Chlamydia bacterium Chlamydia trachomatis estimated 92 million new cases worldwide in 1999 common cause of pelvic inflammatory disease with subsequent risk of infertility; vertical transmission causes conjunctivitis, blindness, pneumonia in infants Trichomoniasis flagellate protists Trichomonas vaginalis estimated 88 million new cases worldwide in 1999 affects women; men are asymptomatic carriers; associated with premature birth and low birth weight; may facilitate spread of HIV Candidiasis (thrush) yeast Candida albicans extremely common see Block 1 Unit 7 (*Source: WHO, 2001) STIs are a major cause of infertility in women and, when transmitted vertically to infants, they often have much more serious effects than they do in adults. They are also associated with a greatly increased risk of HIV infection: people with gonorrhoea have a threefold increased risk of contracting HIV; people with syphilis have a fourfold increased risk. The causal basis of this effect is not fully understood but is likely to involve several factors; for example, the genital lesions caused by STIs may make people more likely to be infected by HIV. STIs and host specificity It is not surprising that many pathogens exploit host sexual activity as a means of transmission. It is a very intimate form of contact, ensuring reliable transmission and, for many species, it is the only context in which hosts get close enough for transmission to occur. • What other advantage does transmission by sexual contact provide for the pathogen? Answer It guarantees transmission to another member of the same species, i.e. it provides host specificity. Thus there is little or no opportunity for STIs to be transmitted between species. As a result, STIs are typically species-specific and other host species do not have to be considered in the context of controlling them. 4.1.2 Adaptations of STIs to mode of transmission STIs share a number of features that can be understood, in evolutionary terms, as adaptations by the pathogens to their mode of transmission. • They have a long incubation period and latent period. • They typically do not cause debilitating illness, except in the very late stages of infection. • The host does not develop immunity to them and so can be repeatedly infected. STIs and low mortality Consider the host-pathogen relationship from the perspective of a venereal pathogen. The ideal host would be one that mates frequently and with many partners; such host behaviour would maximise its transmission to new hosts. In the real world, however, most hosts do not conform to this ideal. For many species, mating is a rare event and may occur only at certain times of year. There are many species that mate quite often, but do not often change their sexual partner. This is important from the standpoint of a venereal pathogen, because its fitness depends on the number of new hosts infected. In general, therefore, the interval between infection of a new host and that host mating with a new partner is long: much longer than that between catching a cold and starting to sneeze. Were a venereal pathogen to cause a debilitating, acute illness, it would severely reduce its chances of being transmitted to new hosts. Theoretical studies have suggested that, if a venereal pathogen reduces the mating success of its host, selection will favour strains of the pathogen that have reduced virulence. The optimal strategy for a venereal pathogen is thus to minimise the damage it does to its host for as long as possible. Consequently, it has to evade rather than seek to outpace its host's immune defences. STIs and evasion of host's immune defences Evasion of the host's defences involves two possible strategies: finding a place where the immune response is relatively weak, or becoming less detectable by the host's immune system. For example: • The genital tract of a female host is a potentially safe place for a pathogen because the host cannot afford defences so effective that they destroy sperm. • Herpes viruses, including herpes simplex and herpes zoster, invade the nervous system of their host, a relatively safe site because the host cannot afford to damage nerve cells by mounting a vigorous immune response. There they remain dormant until the host's immune system is compromised by stress or illness. • HIV incorporates itself into the cells of the host's immune system. • The syphilis bacterium sheds the surface molecules that enable its host to recognise it. • The gonorrhoea bacterium changes its surface molecules so frequently that the host immune response does not catch up. Interestingly in this context, the pilus genes of Neisseria gonorrhoea are one of the few places where introns occur in prokaryotic genes. The presence of introns introduces the possibility of varied splicing patterns which may contribute to variation of the N. gonorrhoea surface. Role of symptomless infected hosts Related to the essentially 'stealthy' nature of sexually transmitted pathogens is the fact that many infected people present no symptoms; they are nonetheless able to transmit STIs to others. For example, 70 to 75% of women and 50% of men infected with Chlamydia trachomatis are symptom-free; 80% of women and 10% of men infected with gonorrhoea are asymptomatic (WHO, 2001). This makes efforts to control these diseases very much more difficult. • In the context of transmission, how would you interpret the subtle symptoms of first stage syphilis? Answer This helps transmission by not affecting host sexual behaviour and the pathogen not being detected by the host or potential mates. STIs and promiscuity Venereal pathogens may have evolved mechanisms by which they manipulate their hosts to mate more often or with more partners. In humans, promiscuity and the incidence of STIs are correlated but there is no reason to suppose that the causal basis of this correlation is anything other than that promiscuous behaviour increases the risk of infection. A few studies of animals have sought evidence that venereal pathogens cause more promiscuous behaviour but none has yet demonstrated such an effect. STIs and comparative studies Given the enormous variation among animal species in terms of how frequently they mate and, more importantly, how many mating partners they have, comparative studies would provide valuable insights into the impact of these variables on the evolution of host-venereal pathogen relationships. The information that is available for animal STIs was reviewed by Lockhart and colleagues (1996). They documented over 200 diseases, spread across a wide diversity of host species and involving a wide variety of pathogens. Among mammals, STIs, in comparison with other diseases, typically cause low mortality, are long-lived in their hosts, invoke relatively weak immune responses, have a narrower range of hosts and show less fluctuation in prevalence over time. This agrees with the observations for human STIs. In the UK, detailed studies have been made of two STIs of the familiar two-spot ladybird (Adalia bipunctata). The fungus Hesperomyces virescens appears to affect ladybirds only in urban habitats and has been studied in London. Samples of ladybirds taken along transects running north to south and east to west across London show that its prevalence reaches a peak, at 50%, near Euston railway station (Figure 5.6). This figure is a line graph consisting of four plots showing the prevalence of sexually transmitted disease in ladybirds at a range of distances from the centre of London, over the summer months between May 1998 and May 1999. The horizontal axis is labelled distances from centre of London and is marked from zero in the middle to 30 kilometres to the left, denoting the southward direction; and 30 kilometres to the right, denoting the northward direction. The vertical axis is labelled proportion of adults infected and is marked from zero to 0.5. In all the four months recorded, the proportion infected was highest near the centre of the city. In May of both years, the proportion infected peaked at about 0.5, while the July 1998 peak was below 0.4 and the September 1998 maximum was only 0.05. In addition, in May of both years, infected ladybirds were recorded more than 20 kilometres both north and south of the centre, while in the September none were recorded beyond 10 kilometres from the centre. One possible explanation for this distribution is that higher temperatures in central London increase activity in ladybirds, enabling them to mate more often, and thus with more partners, than in rural habitats. Social factors The epidemiology and control of STIs are subject, perhaps more than other diseases, to powerful social factors. First, the social stigma attached to STIs makes infected people very reluctant to report them. Second, programmes to control STIs involve intense initiatives to educate people, to improve reporting rates and to carry out contact tracing. Such measures tend to collapse at times of social disruption. For example, there was a massive increase in the incidence of STIs in Eastern Europe following the break-up of the USSR. Major outbreaks of STIs also occur at times of war; additional factors here are the close association between soldiers and prostitutes and the propensity of invading armies to commit rape. 4.2 Vector-borne diseases You have already studied some of the important infectious diseases of humans transmitted by vectors (see Table 5.3). Remember that a vector is a living creature that can transmit infection from one host to another. Mosquitoes malaria, dengue fever, filariasis, yellow fever Sandflies leishmaniasis, sandfly fever Other flies trypanosomiasis, onchocerciasis Fleas plague, rickettsial infection Ticks relapsing fever, rickettsial infection, Lyme disease Mites rickettsial infection Lice relapsing fever, typhus While vectors are normally considered in the context of transmission from one human host to another, they can also transmit pathogens between different host species and are thus a factor in zoonotic diseases. Historically, vector-borne diseases, including malaria, dengue fever, yellow fever and plague, were probably responsible for more human disease and death in the seventeenth century through to the early twentieth century than all other causes combined (although respiratory infections and diarrhoeal diseases, especially among babies, may have been as important). The evolutionary biology and ecology of vector-borne diseases is much more complex than that of directly transmitted diseases, because it involves an interaction between three, rather than two, partners. It is not only the relationship between pathogen and final host that has to be considered, but also that between pathogen and vector and that between vector and final host. Insect vectors have been described as 'flying syringes', the implication being that they provide a very reliable mode of pathogen transmission, regardless of the mobility of the human host and one which, moreover, evades many of the host's first-line defences. Because vector-borne pathogens are dependent on their vector for transmission to new hosts, it is widely assumed that it is in their best interests not to harm the health of the vector. Insects have complex immune systems, similar in some ways to those of mammals (although insects lack lymphocytes and antibodies) that defend them against pathogenic microbes. The insect vectors' immune systems respond to the pathogens they carry. For example, mosquitoes mount an immune response to malarial pathogens, indicating that they are not simply passive in transmission. In response, vector-borne pathogens are covered by surface coats that not only protect them from the vector's immune system but also interact with specific vector tissues to find those sites in the vector's body where development and reproduction of the pathogen take place. Overall, the question of whether or not malaria pathogens have a harmful effect on their mosquito vectors remains unresolved. The number of studies that have found evidence of detrimental effects is roughly equal to those that have not. Several harmful effects of malaria pathogens on mosquitoes have been identified, including tissue damage, loss of protein and glucose to the benefit of pathogens, and an increased risk of death resulting from the fact that infected mosquitoes feed more often and may be more easily swatted! To be successfully transmitted by a vector, a pathogen must be able to evade or resist defensive responses to it that are made by its vector. Thus, pathogens must acquire adaptations that enable them to be vector-borne. For example, the plague bacillus Yersinia pestis evolved relatively recently from Y. pseudotuberculosis, which causes a relatively mild food- and water-borne gut disease. Genetic comparison of these two bacteria has revealed that Y. pestis differs from Y. pseudotuberculosis by the inclusion in the former of two plasmids. One of these plasmids codes for an enzyme that protects the bacillus from digestion by its rat flea vector. This ability to be transmitted by a vector may have favoured the evolution of more virulent forms of Yersinia, and thus the emergence of plague. Host-pathogen interaction and control measures The fact that vectors are not simply passive carriers of pathogens, but have a complex interaction with them, is very important for the development of new control measures for diseases such as malaria. Current efforts to control malaria are hampered by the evolution of drug resistance in Plasmodium, the evolution of insecticide resistance in mosquitoes, and the lack of an effective vaccine. A potentially effective new tool is to use genetic modification to strengthen the response of mosquitoes to Plasmodium. Such genetic modifications have already been achieved, in the laboratory, to produce mosquitoes that are relatively inefficient at transmitting the malaria pathogen. Transferring this technique to the field is fraught with problems, however, not least because of public and scientific concerns about the possible long-term effects of releasing genetically modified organisms into the wild. Recent evidence suggests that the rodent malarial pathogen Plasmodium berghei itself limits the degree to which it infects its mosquito vector Anopheles stephensi. Only a small proportion of the gametocytes that enter the mosquito's mid-gut after a blood meal survive. The mid-gut is a hostile environment for the gametocytes but more than 50% die naturally by apoptosis or programmed cell death. This appears to be an adaptation on the part of the pathogen that limits the burden that it imposes on its mosquito vector. Most significantly, it offers an interesting avenue of research into new ways to control malaria; if mechanisms that can increase the proportion of pathogens that undergo apoptosis can be found then malarial proliferation in mosquitoes might be reduced. Vector-borne diseases increasing in the modern world Some vector-borne diseases are emerging or resurging for a complex variety of reasons. These include urbanisation, deforestation and changing agricultural practices, but a major reason is the evolution of drug-resistant pathogen strains and of pesticide resistance in vectors. There are also alarming predictions about the long-term effects of climate change; global warming, it is argued, will cause vectors to expand their range, bringing malaria to temperate regions where currently it cannot survive. The data available on the impact of climate change on vector-borne diseases are inconclusive. This should be seen as 'absence of evidence', rather than as 'evidence of absence' for such an effect. Some emerging infections of humans are primarily diseases of animals that have become more common in recent years because humans have increased their frequency of contact with their vectors. For example, tick-borne encephalitis and Lyme disease (Block 1 Unit 5 and Figure 5.7 below) have increased in northern temperate regions, including Europe and North America, since the 1980s, because of greater human contact with the ticks that transmit them. The incidence of Lyme disease has increased steadily since its discovery in the USA in 1975. In the eastern USA, a lot of farmland has fallen into disuse, encouraging an increase in populations of wild deer, which carry Ixodes ticks; humans pick them up from long grass where deer have been grazing. The two-year life cycle of the deer tick shown in Figure 5.7b illustrates the various stages where human infection is increasingly possible: • Adult female tick lays eggs on the ground in early spring. • By summer, eggs hatch into larvae. • Larvae feed on mice, other small mammals, deer and birds in the late summer. • In early autumn, they moult into nymphs and are then dormant until the next spring. • Nymphs feed on rodents, other small mammals, birds and humans in the late spring and summer and moult into adults in the autumn. • In the autumn and early spring, adult ticks feed and mate on large mammals (especially deer) and bite humans. • The adult female ticks then drop off these animals and lay eggs in spring, completing a two-year life cycle. Finally, the fact that vectors can transmit pathogens between different host species raises the possibility that they have been responsible for the transfer of animal diseases to humans. For example, the blood-sucking stable fly (Stomoxys calcitrans) is capable of transmitting HIV and, because it feeds on the carcasses of dead apes offered for sale as 'bush meat' in African markets, as well as on humans, it may have been the agent that passed HIV from primates to humans.

Population dynamics and compartmental modelling

Unit 2 of this block introduced the idea of compartmental models with each compartment containing one category of host (e.g. 'susceptible' or 'infected'). This section develops a compartmental model drawing on ecological information such as the life cycle of the pathogen and the interaction between the host and the pathogen. The section on patterns of abundance (Section 1.2) introduced the idea that a series of places (including various categories of host) in which the pathogen can exist (and various places where they do not exist) can be identified. This links back to the idea of compartments discussed in Unit 2. If we can determine the abundance of the hosts representing each of the compartments at one point in time, and the rate at which the hosts flow between compartments, then we can begin to model the population dynamics of the pathogen and host. These models help us to understand why some pathogens spread rapidly through a host population while others remain at more or less similar values, or decline. Such knowledge will be extremely useful when you have to consider ways of combating a disease. 2.1 A compartment model based on malaria To illustrate this approach, this section builds a compartmental model using the example of malaria and following the categories of host identified above. This example immediately creates a new level of complexity because there is a human host and an insect vector in the life cycle of the pathogen, giving rise to eight compartments (see a-h in Table 3.2). Table 3.2 Organisation of compartments for a model of the dynamics of malaria Category Human host compartment Insect vector (females) compartment Pathogen-free (susceptible) a b Pathogen-infected (latent) c d Pathogen-infected (infectious) e f Recovered g h Total number (a + c + e +g) hosts (b + d + f + h) insect vectors This compartmental model will be used to explore the dynamics of the malaria host-vector-pathogen interaction, i.e. to see how numbers in the different compartments change over time. For example, you might ask, under what conditions and how rapidly would a malaria infection spread through a human population? The compartmental model will be used in the following section to answer this question. 2.1.1 A scenario for the model building Imagine the scenario of an isolated village with resident populations of 99 humans and 1000 female mosquitoes who are all free of Plasmodium infection. In this scenario you should assume that: • humans of different age and gender are equally likely to be infected with the malaria parasite (a potentially unrealistic assumption, but one of a number that have to be made to simplify this example scenario) • the mosquito population is more or less constant regardless of the presence of malaria • there are always 1000 female mosquitoes in the village. Now imagine that one recently infected human enters the village. The beginning of Day 1 will be designated as the point in time that they become infectious, and at this time they have not been bitten by a village mosquito. The values in the compartments for the start of Day 1 are given in Table 3.3. Table 3.3 Start of Day 1: newly arrived human becomes infectious Category Human host compartment Insect vector (females) compartment Pathogen-free (susceptible) 99 1000 Pathogen-infected (latent) 0 0 Pathogen-infected (infectious) 1 0 Recovered 0 0 Total number 100 1000 To proceed further with the dynamic modelling exercise, a few more assumptions have to be made and some more data are needed. It is valuable to list these factors, in order to see how far the model might depart from reality and to note the variability that is inherent in the malaria host-vector-pathogen system. Assumptions In addition to the previous stated assumptions regarding the equal susceptibility of all hosts to infection, and a constant number of female mosquitoes, it will be considered that: • there is no previous history of malaria in the village • the infected human arrives at the beginning of the infectious period. Data required to proceed with the dynamic compartmental model include: • length of life and survival of mosquito • number of people bitten per day • duration of components of life cycle of the pathogen. The required data are highly variable and depend on the species of pathogen and mosquito and the geographical location under consideration. The variation as well as the average values will be considered as the model is constructed. How long does the mosquito live? Female mosquitoes have a daily mortality rate of between 5 and 25%. In other words, somewhere between 95% (0.95) and 75% (0.75) of female mosquitoes survive until the next day. This means that if we start with 1000 mosquitoes, there will be somewhere between 1000 × 0.95 = 950 and 1000 × 0.75 = 750 remaining after one day. (This calculation ignores recruitment of new mosquitoes - you will see the reason for this later in the discussion.) • What is the maximum and minimum number of mosquitoes after two days? • Maximum after two days = 950 × 0.95 = 902.5 (or 1000 × 0.95 × 0.95) Minimum after two days = 750 × 0.75 = 562.5 (or 1000 × 0.75 × 0.75) In fact, the number of mosquitoes alive after n days can be expressed as a simple equation, given an initial number of mosquitoes in the scenario: (Eqn 3.1) number of mosquitoes alive after n days = initial number of mosquitoes × (fraction surviving)n • Using Equation 3.1, what would be the number of mosquitoes alive after 3 days if the daily survival rate was 90% and the initial number was 1000? • Number alive after 3 days: These values can be incorporated into the dynamic compartmental model. After every day, we can multiply the number of mosquitoes by the survival rate. We will need to do this because we want to track the abundance of the infected mosquitoes. Appropriate values of daily survival rate will be chosen below. How many people are bitten per day? Female mosquitoes mostly feed every two to four days. We also have to take into account the preferences for human blood. Anopheles gambiae feeds on average once every two nights and prefers feeding on humans, whereas A. culicifacies feeds once every three nights but feeds predominantly (about 80% of the time) on cattle with only about 20% of feeds on humans. This information can be used to calculate a daily bite rate on humans by female mosquitoes. • For these two species, what is the (average) daily bite rate on humans? • A. gambiae: 0.5 bites per day (assuming 100% feeding on humans) A. culicifacies: 0.33 × 0.2 (20% preference for humans) = 0.066 bites per day. In our compartmental model, we will work with an average daily bite rate of 0.2, i.e. somewhere between the rates for A. gambiae and A. culicifacies. Duration of components of life cycle of pathogen This information is needed in order to determine the length of the latent and infectious periods in both the human and the insect vector. The following information refers to Plasmodium falciparum unless stated otherwise, and its life cycle is shown in Figure 3.5. Gametogenesis of P. falciparum in the mosquito takes about 18-24 hours and sporogony is about 9-10 days (Figure 3.5, stages E and F) and is longer for other Plasmodium species. Thus, it is reasonable to assume that the minimum time for a mosquito to become infective after biting an infected human and itself becoming infected is about 11 days. This is the duration of the latent period in the mosquito. Sporozoites can remain in the mosquito salivary gland(s) for up to 59 days. This means that once a mosquito has become infective, it is highly likely to remain infective until it dies. • Why is the mosquito highly likely to remain infective until it dies? • Because even with a high daily survival rate of 0.9, the probability of being alive after 59 days is 0.9 multiplied by itself 59 times, i.e. 0.959 = 0.002: a very small number. In humans, passage of the sporozoites through the blood system to the liver (Figure 3.4, stage A) is quite rapid. After 45 minutes, most sporozoites will have left the blood. Hypnozoites usually remain in the liver from 9 to 16 days (stage B), although some may remain there much longer. The time from schizont to gametocyte is about 48 hours and gametogenesis takes 10-12 days. The minimum duration of the latent period in humans is therefore about 9 + 2 + 10 = 21 days. We will assume for convenience a latent period in humans of 22 days (i.e. twice the latent period in the mosquito). The delays that occur in the liver mean that humans may remain infective for many weeks following the latent period. There may also be dormancy and reactivation of the pathogen. The following three sections consider what is likely to happen to the numbers of hosts and vectors in each compartment after the point at which the one infected human becomes infectious. 2.1.2 Initial human host becomes infectious Day 1: the cycle begins After one day, we expect that each of the 1000 female mosquitos will have bitten 0.2 humans (on average). Therefore, a total of 1000 × 0.2 = 200 humans will have been bitten. As there are only 100 humans in the village, this means that every human is bitten twice each day by different mosquitoes. This bite rate will remain constant as long as there is no change in the number of humans and female mosquitoes (and the assumptions of the model hold). As one of the humans is infective, at the end of one day, two of the mosquitoes have become infected (assuming that the bite results in successful transfer of the pathogen from the human to the mosquitoes). So, at the end of Day 1, we have the situation in Table 3.4. Table 3.4 Compartmental model at the end of Day 1 Category Human host compartment Insect vector (females) compartment Pathogen-free (susceptible) 99 998 Pathogen-infected (latent) 0 2 Pathogen-infected (infectious) 1 0 Recovered 0 0 Total number 100 1000 Day 2 onwards: increasing numbers of infected mosquitoes At the end of Day 2, two more mosquitoes have become infected as a result of biting infected humans. There will now be four infected mosquitoes. (Assume that the infected human remains infectious in the village for two days and then leaves the village). Now, let us move on to the end of Day 11. We can estimate the number of infected mosquitoes alive after 11 days using Equation 3.1: number of infected mosquitoes after 11 days = 2 × (0.9)11 = 2 × 0.314 = 0.628 We also have the two mosquitoes that were infected on Day 2. The number of these alive at the end of Day 11 is 2 × (0.9)10 = 0.697. (Note that we are also assuming that the female mosquitoes that became infected were newly emerged individuals.) These data can now be entered into the compartmental model (Table 3.5). Table 3.5 Compartmental model after 11 days Category Human host compartment Insect vector (females) compartment Pathogen-free (susceptible) 99 1000 − 1.325† Pathogen-infected (latent) 0 0.628 + 0.697 = 1.325 Pathogen-infected (infectious) 0 0 Left village* 1 0 Total number 100 1000 *Note that the 'Recovered' category has been replaced by an 'emigration' category - we have assumed that the infected human left the village after two days. †Recall the assumption that the female mosquito population stays at 1000. In fact, the number of susceptible mosquitoes will not affect the model (as long as the number is large). Legitimacy of model assumptions At this point, you may be wondering about the legitimacy of using fractions of mosquitoes and fractions of humans! For a general mathematical model, this is not a problem. The interpretation in these cases can be the probability that a mosquito or human is alive as opposed to the actual number of mosquitoes or humans. It can then be applied to a problem with, say, 10 000 or 100 000 humans. We have to be more careful with specific models that are applied to real populations of a certain size, especially where the number of individuals is small. 2.1.3 The first infected mosquitoes become infectious On Day 12, we move into a new dynamic phase in which the first infected mosquitoes become infectious. For simplicity, let us say the mosquitoes infected on Day 1 and Day 2 became infectious on the same day and state that they have the abundance calculated at the end of 11 days. The number of humans bitten by the infectious mosquitoes on Day 12 is therefore 1.325 × 0.2 (the bite rate determined in Section 2.1.1) = 0.265 (Table 3.6). Table 3.6 Compartmental model on Day 12 Category Human host compartment Insect vector (females) compartment Pathogen-free (susceptible) 99 − 0.265 = 98.735 1000 − 1.325 Pathogen-infected (latent) 0.265 0 Pathogen-infected (infectious) 0 1.325 Left village 1 0 Total number 100 1000 As the mathematics rapidly becomes complicated, we need to make some simplifying assumptions that will allow us to continue making predictions about the population dynamics without qualitatively affecting the model. One simplifying assumption is that the infectious mosquitoes only bite over two days (Days 12 and 13) and then die. (The bite rate may be an overestimate, as any one mosquito only needs to feed every two to four days. On the other hand, an infected mosquito only transfers about 10% of its sporozoites in one bite.) Therefore, we need to calculate the number of humans bitten by the infectious mosquitoes on Day 13: number of Day 1 infectious mosquitoes surviving = 2 × 0.912 = 2 × 0.282 = 0.564 number of Day 2 infectious mosquitoes surviving = 2 × 0.911 = 2 × 0.314 = 0.628 total number of infectious mosquitoes on Day 13 = 0.564 + 0.628 = 1.192 The number of humans bitten by infectious mosquitoes on Day 13 is therefore 1.192 × 0.2 (bite rate) = 0.238 (Table 3.7), giving a total of 0.503 infected humans. Table 3.7 Compartmental model on Day 13 Category Human host compartment Insect vector (females) compartment Pathogen-free (susceptible) 99 − 0.503 = 98.497 1000 − 1.192 Pathogen-infected (latent) 0.265 + 0.238 = 0.503 0 Pathogen-infected (infectious) 0 1.192 Left village 1 0 Total number 100 1000 After Day 13, we are assuming that the infectious mosquitoes have died. So, on Day 14, we have the situation in Table 3.8. Table 3.8 Compartmental model on Day 14 Category Human host compartment Insect vector (females) compartment Pathogen-free (susceptible) 98.497 1000 Pathogen-infected (latent) 0.503 0 Pathogen-infected (infectious) 0 0 Left village 1 0 Total number 100 1000 • How long will the compartment values remain the same as Day 14? • Until the first newly infected human becomes infectious, which is 22 days after first being bitten by an infectious mosquito. This occurred on Day 12, so the first newly infected human becomes infectious on Day 34. 2.1.4 Other infected humans become infectious Let us assume that humans infected on Days 12 and 13 become infectious on Day 34. This day will be like Day 1, when the infectious human arrived in the village. On Day 1 there was one infectious human who got bitten by two mosquitoes. The difference on Day 34 is that there are 0.503 humans who are infectious (Table 3.9). Table 3.9 Compartmental model on Day 34 Category Human host compartment Insect vector (females) compartment Pathogen-free (susceptible) 98.497 1000 Pathogen-infected (latent) 0 0 Pathogen-infected (infectious) 0.503 0 Left village 1 0 Total number 100 1000 • Contrasting Table 3.9 with Table 3.3, what do you predict will happen to the pathogen in this particular scenario? • The pathogen population is not likely to be sustained in the village because the number of infectious human hosts (0.503) is less than 1, i.e. the initial number of infectious human hosts. We have reached an important conclusion about the dynamics of the pathogen. In this scenario, the pathogen population is predicted not to persist in the host population (the village). Indeed, we would predict that it would eventually become extinct. 2.1.5 Variables affecting the size of the pathogen population So, under what conditions might the pathogen population increase? A useful quantity to calculate would be the minimum requirements for one or more person to be infectious on Day 34. To do this, we have to look at each of the variables to see which, if any, might naturally alter and therefore can be changed in the model to favour increase in the pathogen population. This will be a useful exercise when we come to think of ways of controlling malaria (as opposed to looking at ways in which it might increase). • Suggest two ways in which the abundance of the pathogen might be increased in the village following the arrival of the infected individual. • By an increase in the daily bite rate or by an increase in the number of times the first infected person is bitten (by higher bite rate and/or by longer exposure before isolation or leaving the village). We have not considered pathogen life cycle, which is already at minimum duration, and mosquito survival, which is already high. Let us take the daily bite rate (b) as the variable to investigate. In order to determine the requirement for a minimum of one infectious person on Day 34, we need to solve the following equation: 1 infectious person on Day 34 = (no. of infectious mosquitoes on Day 12 (1.325, see Table 3.6) × bite rate, b) + (no. of infectious mosquitoes on Day 13 (1.192, see Table 3.7) × bite rate, b) This can be written as an algebraic equation: For convenience, let us take b as being 0.4. (Note that if the mosquitoes only bite once, we have 1 = 1.325b, i.e. b = 0.75.) • Is this value of bite rate (0.4) unrealistically high? • No, it is less than that recorded for A. gambiae (0.5). If b is greater than 0.4, the pathogen population will increase in the host population. If b is less than 0.4, the pathogen population will decline. The daily bite rate value of 0.4 is therefore a threshold for persistence of the pathogen (and therefore operates in a similar way to R0). So, if all the other variables of the model are held at the assumed values (and the general assumptions of the model hold), daily bite rate represents an important controlling variable on the success of the pathogen. • Considering the potential control of the disease, how could daily bite rate be decreased below the critical value of 0.4? • One straightforward method is to use mosquito nets over beds or hammocks, to coincide with the peak biting times (early evening). The use of mosquito repellents can also dramatically reduce the frequency of bites. 2.2 Usefulness of the model in practice The development of this model of malaria infection has been helpful in showing the difficulties of working with complex host-pathogen systems and the potential predictive power of mathematical models of disease (see Box 3.2). This is especially important for a disease that continues to kill about one million people per year (Murray et al., 2012). This particular model is further justified because of the role of mosquitoes (and daily bite rate) in a number of globally important diseases in addition to malaria, such as dengue fever (Figure 3.6), yellow fever and filariasis. The construction of mathematical models to understand the dynamics of malaria first began in the early twentieth century with the pioneering work of the British doctor Sir Ronald Ross (1857-1932). In the late 1890s, he had demonstrated the life cycle of the malarial parasites in mosquitoes, thereby confirming the role of the vectors in the disease. Ross dedicated much of his subsequent working life to understanding ways to control malaria. He did fieldwork in many parts of the world, including the Middle East and West Africa, and developed mathematical models of malaria epidemiology, published in a series of works from 1906 to 1916. He not only sustained a wider interest in mathematics but also received critical acclaim for his poetry! Ross's contribution was recognised in a series of awards, the most prestigious being the Nobel Prize in Physiology or Medicine 1902 which was awarded to him, 'for his work on malaria, by which he has shown how it enters the organism and thereby has laid the foundation for successful research on this disease and methods of combating it' (The Nobel Foundation, 2012). In fact, many aspects of modern epidemiological models of malaria transmission have their roots in Ross's work.

Sustaining vaccination programmes

Until May 2011, when UNICEF published the prices that it pays pharmaceutical companies for vaccines, there was exceptional secrecy about the costs, which manufacturers claimed was commercially sensitive information. UNICEF's decision to disclose its vaccine costs was taken in the hope of creating a more competitive market that persuades the most expensive vaccine suppliers to reduce their prices. Costs paid by UNICEF in 2010 ranged from US$3.50 per dose for pneumococcal vaccine to under 5 US cents per dose for tetanus toxoid. There are signs that UNICEF's strategy is working. In June 2011, several pharmaceutical companies agreed to significant price cuts for vaccines against rotaviruses and human papilloma virus, and a pentavalent vaccine like the one used in Ethiopia, which protects against diphtheria, tetanus, pertussis, hepatitis B and Haemophilus influenzae type b. UNICEF gets much of its vaccine funding from the Global Alliance for Vaccines and Immunization (GAVI) Alliance. This organisation is a public-private partnership between national governments, international agencies such as the WHO, UNICEF and the World Bank, private donors, businesses and charitable foundations. Since its launch in 2000 with an initial donation of US$750 million from the Bill and Melinda Gates Foundation, GAVI has become the main source of funding for vaccines and equipment, staff training and infrastructure such as transport, cold storage, advertising, health education, record-keeping and so on in the world's poorest countries. However, despite the progress made by GAVI and the many entirely charitable organisations providing vaccination programmes in LMICs, the vaccination gap remains huge. An estimated 19.3 million children worldwide remained unimmunised at the end of 2010, and 80% of them were born in GAVI-eligible countries. The reasons for the vaccination shortfall are complex and interacting. Cost is clearly one of them. Some of the world's poorest countries may not afford the co-payments that form part of the GAVI funding model - a requirement that is meant to satisfy sustainability criteria for vaccination programmes when GAVI funding ends. There is concern that some countries may have delayed applying to GAVI for funds to introduce new vaccines (e.g. against pneumococcal or rotavirus infections) because of the need for co-payments. A further problem is the chronic shortage of health workers, health service managers, refrigeration and other facilities needed to deliver vaccination programmes, especially in remote rural communities. Most vaccines are damaged by temperatures above 8 °C and some (e.g. measles vaccine and BCG) are sensitive to light. Without electricity, gas or kerosene to power refrigerators and cold-storage transporters, vaccines are not accessible to some of the world's poorest children.

Parasite evolution

Until around the mid-twentieth century, a central axiom of textbooks on parasites was that a well-adapted parasite would have little effect on its host. It was assumed that natural selection would drive the evolution of the host-parasite relationship to one of a stable coexistence. However, if a successful parasite is one that exploits its host for its own nutritional needs, it is more likely that natural selection would favour the parasites that were more successful at extracting nutrients, and that such adaptations would be carried by their offspring. Thus, such evolutionary adaptation is likely to result in more, rather than less, harm to the host (Anderson and May, 1982). In addition, an analysis of the relationship between stress and parasitism by Esch et al. (1975) indicated that host susceptibility to parasite infection increases when the host becomes physiologically stressed. As it is likely that the very event of parasitic infection would increase stress, hosts may thus be more susceptible to further infection. So far, this argument only reflects what might happen, through natural selection, to the parasite. But what of the host? • In what ways do you think that natural selection would drive host adaptations that reduce the effects of parasitic infection? • By increasing the effectiveness of the host's immune system. Natural selection would favour the survival of, and the production of viable offspring by, those hosts that, by chance, possessed a more effective immune response. Thus, far from a balanced coexistence between host and parasite, there is often a kind of evolutionary 'arms race' between them. Such an idea is reflected by the 'Red Queen' hypothesis. (This hypothesis derives its name from Lewis Carroll's book Through the Looking Glass, where the Red Queen tells Alice to run faster and faster to stay on the same spot.) The strategies adopted by parasites can fall between two extremes. Some parasites may adapt to be more virulent, reduce host mobility or even kill the host, whereas others may adapt to have less virulence, so that they survive in the host for a longer period of time. • Suggest a benefit for either of these strategies based on the life cycles you have read about in this unit. • Life cycles that depend on the definitive host being infected by a cyst stage, such as in tapeworms, would be enhanced by the easy capture and consumption of the intermediate host by predators, for example if the host was less mobile or even dead. Those that depend on the continual release of an infective stage, such as schistosome cercariae, would benefit from the prolonged survival of the snail intermediate host. In most cases, the long survival of the definitive host enhances continued egg release from the mature parasite. Constant mutation of genotypes in hosts, to create variation in the immune system, is necessary to counter mutational changes in the parasites' defence mechanisms. • Which reproductive process would increase genetic variation in a host population? • Sexual reproduction. It is thus very possible that sexual reproduction evolved in organisms as an adaptive response to infection by parasites (see Morran et al., 2011, and Hamilton, 1990). A more detailed analysis of the concept of coevolution is considered in Block 3.

Intrinsic factors affecting dynamics

Where there are no apparent extrinsic events driving the dynamics of pathogen abundance, you need to consider what intrinsic processes might be driving the dynamics, i.e. processes derived from the interaction between host and pathogen. 3.2.1 Cycles of abundance The search for intrinsic mechanisms that generate cycles of abundance in organisms is an issue that has fascinated ecologists for many years. It is a phenomenon that stretches far beyond host-pathogen interactions. Cycles of abundance can be found in organisms as diverse as lynx, herbivorous insects and grouse (Figure 3.9). What do the ecological systems behind the data shown in Figure 3.9 have in common with the host-pathogen system? First, they all involve interactions where one species gains (+) and the other species loses (-) as a result of the interaction. But that is not unusual; many species are involved in such interactions. A second important property is that all the predators/pathogens/herbivores (i.e. those species that gain from the interaction) specialise on a small number of host or prey species. Indeed, they usually specialise on just one species. For example, Canadian lynx (Lynx canadensis) (where they show cycles of abundance) feed predominantly on snowshoe hares (Lepus americanus) (Figure 3.10), and the larch bud moth (Zeiraphera diniana) feeds only on larch (Larix decidua). Grouse specialise on heather (Calluna) species, but in their case it is probably a parasitic nematode that specialises on them that is important. In the early 1990s, Dobson and Hudson used a combination of experiments and mathematical models to show that cycles of abundance in grouse could be due to the grouse's interaction with a parasitic nematode. The lynx and the snowshoe hare system A consequence of this specificity is that changes in the abundance of one species are expected to lead to changes in the abundance of the other. To illustrate this, imagine a system of one predator and one prey species (the lynx and the hare). In this simple scenario, lynx only eat hares and hares only eat grass. Assume that grass is abundant and unaffected by the numbers of hare. • What do you predict would be the effect of an increase in the number of lynx? • They would eat more hares and the number of hares would fall. • What would be the effect on the lynx of this reduction in the number of hares? • The lynx would have less to eat and therefore their survival and/or fecundity would decline. • What would be the effect of reducing the survival and/or fecundity of the lynx? • They would reduce in abundance. • What would be the effect of reducing the number of lynx? • Hares would increase in abundance. You will see from this set of questions and responses that it is easy to generate an intuitive argument (the real situation is inevitably more complex than this!) to explain why lynx and hare would cycle in abundance. This is particularly true if there is a delay in response of one species to changes in abundance of the other. For example, while reduced numbers of hares this year may result in the reduced fecundity of lynx in the same year, this is not manifested as a change in the abundance of mature lynx until several years later. The grouse-parasitic nematode system Very similar arguments were applied by Dobson and Hudson (1992) to the grouse-parasite system. They showed that birds treated with chemicals to kill the parasites had higher adult survival and higher hatching success. Furthermore, they showed that there were significant delays in the parasite life cycle, caused by the ability of the larval parasites to arrest their development after infecting a host. The authors' detailed mathematical studies demonstrated that cycles of abundance with a period of 8 or 10 years could be produced by a particular combination of larval arrestment duration (typically 2-4 months) and host birth rate. 3.2.2 Dormancy Delays in the pathogen life cycle are found widely, including in human pathogens. They occur due to the duration of development and reproduction of the pathogen in the host (i.e. the latent period) and due to dormant periods, e.g. between the stages of syphilis. Dormancy also appears to be an important feature of prion and virus life cycles. What might be the adaptive explanation for long periods of dormancy? Clues come from comparison with other ecological systems where dormancy is a feature, and two common examples come to mind. The first is annual plants, the seeds of which may remain dormant in the soil for many years, i.e. for periods of time which far exceed the generation time of the organism. The second example is the dormancy of cicadas (Figure 3.11), a type of plant-feeding insect whose loud noise in late afternoon and early evening is well known to anyone who has travelled in tropical and subtropical regions. These two examples represent two different reasons for dormancy. In the first, the benefit of dormancy is that the seeds of many annual plants are only able to germinate under certain conditions such as following physical disturbance, flooding or intermittent rains. These events may be infrequent and unpredictable in occurrence, e.g. in desert environments. Hence there is a selective advantage to plants that can persist during unfavourable conditions in the dormant state (which usually acts to protect them against the adverse environmental conditions) and respond quickly to intermittent favourable conditions. In the second example, selection has favoured cicadas that emerge infrequently and thereby avoid predation. This is further supported by the emergence times of cicadas, for example every 13 or 17 years in Indiana in the USA. • Why is emergence every 13 or 17 years favourable to a potential prey species? • Because 13 and 17 are both prime numbers and cover a protracted period of time, which means that predators can only specialise on these prey by emerging at exactly the same time and with the same period. Predators that emerge with any other (non-prime number) time period (or show peaks, e.g. every five or ten years) would rarely coincide with the peaks of prey abundance and therefore be unable to specialise on the prey. Applicability of dormancy explanations Does either of these explanations for dormancy in non-pathogen species help with the interpretation of patterns of dormancy in pathogen species? It is possible that both are relevant. The first explanation may be appropriate to pathogens that have specialised requirements within the host, e.g. contact with a limited set of cell types, or require the host to be at a particular developmental state, but have a broad type of transmission or unspecialised movements within the host. Thus, the pathogen cannot guarantee when and where it enters the host and needs to enter a dormant state before it encounters a part of the host in a receptive or suitable condition. For example, dormancy in malaria may be related to its exit strategy from the host. Hypnozoites of Plasmodium vivax and P. ovule in the liver can be reactivated up to 18 months later; this may be seasonal and coincide with reappearance of the vector. The second explanation may also be correct. The pathogen is seeking to evade the host's defences and may be aided by long periods of dormancy during which it remains undetectable by the host's defences. 3.2.3 Comparing the grouse-parasite system to the human system Given the presence of these delays, can the population dynamics argument applied to grouse and their parasites be applied to humans and their pathogens? Humans have host-specific pathogens that can have an effect on survival and fecundity, and those pathogens can also have periods of dormancy in the host. However, a major difference from the grouse-parasite system is that humans have a much longer generation time and effects on human fecundity are not going to result in numerical responses in the manner of the grouse. In other words, humans cannot show great variation in 'hatching' success! Another problem is that humans may be interacting with a much wider array of pathogens than the grouse. Or perhaps more accurately, there is no single pathogen in humans that is of such overwhelming importance as the parasitic nematode is to the grouse (99% of juvenile birds in a sample of 2723 were infected; Dobson and Hudson, 1992). If there were situations in which humans were mainly interacting with one pathogen whose impact on human populations was high in terms of survival and/or fecundity (TB in pre-industrial England may have been a candidate), then cycles of host abundance might be seen. But such dynamics might be played out over tens or hundreds of years and therefore possibly be undetectable on the timescales of scientific studies. With humans, the most interesting host-pathogen population dynamics are probably happening at the cellular level within a single individual, where the immune system is interacting with an array of pathogens. It is the cellular level to which we now turn our attention.

Giardiasis

Chronic diarrhoeal disease caused by Giardia spp. Flagellated protists that parasitise the intestines and cause the chronic diarrhoeal disease known as giardiasis belong to the genus Giardia. These organisms were discovered by Vilein Lambl in 1885, and may have been seen even earlier by van Leeuwenhoek, Anton van Leeuwenhoek is credited with the discovery of microbes in 1675 (Unit 1, Section 1.2). but they were not recognised as pathogens and linked to the cause of giardiasis until the 1960s. More than 40 species have been recorded globally, but humans are usually infected by Giardia lamblia. (Note that this organism appears as Giardia duodenalis or Giardia intestinalis in some texts.) The genus Giardia is very ancient in evolutionary terms and, interestingly, Giardia spp. have no mitochondria. They do, however, possess a structure called a mitosome, which is derived from a mitochondrion; this is possibly an adaptation to a mostly anaerobic way of life. As you will see below, each stage in the life cycle of G. lamblia is finely tuned to its environment. • From what you have learned so far, do you think Giardia have always been parasites? Explain your reasoning. • The protists are extremely ancient and evolved long before mammals. Therefore it is likely that Giardia were free-living in the past and have only recently adapted to the mammalian gut. Infection occurs when organisms are ingested as cysts (which have four nuclei) in food or water contaminated with faeces from humans or other mammals. Following ingestion, excystment is triggered by the low pH in the stomach, although the trophozoites (feeding stage, with two nuclei each; see Figure 6.1) do not emerge until the duodenum is reached. Trophozoites do not emerge in the stomach as the gastric acid would kill them. Emergence in the small intestine is stimulated by the increase in pH and by the proteases found in the duodenum. The pear-shaped trophozoites then use their four pairs of flagella to swim through the liquid of the small intestine, and to attach to the mucus coating of the gut epithelium. Some trophozoites penetrate the mucus layer and, using specific receptors and their sucking pads, attach to the epithelia of the duodenum and jejunum (the upper and middle part of the small intestine, respectively). Once the trophozoites are attached, they multiply by binary fission (asexual reproduction in which a cell splits into two daughter cells) and may reach sufficient numbers to disrupt the absorption of fat from the intestine. A week or two after infection this results in loose, fatty, stools with an offensive smell. Diarrhoea can continue for several weeks and may lead to weight loss and anaemia. If trophozoites are carried to the far end of the small intestine, the increase in pH and a high bile concentration trigger encystment. If encystment did not occur, the trophozoites would die, since they cannot survive outside the host under normal conditions. An encysting trophozoite lays down a thick, fibrous protein coat around itself and undergoes one round of asexual reproduction inside this protective coat, to give four nuclei. The cysts, which persist in freshwater, are then discharged in the faeces, and as few as ten of them are required to infect another individual. Giardiasis is particularly common in children in low- and middle-income countries (LMICs) with prevalence rates of 20-30%. Here, transmission is usually by ingestion of cysts present on hands or fomites. It also affects children (especially those in day nurseries) and adults in high-income countries (HICs), where the route of transmission is more commonly by ingestion of contaminated water. The cysts can remain viable in soil for several months and are able to withstand the concentrations of chlorine used to disinfect drinking water. Cysts can be removed from drinking water by filtration, although this is not always economically possible. Diagnosis is by detection of cysts in stool samples. The infection may be self-limiting (i.e. resolves by itself) but can also be treated with the antiprotist drugs such as metronidazole (see Table 6.1).

Serovars

Groups of closely-related microorganisms distinguished by a characteristic set of antigens. (Also sometimes referred to as serotypes.)

Parasite transfer

Infectious agents can generally be described as being communicated from host to host by horizontal transmission (i.e. person-to-person spread) or vertical transmission (from parent to offspring in utero, etc.), as you read in Unit 2 Section 4. Parasitic helminths are transmitted horizontally. Flukes (including all those that are human parasites), tapeworms and roundworms have elaborate and complicated transfer mechanisms compared with microbes. This is because they are too large (see Unit 2, Table 2.1) to become airborne or to penetrate by direct contact into a host, even if they could be transferred directly, e.g. by water. A few exceptions exist (Schistosoma, Necator and Ancylostoma) but in these species it is only the very small larval stages that are able to directly penetrate a host. Thus, many parasites have evolved special morphological and physiological adaptations to aid their transmission. Alongside these features is the ability of larger parasites to survive in the definitive host for many years. In addition, many of the pathobiological effects on the host may be part of an adaptive mechanism to change host behaviour in favour of transfer of the parasite to another host. To appreciate some of the fascinating, if sometimes bizarre, ways in which parasites ensure their transfer, examine Figure 8.1. List the main features that you consider to be similar in the three life cycle strategies, and those that you consider to be different. • Similarities: • Adult parasites exhibit sexual reproduction. • Eggs may be released into water. • Larval stages are involved in transfer between hosts. • Larval stages usually enter the definitive host (i.e. final host) in food (although penetration by larvae may occur in some flukes and roundworms). Differences: • In flukes and tapeworms, an intermediate host is always involved in the life cycle. Roundworm larvae do not usually use an intermediate host although, in a few groups, a vector is involved. The vector usually effects transfer actively, e.g. by biting, whereas if the larval stages are free-living, transfer to the human host is passive, e.g. by being eaten. • Asexual reproductive stages occur in all flukes and some tapeworms but only sexual reproduction occurs in roundworms. When an intermediate host or a vector is involved, this is said to be an indirect life cycle. In all other cases, it is a direct life cycle. In terms of human parasites, a sound knowledge of the life cycle of the parasite and any behavioural aberrations of hosts is essential for the employment of effective methods of prevention and control. Some 287 helminth species have been reported in humans (Taylor et al., 2001), of which significant disease is caused by 25 fluke species, 14 tapeworms, and 36 roundworms (Mata, 1982). In Sections 3-5 we examine a few of these pathogenic helminths in more detail. 2.1 Increasing the chance of transfer The chance of one single offspring of an adult parasite finding another host before it dies is exceedingly small. To compensate for such mortality, parasites produce very large numbers of offspring at certain stages of their life cycle. Hundreds, sometimes thousands, of eggs are released after each mating by the adults. Roundworms usually, but not always, have separate sexes, but flukes and tapeworms almost always show hermaphroditism, although cross-fertilisation is the norm. Parasites also exhibit a variety of complex larval stages and adaptive features that enhance survival and transfer between hosts. The rest of this section provides a brief overview of this process for flukes (the subject of Section 3 of this unit), tapeworms (Section 4) and roundworms.

Role of zoonoses in communities

In reviewing the role of other species in host-pathogen interactions, it is useful to revisit the concept of zoonoses, which was introduced in the first unit of Block 1. • Recall the definition of zoonoses (singular: zoonosis). • A disease caused by an infection that can be transmitted to humans from other vertebrates under naturally occurring conditions. This link is represented in Figure 3.14 by the arrow from alternative host(s) to human host. Zoonoses can be classified into four groups, depending in part on the manner of transmission: • direct zoonoses • cyclozoonoses • metazoonoses • saprozoonoses. Direct zoonoses involve transmission from the vertebrate host to a susceptible human by direct contact, fomites or a mechanical vector. In this example, there is no developmental change or propagation of the organism during transmission. Examples of direct zoonoses include rabies and brucellosis. Cyclozoonoses require more than one vertebrate host but no invertebrate host and include human tapeworm infections. In metazoonoses, the agent multiplies and/or develops in an invertebrate host before transmission to a vertebrate host is possible. Examples include arboviruses, plague and schistosomiasis. • Why is malaria not an example of a metazoonosis? • Because, although the pathogen develops and multiplies in an invertebrate vector during transmission, it is usually transmitted between humans and not from a vertebrate host. The final category of zoonoses is saprozoonoses. In this case, non-animal development sites or reservoirs are required, such as food plants, soil or other organic material. Examples in this category include some mycotic diseases. 5.2 Radiation of host-pathogen communities Examples such as HIV show that pathogens are continually switching to new hosts. In so doing, they are creating new host-pathogen communities. One way of understanding the changing patterns of host-pathogen communities is to consider the different communities that have arisen among closely related pathogen species. In evolutionary terms, we can talk about the radiation of host-pathogen communities, in the same way that we might discuss the radiation of non-pathogen species, e.g. Darwin's finches on the Galápagos Islands. The numerous species of trypanosome provide an excellent example of the radiation of host-pathogen communities. • Recall briefly the characteristics and different types of trypanosome. (You read about these organisms in Block 1 Unit 6.) • Trypanosomes are single-celled, animal-like cells with flagella. They include Trypanosoma species, which cause a variety of diseases including trypanosomiasis (sleeping sickness). Another group of trypanosomes are in the genus Leishmania, which cause a whole series of different types of disease. (See Block 1 Unit 6, Section 2.3.) The extraordinary diversity of host-pathogen communities associated with Trypanosoma and Leishmania is illustrated in Figure 3.15. For Trypanosoma, the interactions include a variety of vectors, especially flies (Glossina species - which include tsetse flies - and horseflies or tabanids) and, in stark contrast, a reduviid bug, which is the vector for Chagas' disease. The other vertebrates involved in these communities include semi-domesticated stock (horses, donkeys, cattle, camels) and wild animals (antelope and deer). Some of the communities do not include humans and the diseases that characterise them are diseases of the other vertebrates. In one case, there is no vector (dourine acute).

Tapeworms

In tapeworms, a larva known as a hexacanth (so named from the Latin hexa - meaning six, because it has six hooks arranged in three pairs) develops inside the egg (Figure 8.3a). If the egg is taken in by the correct intermediate host, the oncosphere is released in the gut, penetrates into the tissues, and forms a cyst stage, the cysticercus or bladderworm. This cyst contains a small invaginated (turned inside out) tapeworm head or scolex (Figure 8.3b). If this intermediate host is eaten by the definitive host, the tapeworm scolex evaginates (pops out) and attaches to the gut wall. In some species of tapeworm, the cysticercus stage exhibits massive asexual reproduction and buds many cysts within cysts (rather like a Russian doll), so that from one invading oncoshpere, hundreds, even thousands, of tapeworm heads (scoleces) develop. These cysts are called hydatid cysts (Figure 8.3c).

UK statutory notification system

In the UK, the statutory requirement for the notification of certain infectious diseases came into being towards the end of the nineteenth century. Diseases such as cholera, diphtheria, smallpox and typhoid had to be reported in London from 1891 and in the rest of England and Wales from 1899. The list of diseases increased over the decades and now stands at about 30. Originally, the head of the family or the landlord had the responsibility of reporting the disease to the 'proper officer' of the local authority. Now this is restricted to the attending medical practitioner, either in the patient's home or at a surgery or hospital. The prime purpose of the notification system is speed in detecting possible outbreaks and epidemics. Accuracy of diagnosis is secondary and, since 1968, clinical suspicion of a notifiable infection is all that is required. Originally, statistics were collected nationally at the Registrar General's Office, which already collected data on births, marriages and deaths. The Office was later known as the Office of Population Censuses and Surveys (OPCS) and is now the Office for National Statistics (ONS), but in 1997 the responsibility for administering the notification system was transferred to the Communicable Disease Surveillance Centre (CDSC). The proper officers, who are usually consultants in communicable disease control (CCDCs), are required every week to inform the CDSC of details of each case of each disease that has been notified. The CDSC has responsibility for collating these weekly returns and publishing analyses of local and national trends. The CDSC in the UK has now been devolved to the four national administrations. Figure 2.4 shows graphs over time of several notifiable diseases in the UK. The interpretation of case reports requires some care. For example, apparent changes in incidence could be due to changes in reporting practice or heightened awareness. Nevertheless, case reports (or rates derived from sentinel reporting schemes) are a valuable tool in identifying trends and other patterns. • The general trend in infectious diseases in the UK during the twentieth century was downwards, at least before the HIV epidemic started. Figure 2.4 confirms that this is broadly true for diphtheria, typhoid, measles, scarlet fever, and whooping cough. However, the graph for polio tells a very different story: there was a big increase in the decade after 1945, which was stemmed by the introduction of mass vaccination. Suggest some potential reasons for this increase. • In interpreting time trends it is important to consider the effects of: a. demographic changes (a larger population will produce more cases, even if the infection rate remains constant) b. changes in the reporting system or in reporting practices (for example, the true or perceived severity of a disease will affect the likelihood of it being reported) c. changes in the clinical manifestation of the infection. In the case of polio, reason 'a.' is insufficient to explain the large variation in reports. The truth is probably a combination of reasons b. and c.: polio became a major public health issue in the post-war period, as the incidence of acute flaccid paralysis - a common symptom of poliovirus infection - increased (although probably not the incidence of poliovirus infection). 4.1.2 Reservations when using clinically based reports Just as there is a danger that case reports can be accidentally misinterpreted as a result of changes in reporting practice, a further problem with reports based on purely clinical criteria is their lack of specificity. Many cases attributed to an infection may, in fact, be due to other causes. This is particularly problematic for some rare infections. Thus, for example, since the introduction of the combined measles, mumps and rubella (MMR) vaccine in the UK in 1988, reports of measles have fallen. However, most of the reported measles cases are not measles at all, but are caused by other infections that produce a rash, such as parvovirus infection. Thus case reports of measles are currently unreliable for studying the epidemiology of measles in the UK. • On the other hand, measles case reports prior to mass vaccination were very useful indeed. Why was this? • Prior to the introduction of mass vaccination, the incidence of measles was very high. Most of the reported rash-like illnesses were indeed measles; other infections accounted for only a small proportion of the total cases. 4.2 Laboratory reports In order to circumvent the problem of non-specific reports, it is necessary to use laboratory-based identification techniques, such as culture, typing or serological testing. Laboratory reports are a second major source of infectious disease data. These are counts of infections that have been confirmed by laboratory identification. Laboratory reports suffer from similar problems of potential bias as case reports, since clinical diagnosis usually precedes laboratory confirmation. Hence changes in clinical awareness or technique will also affect laboratory reports. Additionally, confirmation might only be sought in serious cases; hence laboratory reports might be less representative than case reports. However, they are undoubtedly more specific than case reports. The major sources of laboratory-based infectious disease data in the UK are the Public Health Laboratory Service (PHLS), which collects laboratory reports from laboratories in England and Wales, and the Scottish Centre for Infection and Environmental Health (SCIEH) in Scotland. After 2003, the PHLS was incorporated into the Health Protection Agency (HPA). Figure 2.5 shows laboratory reports for selected infections. • The infections shown graphically in Figure 2.5 have the following characteristics: • rotavirus (a) is a viral enteric infection, accounting for much paediatric diarrhoea • Clostridium difficile (b) is associated with antibiotic use, and is of particular concern in hospitals • Salmonella derby, Shigella sonnei and Salmonella typhimurium DT104 (c, d, and e, respectively) are all bacterial enteric infections • influenza B (e) is a respiratory infection. For each infection, comment on (i) the overall trend in reports and (ii) any epidemics or cyclical effects. i. C. difficile and S. typhimurium DT104 both show increasing trends. It is not clear from these data alone whether the trend is genuine, or due to improved reporting. None of the others shows clear trends. ii. Rotavirus and S. typhimurium DT104 have marked annual cycles, peaking in winter and spring for rotavirus and summer for S. typhimurium. Influenza B has irregular epidemics. There appears to have been an epidemic of S. sonnei in 1991-1993. In spite of the small numbers of laboratory confirmed infections, it looks like S. derby has epidemic peaks in autumn. There are no obvious epidemic fluctuations for C. difficile.

Leishmaniasis

Leishmaniasis currently threatens about 350 million people in 88 countries around the world. There are three major forms of the disease: cutaneous, muco-cutaneous and visceral infection, caused by 20 species of Leishmania that are pathogenic for humans. The most important of these species are shown in the Interactive Taxonomy diagram. This variability of the clinical features results from both the diversity of the Leishmania species and the immune responses of the hosts. All species of Leishmania that are infective to humans are transmitted by the bite of female sandflies of the family Phlebotominae with the genus Phlebotomus performing that function in the 'Old World' (Africa, the Mediterranean basin, the Middle East and the Indian subcontinent), and Lutzomyia in the 'New World' (the Americas). Sandflies are tiny insects (2-3 mm long) found mainly in tropical areas but also extending to more temperate zones. Until 2003, when a new species of Leishmania was found in red kangaroos in the Northern Territory, there had been no record of Leishmania in Australia. This was thought to be because of a lack of a suitable insect vector. However, painstaking research found the vector of this new species to be a daytime biting midge and not a sandfly (Dougall et al., 2011). Globally, over 30 species of sandfly are known to transmit the parasite to humans. In many areas the disease is zoonotic, with both wild (sylvatic cycle) and domestic animals (domestic or urban cycle) acting as reservoir hosts for Leishmania. This factor makes the disease very difficult to control. The infective promastigote stage of the parasite is injected into the bloodstream of a susceptible host with saliva as the sandfly feeds (Figure 6.7, Step 1). The motile promastigotes are taken up by macrophages where they rapidly revert to the amastigote form and begin to divide by binary fission. Eventually the macrophages lyse, releasing many amastigotes, which are taken up by more macrophages and the cycle is repeated Figure 6.7, Stages 2-4). The parasite is able to survive within macrophages because it overcomes their normal killing mechanism by raising the pH of the normally acidic phagolysosome inside the macrophage, creating an alkaline environment in which the protease enzymes are unable to function. Sandflies become infected when they ingest infected macrophages along with the blood meal (Figure 6.7, Stage 5). On release in the stomach of the fly the amastigotes transform into promastigotes and multiply by binary fission before migrating to the proboscis (Figure 6.7, Stages 6-8). The three forms of Leishmaniasis are described below, along with information about treatment and control. A more detailed summary of the various species that fall into these categories is provided in Section 2.4.2. Cutaneous leishmaniasis Activity ranges from a single sore or ulcer characterised by an elevated rim at the site of the sandfly bite (cutaneous infection) to multiple nodules spread over the body (diffuse cutaneous infection). The single sores usually heal within a year leaving an unsightly scar and the person is then immune to further infections. Although the disease is self-limiting, the ulcer can be treated topically with drugs based on the chemical element antimony (Sb), namely the antimonials. An early form of vaccination involved taking infected material from a sore and rubbing it into a scratch made somewhere on an uninfected individual's body where it would not be seen. Diffuse cutaneous leishmaniasis is a chronic, progressive disease which is difficult to treat. Current drugs are the antimonials but these are gradually being preplaced by amphotericin B and miltefosine. Muco-cutaneous leishmaniasis This is a chronic, disfiguring disease affecting the mucous membranes around the nose, mouth and upper respiratory tract. Parasites disseminate from the primary cutaneous lesion that slowly but spontaneously heals to form other chronic ulcers. These ulcers appear after months or years on the skin, mouth and nose and are symptomatic of the destruction of the underlying tissue. Treatment is by systemic drugs such as antimonials and amphotericin B. Visceral leishmaniasis This is a systemic disease characterised by persistent fever, enlarged liver and spleen (hepatosplenomegaly) and emaciation. It is commonly known as kala azar meaning 'black fever' in Hindi due to the dark pigmentation of the skin which often occurs. Around 90% of cases occur in just six countries: India, Bangladesh, Nepal, Sudan, Ethiopia and Brazil, and it causes over 50 000 deaths every year. There is an incubation period of 2-6 months between infection and the manifestation of symptoms. After this time the condition progressively worsens and is fatal if left untreated. Treatment involves prolonged administration of the toxic pentavalent antimonials or the newer but more expensive amphotericin B and miltefosine. Miltefosine is the first drug against visceral leishmaniasis that can be taken orally, which should result in an improvement in compliance in those instructed to take it. Leishmania tropica - Mediterranean basin; Middle East; India Anthroponotic in urban areas. Also zoonotic with dogs as reservoir hosts. L. major - dry, arid areas of North, Central Sub-Saharan, and East Africa; Middle East; parts of Central Asia; India Zoonotic with rodents such as rats and gerbils as principal reservoir hosts L. infantum - Southern Europe; Eastern Mediterranean; China Zoonotic with dogs and rats as hosts and children as principal human hosts Cutaneous (Old World) Leishmania tropica L. major L. infantum L. aethiopica East Africa Zoonotic with hyraxes Cutaneous (New World) L.(Viannia) braziliensis L.(V) guyanensis L.(V) panamensis L.(V) colombiensis L.(V) shawi L.(V) lainsoni L.(V) naiffi L. peruviana L. mexicana L. mexicana Central and South America Zoonotic: (dogs, rodents, monkeys, sloths, marmosets, agoutis, armadillos, opossums) depending on region Diffuse cutaneous (Old World) L. aethiopica East Africa Zoonotic with hyraxes Diffuse cutaneous (New World) L. mexicana Central America Zoonotic with forest rodents Muco-cutaneous (Old World) L. aethiopica East Africa Rare, but zoonotic with hyraxes Muco-cutaneous (New World) L. braziliensis/espundia Brazil; Bolivia; Peru Zoonotic with forest rodents Visceral (Old World) L. donovani Mainly in India; Nepal; Bangladesh; Sudan; Ethiopia Anthroponotic in India and Sudan. Zoonotic with dogs as reservoir hosts eleswhere. L. infantum Southern Europe; Eastern Mediterranean; India; Africa; Eastern Asia and parts of China Zoonotic with dogs, foxes and jackals as reservoir hosts. Children are main hosts Visceral (New World) L. chagasi (L. infantum) Central and South America, especially Brazil Zoonotic with dogs and foxes as reservoir hosts

Mathematical modelling

Okay, so far on your course [module] you will have come across several important concepts in infectious disease epidemiology such as the herd immunity threshold and the basic and net reproduction numbers. Both of these determines the pattern in the instance of immunising infections and also the difficulty in controlling the infection. In this session we will use mathematical modelling to revise these concepts and also to highlight some of the insights that modelling can provide into the dynamics and control of infections. Slide 2 (00:30) Now, many countries routinely vaccinate young children against common infections such as measles, mumps and rubella. Before the introduction of vaccination, notification rates for these infections were very high. This slide shows notification rates of measles in England and Wales from 1914 and there we see, before the introduction of vaccination in the 1960s, regularly saw peaks in the notification rates reaching roughly 700 per 100 000 per quarter. So these are very high notification rates. The introduction of vaccination from the late 1940s led to huge reductions in the notification rate. We also see that vaccination had some other interesting effects. For example, the time interval between peaks in the epidemics actually increased. This slide shows that several similar effects were also seen for other infections. Slide 3 (01:20) On the right-hand side I've included the notification rates for whooping cough also known as pertussis. And here we see that again the time interval between peaks increased following the introduction of vaccination and also, as you would hope, the notification rates decreased. One of the interesting aspects of these figures is that we see that, even if the vaccination coverage is less than one hundred per cent, notification rates could actually drop to very low levels. This suggests that it may be possible to control the transmission of an infection without vaccinating everyone. Now the level of coverage for this to occur is known as a herd immunity threshold. It's also known as a critical immunisation threshold and you will probably have met this already in your previous studies. Now the effects of vaccination on the cycles of instance have been studied extensively using mathematical models. Slide 4 (02:14) We will begin by using modelling to revise the concept of the herd immunity threshold using a simple model, and exploring what happens after we introduce an infectious person into a totally susceptible population. We will then extend the model to think about long term dynamics of infections and the relationship between the herd immunity threshold and the peaks in the instances of immunising infections before going on to use the model to think about the effects of vaccination. The theory described here is not new and it was known many years ago. It was elaborated in a book by Anderson and May, which is provided in the last page of references. Slide 5 (02:46) We will begin by considering the simplest scenario: what happens when you introduce an infectious person into a totally susceptible population? We would like to make predictions of the size of the susceptible and immune population over time together with the numbers of infectious people. Slide 6 (03:00) To do this we will use a model which has the same structure as the diagram on the top of this slide. Here we stratify people into those who are susceptible, those who are pre-infectious and those who are infections and those who are immune. The model needs several input parameters. For example, we need to know what is the base reproduction number. For measles it's thirteen. We need to know the pre-infectious period so that's the time interval between infection and onset of infectiousness, which is eight days for measles. We need to know the infectious period. We need to know the total population size and we also need to know what proportion of the population is immune at the start. Now according to the theory that you will have covered elsewhere on the course [module], the herd immunity threshold is equal to one minus one over R0. So if we substitute the value for R0 of thirteen for measles into this expression, we will obtain a value for the herd immunity threshold of ninety-three per cent. So that means that we need to vaccinate at least ninety-three per cent of the population in order to control transmission. We will now link into a real model and see what it predicts when we incorporate these parameters. Slide 7 (04:11) Okay. I'm now going into the demo of the model. Okay. Now this window shows the general structure of the model and it's identical to the structure of the model that we showed on a previous slide. Slide 8 (04:24) The parameters are listed here. And here we see that the start time is zero and we also have the basic reproduction number which is thirteen. We have the pre-infectious period, which is seven, and we have - sorry - the pre-infectious period is actually eight. And the infectious period which is seven - and the proportion of the population that is immune at the start is zero. Let us run the model and see what it predicts. So if I click on the run button, here we see what the model predicts about the number of new infections occurring over time. We see the number of newly infectious people going up and go down, as we would expect for a normal epidemic. It's also interesting to see what happens to the proportion of the population that is immune over time. I'm going to click on this button and here we see that the proportion of the population that is immune goes up and it ultimately reaches something which is very close to one hundred per cent. Now one of the interesting things to point out in this model is that, when the epidemic peaks, what is the value for the proportion of the population that is immune? So here - I'm pointing to the point where the epidemic peaks - and we see that the proportion of the population that is immune is actually roughly ninety-three per cent. Or ninety-two per cent, which is consistent with what we predicted using our simple equation for the herd immunity threshold. Let's see what happens to the proportion of the population that is susceptible. And that is just calculated as one minus the proportion of the population that is immune and we see that this actually goes down when the peak, when the epidemic peaks, the proportion susceptible is roughly about 0.07, it's actually about seven per cent. At this point it is useful to recall another important statistic that you will have come across in your studies and that is the net reproduction number. That's defined as the average number of secondary infectious individuals resulting from each infectious person in a given population, that is, a population in which some of the people may already be immune. And, as you will have learned elsewhere, this is calculated as the basic reproduction number multiplied by the proportion susceptible. So if I click on the button for the net reproduction number - that's RN - and here we see what the model predicts for the net reproduction number as given by the blue line here and we see that, at the start, the net reproduction number is very high but at some point the net reproduction number gets below one. And that is actually at the same point at when the epidemic peaks. So, when the epidemic is set to peak, the net reproduction number equals one so each infectious person is leading to one secondary infectious person. Where the net reproduction number is below one, you see that the number of new infectious people is actually going down. That's because each infectious person is leading to less than one secondary infectious person. It's also interesting to see what happens if we consider another infection and that is influenza. Now to do this we need to change some of the input parameters in our model. So we need to change the pre-infectious period which is two days for influenza. So I change it here. We need to change the infectious period which is also two days on average and also the basic reproduction number which is two - roughly two. Okay. Now according to the theory so beforehand we saw that the herd immunity threshold is calculated as 1 minus 1 over R0. If we substitute a value for R0 of two into that equation we would predict the herd immunity threshold is roughly 0.5. Okay. So let's run the model and see what it predicts for our epidemic curve. Okay. I'm going to de-select the proportion immune and the net reproduction numbers and just run the model again. And this is what the model predicts about the number of new infectious people over time. And, if I click on the button for the proportion immune, we see that again at approximately the same time that the epidemic peaks, the proportion of the population that is immune is roughly 0.5 and that is exactly the same value as a herd immunity threshold that I just calculated. Now at present we're assuming that everybody is susceptible in the population and this might be the case if we have a new strain of influenza such as what might occur if we have a pandemic of influenza about to occur. Now what would happen if some of the people were immune at the start? This might occur if for example we had experienced the first wave of an influenza pandemic and then went on to experience a second wave so that some people may still be immune because of infection that they acquired during the first wave. Slide 9 (09:41) Now this can actually occur in reality. I'm going to link into slide nine in my PowerPoint presentation and show you that this can occur in practice. Slide 10 (09:42) And here we see a slide for influenza in Scandinavia and the US during the period 1918 to 1919 and here we saw that there were two waves of the pandemic in each setting. So, in the Scandinavian setting, we had one - one wave occurring in late 1918 and early 1919 and similarly for Scandinavian cities. I'm now going back to my model presentation ... Slide 11 (10:13) Okay. So we are going to explore what happens if some of the people in the population were immune at the start. Okay? So I'm now going to now vary the value for the proportion immune and see what happens. So if I click on the button here ... proportion immune ... Slide 12 (10:31) So this is interesting, isn't it? So here we see that the actual - the peak of the epidemic actually goes down but it also occurs later. Suppose we carry on increasing the coverage. Suppose we carry on increasing it say to about thirty per cent so the peak goes down further. Suppose we now increase the coverage at the start so that more than fifty - so that fifty - over fifty per cent of the population is immune at the start. And what we see here is that no epidemic appears to occur at all. Now the reason for this is if a sufficient proportion of the population is immune at the start, this means that every reproduction number at the start would have been below one. And we can see this if we add the net reproduction number to the plot. So at the start with an effective coverage of 0.52 we see that the net reproduction number is below one and it continues to go down and the number of new infectious people we hardly see any transmission occurring in the population. So this highlights that in theory we could control transmission of influenza with a low coverage. However, the reality is somewhat more complicated than this because the serial interval for influenza is very short. So if you recall serial interval is defined as the time interval between successive cases in a chain of transmission. For influenza it can be a matter of days - so maybe four days, maybe three days - but it's very short and that is why it's very difficult to control influenza just by vaccination. There are also other complications and I don't want to go into too much detail about these now. One of them is influenza strains change over time and there's also the issue of age-dependent mixing so the amount of contact between people depends on the age of the people. Slide 13 (12:30) Now the previous example considered what happened when we considered a totally susceptible population and you also just considered the short-term patterns. However, many infections are endemic as was the case before vaccination was introduced in England and Wales for measles and pertussis and other infections. In order to predict what will happen for these infections using our model, we need to modify our model and include births and deaths. Slide 14 (12:59) So suppose we extend our model to include births and deaths. The top diagram shows how we need to change our model so the diagram is similar to what we had before except that we've included an extra arrow for the births - so people coming in - and also we have arrows going out of each compartment from the susceptible pre-infectious, infectious and immune categories representing death. Slide 15 (13:22) And now we're going to deal with the same model we dealt with previously but we've adapted it to include births coming in and people dying. So we have new births coming into our model diagram and we have people dying from each of the compartments. And if you run the model ... I'm going to go to this window which shows predictions and click on the run button ... And here we see that the model predicts that the number of new infectious people will cycle over time in a similar pattern to what we see in the real data. Suppose we add what happens to the proportion susceptible and the proportion immune in the population. Okay. And here we see that these values also cycle over time. And they're very closely related to the values that we calculated for the herd immunity threshold so the proportion immune - it seems to cycle around the value of 0.93 which is the value for the herd immunity threshold that we calculated previously. Let's look at these patterns in further detail. This shows what happens if we focus on a specific time period say between forty and fifty years after we introduce one infectious person into the population. And this shows what happens to the number of new infectious people. And the red line shows what happens to the net reproduction number. What is interesting in this figure is that if we look at the number of new infectious people, it involves the number of new infectious people over time is increasing, the value for the net reproduction is actually going up and it's going down. But throughout this period the net reproduction number is actually above one. And this is consistent with what we might expect. Each infectious person is leading to more than one secondary infectious person and therefore the number of new infectious people should go up. Conversely, when the number of new infectious people is going down, we see that the net reproduction number is also going down and going up but, throughout this period, its value is consistently below one. Let's look at what happens to the proportion susceptible in the population. So here the black line again, that still shows the number of new infectious people, but the green line shows what happens to the proportion that's susceptible and here we see that, whilst the number of new infectious people is going up, the proportion susceptible is also going up and going down but it's above some value. And it turns out that that value is 0.07. And whilst the number of new infectious people is going down, the proportion susceptible throughout this period is below 0.07. If you recall, that is exactly the same value that we calculated for the epidemic model for the instance increase, the proportion susceptible must be bigger than one over R0. Let's look at what happens to the proportion immune over this time period. So here we see again the number of new infectious people over time. The proportion immune whilst the number of new infectious people is going up the proportion immune is below some value and that value turns out to be 0.92, which is similar to the value for the herd immunity threshold that we calculated. And whilst the instance is - increasing the proportion immune is below 0.93 whilst the instances decreasing the proportion of immune is above 0.93. Slide 16 (15:00) We can summarise what we've seen in this model, using this slide. So the red line shows the number of new infectious people over time and the blue line shows what is happening to the proportion susceptible over time. So whilst the instance of new infectious people is at a peak, proportion susceptible equals one over R0 so at this point the net reproduction number equals one. Okay? Let's now plot what happens to the net reproduction number over time. Whilst the instance is at a peak, the net reproduction number equals one. When the instance is at a trough, the net reproduction number equals one as well. Whilst the instance is decreasing, the net reproduction number is below one. Whilst the instance is increasing, the net reproduction number is above one. Also see the proportion immune. Whilst the instance is at a peak, the proportion immune equals one over - where the instance is at a peak, the proportion immune equals one minus one over R0 which equals the herd immunity threshold. And the same for when it's at a trough. When the instance is decreasing, the proportion immune is above that value. When the instance is increasing, the proportion immune is below that value. So we see that the number of new infectious people cycles over time and also the net reproduction number, the proportion immune and the proportion susceptible cycle over time. However, the peaks and the troughs in these cycles do not coincide. Now we can build on these simple results to explain why the time period between epidemics differs between settings and for infections and what happens when you introduce vaccination. So let's explore what we might see for the same infection or for different settings by changing the value for R0. Slide 17 (18:42) I'm now going to return to the model. And explore what happens if we vary the value for the base reproduction number. Okay. So I'm going to run the model now again. And this is what we saw previously when the base reproduction number equals thirteen. Now suppose we change the base reproduction number to be eighteen. So that will be very high. And here we see that the cycles appear to occur more frequently than they did when the base reproduction number equals eighteen. Let's see this in a bit more detail. If we compare - if we overlay - the plots this is what happens if we have a base reproduction number of eighteen and we change it to be ... thirteen the blue line shows what we see. So here we see that the peaks occur less frequently for a lower value for the base reproduction number than when the base reproduction number is very high. We can explain why this occurs using some very simple calculations. Slide 18 (19:49) Now the proportion of the population that needs to be susceptible for an epidemic to occur is equal to one over R0 as we saw earlier. If the base reproduction number equals five, this proportion equals one over five. That is to say it is 0.2. If the base reproduction number equals thirteen this proportion equals one over thirteen or 0.07. The size of the susceptible population changes due to the entry of susceptible newborns and the removal of susceptibles through infection. Slide 19 (20:19) Now fewer newborns are needed for the proportion susceptible to reach 0.07 than for it to reach 0.2. So the timing between epidemics is shorter if R0 is high than if R0 is low. Suppose we now consider what happens if we introduce vaccination. As discussed, the proportion of the population that needs to be susceptible for an epidemic to occur equals one over R0 and the susceptible population changes due to the entry of newborns and the removal of susceptibles through infection. Now vaccination of newborns means that the rate at which susceptible are added to the population is reduced, so it takes longer to reach the threshold required for an epidemic to occur. This means that the time interval between successive epidemics increases with the vaccination coverage. Let's test this theory using a model. Slide 20 (21:11) This slide shows how we need to adapt the model that we had previously to include immunisation. The diagram is identical to what we had before except that now the arrow which was labelled just 'births' is relabelled to 'susceptible births' and we have another arrow reflecting 'immunised births'. So let's see what happens when we incorporate vaccination into our model. Slide 21 (21:33) So now on the demo. So this window shows the model diagram. And we have susceptible births coming in and immunised births coming into the immune category. Now suppose we change the vaccination coverage so at the moment the vaccination coverage is said to be zero and we run the model. This is what the model predicts. It's similar to what we saw before. Now suppose we change the vaccination coverage to be fifty per cent so I'm going to change the value for 'eff_vacc_cov' to be 0.5. And here we see when we introduce vaccination after one hundred years after the start of the model runs, the number of new infectious people goes down. Suppose we increase the vaccination coverage further. Suppose we increase it to be eighty per cent. And this shows what the model predicts. So we see that the number of new infectious people goes down but we also see that the time period between successive epidemics also increases. Suppose we increase the vaccination coverage to be ninety per cent. And here we see that - you can just vaguely see - here we can see that we can see an epidemic occurring after another period of about forty years and the instance has dropped to very low levels. Now, finally, suppose we look at what happens if we introduce vaccination at a coverage which is above the herd immunity threshold, say ninety-five per cent. And here we see that we've controlled transmission completely. It would also be interesting to look at what happens to the proportion susceptible and the proportion immune and the net reproduction number in a population. I'm now going to switch to a different plot ... and let's change the ... at the moment the coverage is set to be zero. Now let's change the coverage to be fifty per cent. So what we see here - this plot shows proportion susceptible in the population and this shows that once we introduce vaccination here, then the average value for the proportion susceptible remains unchanged. Now this may seem surprising since we've introduced vaccination. However, to explain this, note that the vaccination coverage was below the herd immunity threshold and therefore the infection is still endemic. You can see this if we go back to page one and here we see that transmission is still occurring even when proportion immunised is bigger than zero. So if the infection is still endemic this means that the average net reproduction number is equal to one because on average each infectious person is leading to one secondary infectious person. As we saw a few slides ago, the net reproduction number is in proportion to the proportion susceptible. So if the average net reproduction number remains unchanged this means that the average proportion susceptible would also be expected to remain unchanged after we introduce vaccination. But this only holds if the coverage is below the herd immunity threshold. Suppose we now change the effect of coverage to be over ninety-three per cent which was the herd immunity threshold which we calculated for our model. And here we see that the proportion susceptible actually goes down. And if we add what happens to the net reproduction number over time, you also that it goes down once you introduce vaccination. Slide 22 (25:53) So we have now seen how modelling can provide insight into the dynamics of infections and the effect of control. Now the examples that we have considered have been fairly simple. We have assumed, for example, that individuals mix randomly. We've also focused on the simplest effects of vaccination. We also know that vaccination has another effect. It reduces the amount of transmission and this leads to an increased proportion of individuals who reach adulthood still susceptible. If the force of infection or the rate at which the susceptibles are infected is still high then we can get into a situation whereby the number of new infections among adults can be greater than that before vaccination. Well, we don't have time to consider that in this lecture and you may like to read about that further elsewhere. So to use the models we've mentioned here, you may like to visit the website that's provided on this slide. You may also find it helpful to read further about modelling and the insights into the dynamics of an infection. There's an introductory book which has been published fairly recently. The book's website also includes additional model files and exercises that you can explore. For further mathematical details you may also want to read the book by Anderson and May published in 1991 which actually laid out much of the theory that is presented in this slide. Many thanks.

Coevolution with hosts

Tadpoles of the North American green frog (Rana clamitans) are parasitised by a trematode (Echinosterma sp.), the dispersal phase of which (cercariae) enter via the tadpole's cloacal opening and invade its kidneys. In a sample of 200 tadpoles, 97% had pathogens only in the right kidney. This lateral bias by the pathogen ensures that its tadpole host, which can survive the destruction of one kidney but not both, does not die, to the benefit of both host and pathogen. Lizards are often parasitised by ticks and mites, which tend to cluster around the host's eyes and ears and in skin folds, to the irritation of their host. The Spanish lizard Psammodromus algirus has a little pocket of skin on each side of its neck that provides an ideal home for the tick Ixodes ricinus. Unless the number of ticks on an individual lizard becomes very high, they remain mostly in the pouches, leaving other parts of the host tick-free. These and numerous other examples illustrate the intimacy that characterises many coevolved relationships between hosts and their pathogens. As emphasised in Unit 3, pathogen-host relationships are fundamentally different from predator-prey relationships in that both partners generally have a shared interest in the host's survival. As a result, adaptations have evolved in hosts that cannot avoid pathogens, and these changes mitigate some of the pathogen's adverse effects. For their part, many pathogens have also evolved adaptations that reduce the threat they pose to the health and survival of their host. However, most of the pathogens that you are studying in this module make people very sick and often kill them, so why are some pathogens benign and others lethal? In this section, you will examine aspects of the life history and ecology of pathogens and hosts that favour the evolution of coexistence, and consider whether this relationship is the inevitable evolutionary outcome of all pathogen-host associations. Coevolution, which was introduced in Block 1 Unit 8 (Section 6), is a word that is widely used in evolutionary biology, often rather loosely, in the context of intimate relationships between species, such as between plants and the insects that pollinate them. Strictly defined, coevolution refers to relationships in which the evolution of a particular trait in one species has led to the evolution of a particular trait in another. For example, in the context of infectious diseases, the evolution of the ability of a host to detect and respond to an infecting pathogen at an early stage has led, in many pathogens, to the evolution of adaptations that enable it to evade host defences. Later, you will think about how coevolution has led to increased specialisation. There are three contrasting models of how hosts and pathogens can coevolve. 1. Mutual aggression o Host and pathogen are engaged in an evolutionary 'arms race', with each species continually evolving in an aggressive manner towards the other (this was the example of coevolution given in Block 1 Unit 8). In the pathogen, selection favours greater exploitation of the host; in the host, selection favours more effective exclusion of the pathogen. 2. Prudent pathogen o Selection in the pathogen favours traits that limit the harm it does to the host. As a result, the host and pathogen survive longer. From the host's point of view, it can invest fewer resources in defence and so comes to tolerate the pathogen. Note that this model assumes host defences against pathogens are costly, an assumption to which you will return. 3. Incipient mutualism o Pathogen and host not only evolve towards doing one another less harm, but also evolve ways to benefit one another. These three models are well illustrated by the many microbes that make up the normal flora that live on and in animals, particularly in the gut. The number of individual microorganisms that make up the normal flora is enormous. The majority of these microbes appear to have no harmful or directly beneficial effect on their host and so illustrate the prudent pathogen model. For some animals, notably ruminant mammals such as cows, certain species of bacteria play a vital role in breaking down specific components of food, such as cellulose, which the host cannot digest itself. Such bacteria are mutualists and, if evolved from pathogens, illustrate the incipient mutualism model. However, occasionally, the gut may be infected by a harmful microbe, such as Escherichia coli O157 or Salmonella, which can kill its host; these organisms exemplify the mutual aggression model.

Geographical patterns

The communities associated with Trypanosoma show interesting geographical patterns. For example, the subspecies of Trypanosoma brucei in East Africa is transmitted by Glossina morsitans and causes acute sleeping sickness. In West Africa, another subspecies of T. brucei (gambiense) is transmitted by another Glossina species (Glossina pallipides, the tsetse fly) and causes chronic sleeping sickness. Finally, a third subspecies of T. brucei (brucei) is widely distributed in Africa, transmitted by the tsetse fly, and causes a disease in domestic stock and wild animals (nagana acute). To complicate the story further, this same disease appears to be caused by two other Trypanosoma species (vivax and congolense), spread by Glossina morsitans. The relatively subtle shifts in trypanosome vector in Africa (e.g. between Glossina species) contrasts with the one New World example, where the vector is an entirely different order of insect. The Leishmania communities in both the New and the Old World all involve sandflies with at least five different pathogen species and three forms of the disease. • What major difference in the composition of Leishmania and Trypanosoma associated communities is apparent from Figure 3.15? • Leishmania communities do not have an alternative vertebrate host, in contrast to many of the Trypanosoma communities. 5.2.1 Phylogenetic trees The interpretation of the evolutionary and ecological patterns of host-pathogen communities has been helped enormously by the analysis of phylogenetic trees (or phylogenies) derived from molecular data (RNA or DNA sequence data, e.g. from ribosomal RNA, or rRNA). These trees show not only what is related to what but also the closeness of that relationship. The statistical analyses used to construct them are extremely complex and you do not need to know about them for the purposes of this module. However, the cautionary tale is that there may be many, equally likely phylogenetic trees. Thus you should treat a phylogenetic tree as the latest in a series of hypotheses about the relationship between the species in question. Phylogenetic trees as predictors of the age of relationships The other important result that can be derived from a phylogenetic tree is the age of the relationships, i.e. given certain assumptions (such as the rate of mutation), they can be used to estimate the time before present at which species diverged. Therefore, in theory, we should be able to estimate how many million years ago Leishmania split from Trypanosoma (assuming they are both derived from a common ancestor). The phylogenetic tree for trypanosomes is depicted in Figure 3.16. This diagram shows that the trypanosomes were monophyletic, i.e. derived from a common ancestor. It also shows the ancient divergence of the so-called 'aquatic clade' from other groups. (A clade is a group of closely related species with a common ancestor.) The aquatic clade comprises species of trypanosome found in marine and freshwater fish and amphibians. These trypanosomes are spread by aquatic leeches. Coevolution of trypanosome vectors Indeed, one of the important findings of the phylogenetic analysis is to show the coevolution of the pathogens and their vectors - thus, closely related trypanosomes (comprising a single clade) tend to be transmitted by the same closely related set of vectors. But it is also predicted that the trypanosomes will coevolve with their vertebrate hosts. This has also been supported by the phylogenetic studies: for example, most of the species in the T. cruzi clade are associated with South American mammals. The exceptions to this are an unnamed species that infects kangaroos and two species of European bat trypanosome. • Why would the kangaroo and European bat trypanosomes be related to the other trypanosomes in the T. cruzi clade? • In the case of the kangaroo, this is because Australia was joined to South America via Antarctica much later than South America was joined to Africa. Thus, the T. cruzi clade may have evolved across South America, Antarctica and Australia. The European bats are mobile and may have been able to disperse over long distances (possibly linking South America and Africa). Coevolution of trypanosomes and hosts The dating of the divergence of these clades suggests that trypanosomes have coevolved with their hosts over several hundred million years. For example, the T. brucei clade is predicted to have diverged from other clades during the Permian Era, when reptiles were the most advanced vertebrates and certainly when none of the current hosts were present. Therefore, as humans evolved in Africa, they were among trypanosomes whose evolution was already several hundred million years old and for which movement into a new (and increasingly abundant) primate host was simply a matter of time. In South America, as humans migrated into the continent 30 000-40 000 years ago, the T. cruzi trypanosomes also had a short host leap from their existing host range, which included primates.

Ecological attributes of host-pathogen interactions

The compartmental model of malaria infection and the potential cycling in abundance of pathogen populations show that the emergent dynamic properties of the host-pathogen system depend on details of the interactions between all the linked populations. In this final section, we will summarise aspects of the ecology and biology of the individuals and populations that characterise host-pathogen interactions and influence most heavily the abundance of the pathogen. 6.1 Body size It is notable that in all host-pathogen interactions the host is very large compared with the pathogen. Where vectors are involved, the vector is of intermediate size. Body size is important for the following two reasons. • Physical constraints: a pathogen needs to survive and reproduce within or among the physiological apparatus of its host. In the most extreme examples - such as viruses - the pathogens need to access molecules such as enzymes, ATP and amino acids from within the cell of its host. This places a clear constraint on pathogen size. • Generation time of the organism: body size is strongly correlated with the generation time of the organism. The smaller the body size, the more rapid the generation time. The pathogen needs to have a more rapid generation time than its host. Otherwise, it will fail to reproduce before the host dies. In many host-pathogen interactions, the pathogen also needs to have a much more rapid generation time than that of its host because it needs to reproduce before the host's defence system has detected it. Ideally, it will have completed several replication cycles before the body has begun to respond. 6.2 Forms of transmission There are two forms of transmission: active and passive. Active transmission Many of the most effective pathogens use active forms of transmission from host to host. This includes using insect vectors and causing changes in the host's physiology and/or behaviour to elicit transmission. • Give an example of changes in the human host's physiology and/or behaviour caused by the pathogen that favour its transmission. • Examples include sneezing (influenza), skin lesions (cowpox, smallpox) and diarrhoea. These changes in host physiology and behaviour may also be linked to using the waste products of hosts (urine, faeces) and causing a more rapid expulsion of these products, as was illustrated for the Vibrio cholerae bacterium in the Cholera Case Study (Section 2). Changes may also be associated with a site specificity on the host that maximises transmission, e.g. syphilis around the genitalia. Transmission may also be increased by aggregation of the host, leading to a higher local density of hosts. Aggregation may be due to work (e.g. malaria spreading through mining communities), poor and crowded housing (TB), lack of sanitation (cholera) or social behaviour (STIs, including HIV). Note that multiple sexual partners constitute a form of aggregation - not necessarily aggregation in one place at one time but aggregation over a short time period. The ecological parallel of transmission in populations that are not pathogenic is dispersal from one suitable habitat to another. Many of the most abundant species in the world are those that can successfully move between habitats. This is especially true today with the increasing fragmentation of habitats. A patchy habitat is essentially what the pathogen experiences - a habitat comprised of a set of individual hosts separated in space. In this way, it is no different from a fly species that reproduces in patches of dung. It must first find the dung and lay eggs in it, complete its life cycle before the dung dries out and then move on to the next patch. Passive transmission The active aspects of transmission have so far been emphasised. But what is passive transmission and are there any examples among pathogens? In passive transmission, pathogens will return to the environment and be picked up by passing hosts. This environment might include areas of soil, vegetation and water bodies. Ticks and some other ectoparasites, e.g. leeches, use this method of dispersal between hosts. They feed on a host, drop off into the vegetation or soil and then later get picked up by a passing host. In some cases, they may move a short distance to maximise their chances of being picked up (they are part active). While these ectoparasites are not infectious agents, they can be vectors of infectious agents, such as Lyme disease. 6.3 Co-occurrence Pathogens need to occur in the same areas and habitats as their hosts. You have already seen how some pathogens exist within a smaller fraction of the geographical range of their host. There are a number of ways in which pathogens cope with this requirement, and these centre around the essential resources used by the host. Exploitation of host's water resources One way in which pathogens can increase their chances of interaction with their host is to exploit the host's water resources. Alternatively, the pathogens exploit vectors or intermediate hosts associated with these resources. • Give two examples of vectors or intermediate hosts associated with water bodies. • Anopheles mosquitoes (malaria vector) and aquatic snails (schistosomiasis or bilharzia intermediate host). In both of these examples, the vectors and pathogens are associated with water bodies close to human habitation or places of work (thereby increasing co-occurrence). In the case of schistosomiasis, the cercariae and their vectors are especially prevalent in irrigated fields where crops are grown or shallow rivers where people wash. Pathogens and vectors associated with water bodies are of enormous global importance in terms of the numbers of humans infected and the amount of mortality or serious illness. Of particular note is the number of diseases associated with mosquito transmission (dengue, malaria, filariasis, yellow fever). When this is extended to other biting Diptera (flies), thereby including leishmaniasis (sandflies) and onchocerciasis (blackflies), the importance of water as a breeding area for flies as vectors of disease becomes overwhelming (Figure 3.17). The reasons for the success of the pathogens that cause these diseases lies in their utilisation of a vector that lives part of its life in water and part of its life aerially, feeding off humans. The pathogen therefore maintains an intimate relationship with both the individual host and one of his or her most essential dietary and domestic requirements - water. Moreover, these Diptera are able to utilise water bodies created by their hosts. Exploitation of host's food resources There are also examples of pathogens that are not exclusively associated with water and may be spread through contaminated food. Ascariasis, typhoid (Salmonella typhi) and cholera are all spread via food (contaminated hands, flies or utensils) and water contaminated with faeces and/or urine (in all cases, transmitted through the faecal-oral route). Thus, the resources of food and water are linked in these examples. 6.4 Specialisation It is expected that pathogens will specialise on one or a few closely related hosts. This prediction results from a consideration of coevolution with pathogens developing ever more complex methods of living in hosts and overcoming their defences. The evolutionary responses of the hosts ensure that many pathogens become locked into an increasingly specialised relationship with one host or a few host species. Again, there are ecological parallels among non-pathogenic species. For example, herbivorous insects often specialise on plant species that are highly toxic to generalist predators. Through coevolution, the plant feeders have overcome the defence systems of the plant (and the plant may have, in turn, evolved more toxic defence systems - which the herbivores have evolved to overcome, and so on). In some cases, the herbivores have used plant defence to their advantage, i.e. to defend themselves. 6.5 Overview and implications for emerging diseases The four criteria that have been highlighted in this section represent ecological and evolutionary constraints on the parasites and hosts. You can take the argument further, and predict the most effective pathogens based on maximising or minimising these constraints. Thus, you could predict that the most abundant ('successful') pathogens would be those in which: • body size is minimised (and replication rate is maximised) • active transmission and co-occurrence are maximised • host (or part of host) specialisation is maximised (especially with respect to defence against hosts or detection by hosts). However, this is overly simplistic and you need to consider trade-offs. For example, smaller body size may be beneficial for faster replication but also carries costs if genome size is reduced. This is apparent in viruses: those with large genomes carry sophisticated host defence countermeasures, but replicate slowly. Viruses with small genomes rely on fast replication and have no spare genes for defence countermeasures. Another way of looking at these criteria is that they represent an evolutionary route for pathogens. Thus, we might predict that pathogens will evolve from large to small body size, from passive to active transmission, from high to low detection and from generalist to specialist. Body size is important for within-host dynamics because it affects reproductive rate, survival and movement in the host. Active or passive transmission is clearly relevant to between-host variables. Co-occurrence is relevant to between-host variables, ensuring that all the members of the host-pathogen community coexist in the same place at the same time. Finally, specialisation, through coevolution, underpins the host defence countermeasures of the pathogen.

Ecology

The study of the relationships between organisms and their environment.

Making sense of mathematical equations

The first thing to do when confronted with a mathematical equation is to look at it and make sure you understand what it is saying. For example, the equation R0 × S = 1 means that if you multiply the reproduction number, R0, with the proportion susceptible, S, you must get 1. Here R0 is the reproduction number of an endemic infection, so it must be greater than 1, while S is a proportion, so it must be less than 1. So, for example, if R0 = 3 then S = 1/3. Or if S = 0.2 then R0 = 5. You might then ask yourself, so what? The reason this equation is important (and worth spending some time on) is that R0 is an important quantity. However, its definition is rather theoretical (see Unit 2 Section 3 of this block), and it is far from clear how to calculate R0 in practice. This equation ties it to S in a very direct way: if you know S, you must also know R0 (at least for homogeneously mixing populations). And the beauty of it is that S is relatively easy to estimate. The next equation makes this explicit: This equation is obtained from the previous equation by dividing both sides by S. The advantage of this second expression is that it puts the relationship in a directly useable form. Given an estimate of S, you can substitute for it in the equation and read off the value of R0. Thus, if S = 0.5, R0 = 1/0.5 = 2. For example, suppose that it is found that 93% of a large, homogeneously mixing population is immune to measles. Since S = 1 − 0.93 = 0.07, it follows that: Thus the basic reproduction number for measles in this population is about 14. This means that, if the population were wholly susceptible, one infectious individual in this population would infect, on average, about 14 others. • Suppose that 62% of a large, homogeneously mixing population is immune to hepatitis A virus infection. Calculate the basic reproduction number for hepatitis A infection in this population. • S = 1 − 0.62 = 0.38. Hence: Thus, if the population were wholly susceptible, one case infected with hepatitis A virus would infect, on average, 2.6 others. Herd immunity level The herd immunity level describes the level of immunity of the population before mass vaccination is introduced. Because the measure describes the proportion of a population that is immune, it can be expressed as 1 minus the proportion of susceptibles in the population (1 - S), and as S = 1/R0, the herd immunity level can be written as: Consider now what happens if a mass vaccination programme is implemented in this population. Assume for simplicity that vaccination occurs close to birth and that some proportion, q, of the population is immunised. If this proportion is less than the pre-existing herd immunity level then the dynamics of this infection are perturbed, but readjust eventually and settle down again. However, if the proportion of the population immunised through vaccination is greater than the pre-existing herd immunity level then there are too few susceptibles left for transmission to be sustained. If there is no immigration of infected cases, the infection eventually disappears. Thus, the herd immunity level of 1 - 1/R0 represents a threshold: increasing the immunity level in the population beyond this level will lead to the elimination of the infection. The herd immunity threshold is also called the critical immunisation threshold, qc, and is expressed as: This is the minimum proportion of the population that must be immunised at birth (or close to birth) for the infection to be eliminated. Activity 4.1 Herd immunity threshold mini-lecture Allow 1 hour Now watch the mini-lecture by Emilia Vynnycky on 'Herd immunity and the effects of vaccination' (Video 4.1). When you have finished watching it, use the information provided to determine the values of qc for: 1. measles (R0 = 10 to 20) 2. mumps (R0 = 5 to 10) 3. smallpox (R0 = 3 to 5). What is the key assumption in calculating the value of qc? Note that this video introduces a new concept, the net reproduction number (Rt), which is defined as the average number of secondary infections that result from each infective at time t, taking into account incomplete host susceptibility to infection and any control measures that are in place. If Rt is greater than 1 at time t, the epidemic will (in general) be increasing. It is analogous to the basic reproduction number, which you encountered in Block 3 Unit 2, except the basic reproduction number encompasses a 'totally susceptible' population and assumes no control measures. Note also that, as in Unit 2 Activity 2.1 of this block, this video contains footage of the Berkley Madonna software being used. Indeed, the final slide of the presentation lists three models that have been demonstrated, and these are available for SK320 students to download and explore. If you wish to access these modules you can download them from the links below or from the 'Study resources' section of the module website. • Click here to download 'epidemic_model-OU.mmd' • Click here to download 'longterm_epidemic_model-OU.mmd' • Click here to download 'lvaccination_model-OU.mmd' Video content is not available in this format. Video 4.1 Mini-lecture on herd immunity and vaccination View transcript View answer 1.3.2 Why R0 is relevant to endemic infections At this point, you might be wondering why we bother to calculate R0 for endemic infectious agents, since the actual population is certainly not 'completely susceptible' as required in the definition of R0. The reason is that R0 summarises the infectious potential of the infective agent in this population, which depends on the agent, its infectious period, and the rate at which effective contacts occur in the population. This information, in turn, is used to quantify the vaccination level required to eliminate infection. • Consider an infection in the endemic steady state. It might appear more relevant to calculate the average number of cases directly infected by a single infectious case in the actual population in which the infectious agent is endemic, rather than in an abstract, 'completely susceptible' equivalent population. Use the logic of this section to explain why this is not the case. • The reason is that, for an infectious agent in the endemic steady state, the average number of cases directly infected by one case must be equal to 1. This is true of all endemic infectious agents, and so this apparently more 'relevant' quantity is, in fact, of no use at all for contrasting different infectious agents.

Epidemicity

The pattern of regular epidemic waves exhibited by many infectious diseases. A striking feature of many infectious diseases, particularly the common infections of childhood, is that they occur in regular epidemic waves, often separated by a period of several years. This phenomenon is known as epidemicity. The interval between successive epidemics is called the inter-epidemic period, or epidemic period for short, and will be denoted T. The regularity of these epidemics is remarkable. For example, prior to the introduction of measles vaccine in 1968, epidemics of measles occurred in the England and Wales every two years, as shown by the graph of notified cases in Figure 4.5. Similar epidemic cycles can be observed for other infections: for example, Figure 4.6 shows a time series of whooping cough notifications. • From Figure 4.6, give a rough estimate of the epidemic period for whooping cough England and Wales. • The epidemic period is about 3-4 years. Both measles and whooping cough display annual as well as longer epidemic cycles (these annual cycles are reasonably clear for measles in Figure 4.5 but rather less clear for whooping cough in Figure 4.6). It is important to distinguish between the two types of cycles, as they have different mechanisms. Annual epidemic cycles Regular annual cycles in incidence may be due to changes in the contact rate or transmission risk, which vary seasonally. For example, the incidence of measles has been shown to peak during school terms, and to decrease during school holidays. This is a clear indication that transmission between children at school plays a key role in the epidemiology of measles. Longer epidemic cycles Seasonal effects cannot explain the occurrence of epidemic cycles at intervals of more than one year. To understand why, and when, such cycles happen, it is necessary to return to the SIR model of Unit 2. Suppose that an infectious individual is introduced into a large susceptible population, in which R0 > 1. • The number of infected individuals, I(t), will grow initially as the infection spreads through the population. • At the same time, provided of course that the infection confers immunity, the number of susceptibles, S(t), will drop as more people are infected. • Eventually, the number of susceptibles will drop to such an extent that most of the contacts made by an infectious individual are with immunes, and hence do not result in new infections. • Eventually, the number of new infections will drop. At this point two scenarios can develop: • If the number of new infections falls rapidly compared with the birth rate, I(t) may eventually drop to zero. This is the phenomenon of stochastic extinction (see Box 4.1). • However, if the birth rate is high enough, the susceptibles, S(t), will be replenished sufficiently rapidly to stem the drop in I(t) before it reaches zero. Thus the incidence of infection will bottom out and begin increasing again. This is the pattern observed for infections like measles and whooping cough in the UK, prior to the introduction of vaccination. 4.1 The inter-epidemic period The epidemic cycle relates directly to the balance of infectives and susceptibles, which is to a large extent governed by a few key parameters (at least in homogeneously mixing populations). It should come as no surprise, therefore, that the inter-epidemic period, T, is related, at least approximately, to some of these parameters. Specifically: where • A is the average age at infection • D′ (pronounced 'dee-dash', or 'dee-prime') is the average latency period • D is the average infectious period • and π is the constant 3.142.... Note that A, D′ and D must be expressed in the same units. Recall that the latency period of an infection is the period between the point at which an individual becomes infected and the time at which that same individual becomes infectious. Therefore, the sum of the average latent period and half the average infectious period (i.e. D′ + (D/2)) is the average time between an individual becoming infected and infecting others; this is sometimes called the serial interval of the infection. The role of the force of infection To make sense of this relationship, note first that the faster the infection spreads in the population, the shorter you would expect T to be. One factor affecting the speed at which infection spreads is the force of infection: the more people are infected per unit time, the faster the spread. Thus infection spreads more rapidly if the force of infection, λ, is high, and hence if the average age at infection A is low (since A = 1/λ). Thus T should decrease as A decreases. This is what the formula for T implies: as T and A are on opposite sides of the equation, so a decrease in A has a similar effect on T. Serial interval Another factor affecting the speed of spread is the serial interval, D′ + D/2. The shorter the time between successive generations, the quicker the spread will be. Thus T should decrease as D′ + D/2 decreases. Again, this is what the formula for T implies. Finally, do not worry about the other aspects of the equation, such as the square root or the number 2π (2π is there because this is a cyclical process - if you have studied geometry, you may recall that the number π is part of the geometry of a circle). The key point is to check that the equation makes sense to you from a qualitative point of view. Note also that the equation is an approximation, which breaks down as the infectious period increases: infections with long infectious periods tend not to display regular epidemic cycles. Table 4.5 gives the values for A, D′ and D for some common infections of childhood in a number of high-income countries. Table 4.5 Typical values of A, D′ and D for selected countries. These values are indicative: A varies between populations, and D′ and D vary between individuals. Infection A/years D′/days D/days Measles 5 5 5 Mumps 5 8 12 Rubella 8 10 12 Varicella zoster 7 10 8 Pertussis 5 9 21 For measles in the UK, D′ + D/2 = 7.5 days = 7.5/365 ≈ 0.021 years. Prior to mass vaccination, the average age at infection was about 5 years. Hence: This accords very well with the observed inter-epidemic period of 2 years. • Use the data in Table 4.5 to calculate the inter-epidemic period of whooping cough (Pertussis). Does this calculation match the observed cycle in Figure 4.6? • The observed inter-epidemic period varies between 3 and 4 years, so again theory is in agreement with observation. The introduction of mass vaccination can have a direct impact on the epidemic cycle. Since vaccination increases the average age at infection, the inter-epidemic period, T, will lengthen, and regular epidemics might even disappear entirely. A final note on modelling epidemics Epidemic models have become an essential tool of epidemiologists, and are commonly used to investigate the ways in which an infection spreads through a population, and to explore the consequences of vaccination strategies. Many epidemic models are based on the compartmental SIR model (see Unit 2 Section 5 of this block). Simple mathematical equations can be used to describe the progress of an infection through a population. These models, known as deterministic models, successfully replicate most of the features of infectious diseases. One feature they cannot replicate, however, is sustained epidemic cycles: the oscillations of deterministic models eventually settle down to an equilibrium level, as shown for example in Figure 4.7. This is a phenomenon that is not observed in real life. Another feature not reproduced by deterministic models is stochastic extinction (Box 4.1). More generally, when the number of cases is small, it is necessary to allow for random fluctuations. More complex models, known as stochastic models, have been developed to take account of such random effects.

Critical immunisation threshold

This is the same this as the herd immunity threshold: the minimum proportion of the population that must be immunised at birth (or close to birth) for the infection to be eliminated.

Trematodes

Trematodes are a class of animal within the phylum Platyhelminthes that contains two groups of parasitic flatworms that are described as flukes. There are two groups of trematode: the Aspidogastrea (obligate parasites of molluscs, turtles and fish) and the Digenea (obligate parasites of molluscs and vertebrates). The latter are the focus of Sections 3.1 and 3.2. A good starting point for examining the cycle is the release of eggs from the definitive human host. Depending on the species of schistosome involved, mature paired worms are found in the blood vessels of the bladder (S. haematobium) or the intestines (S. mansoni), and the females release fertilised eggs from these sites. Much of the pathology of the infection is caused by the way in which eggs exit the host. They have sharp spines (shown diagrammatically in Figure 8.7 and as micrographs in Figure 8.8) that tear through the blood vessels of either the bladder wall into the urine (hence the appearance of blood in the urine in S. haematobium infestations), or through those of the intestinal wall into the gut contents, where they are passed in the faeces. If passed into freshwater, the egg releases a free-swiming miracidium that finds and invades a water snail host. Unlike many flukes, no redia stage is produced by S. haematobium or S. mansoni, but prolific asexual reproduction in the snail results in further (daughter) sporocysts and many thousands of cercariae, which are released into the water. As each snail is probably infected by many miracidia, hundreds of thousands of cercariae may be released periodically into these watercourses. Cercariae of S. mansoni swim by means of powerful tails (Figure 8.9) and, unlike most flukes, do not enter another host to encyst as metacercariae, but penetrate directly through the skin of a human host, usually via a hair follicle: a process that takes less than a minute! Once in the human host, the tail is lost and these young immature flukes migrate in the circulation via the lungs (within a week of infection) and heart, to the liver (2 weeks after infection) where they mature and pair up with a fluke of the opposite sex. This is another difference between S. haematobium, S. mansoni and other types of fluke: there are separate sexes in the case of the two species considered in this section. The two organisms remain 'in copula': the short, stubby male holding the longer, thinner female in a special ventral, gynaecophoric canal (Figure 8.10). The pair then migrate to the blood vessels around the intestine or bladder, depending on species, and begin egg production 3-5 weeks after the initial infection. S. mansoni females produce around 300 eggs per day, whilst S. haematobium can achieve 3000 eggs per day. Because adult worms live for an average of 5-7 years, and as there may be a number of pairs in the blood vessels, many thousands of eggs are released each day from an infected individual. In many LMICs, especially in the rural areas, there is no safe, clean water or good sanitation procedures. The accessible water is used intensively by members of the community for bathing (Figure 8.11), washing and crop irrigation. It is not hard to imagine that such water becomes contaminated with urine and faeces (directly or via surface runoff from latrines) and, should any of the users be harbouring blood flukes, eggs will be released into the water. Examine Figure 8.12, which provides some data on incidents of blood fluke infections in relation to an infected water source in a region of Nigeria. • What conclusions can you draw concerning the prevalence of infection in these communities and how the disease may enter the water source? • Prevalence appears to be closely related to the distance of each community from the nearest infected water - although it rises between 410 m and 600 m, it falls off with distance outside this range. • In this case, as S. haematobium infection (which infests the bladder) is more prevalent than S. mansoni infection (which infests the gut), urine contamination of infected water is more likely than faecal contamination. • From your knowledge of the life cycle of S. haematobium and S. mansoni (see Figure 8.7), how is the snail intermediate host involved? • Eggs hatch to release a free-swimming miracidium larva. This larva encounters a snail (probably by chemoattraction), penetrates the tissue and develops into a first stage (mother) sporocyst larva. These develop into cercariae larvae that are released from the snail. • What three conclusions can you infer from the data concerning the infection in children and adults? 1. Children are not immune to infection and probably become infected when old enough to play in infected water (the peak of infection is in the early teens). 2. The decline in prevalence of infection with age during adulthood is probably due to reduction in contact with water associated with changed leisure and work patterns. 3. After the age of about 15 there is a decrease in the amount of eggs released in the urine. The lower egg output in older groups of individuals suggests that adults may have acquired immunity to infection. A considerable amount of research has been undertaken concerning natural immunity to schistosomiasis, and this is summarised in the review article in the 'Further reading' section for this unit (see the 'Study resources' page of the module website). Human hosts appear to be able to resist further infections, even though adult worms survive in their blood vessels. This phenomenon is known as concomitant immunity, where the immune response acts against new invading cercariae. Why are the adult worms not challenged successfully by the immune system in this situation? One answer lies in the fact that worms may adopt a disguise by coating themselves with host macromolecules. More recent work has suggested that some worms may be able to suppress their own antigens or even release substances that may inhibit or, at least, suppress their host's immune response. Individuals demonstrating a resistance to infection have high levels of IgE antibodies whereas those who are susceptible show no increase in IgE but elevated levels of IgG and IgM antibodies. (The various types of antibodies and their roles in protecting against infections and parasites are discussed in Block 2.) Immune response and its effects The body's immune response to schistosomiasis is also part of what makes the condition such a ravaging disease. Eggs can be swept away by blood flow and end up in a number of body locations, particularly the liver and spleen. The host's immune cells then attack the eggs forming cellular granulomatous masses around them. This process traps the eggs in the tissue that they are found on, leading to fibrosis and a huge increase in size of both liver and spleen: hepatosplenomegaly (Figure 8.14). Such symptoms occur a few years after the initial infection, are extremely debilitating, and occasionally cause death. Associations between schistosomiasis and the occurrence of some cancers have also been identified. 3.1.3 Control and prevention Next to malaria, schistosomiasis is the world's most debilitating parasitic tropical disease and consequently a good deal of effort has gone into methods of its control and prevention. • Consider the schistosome life cycle (Figure 8.7) and suggest four key places where control could be effective. 1. Prevent contamination of water by eggs, by providing proper sanitation and clean water. 2. Eliminate the snail host (although in practice this is virtually impossible, given the limited chemicals available to kill molluscs). 3. Prevent humans entering water where infected snails have been identified. 4. Administer chemotherapeutic drugs to infected individuals. In some parts of the Caribbean, prevalence of schistosomiasis has decreased due mainly to improved sanitation and education of the communities, but in other areas of the world it has increased. The construction of dams and the introduction of irrigation schemes to improve water resources in Africa have led to increasing transmission of S. mansoni in places like Senegal and Ghana, as the changes have extended the range of the snail vector. There is concern that large-scale schemes such as the building of the Three Gorges Dam in China may also have a similar effect on the prevalence of S. japonicum. It is important in any control programme to take into account the habitats of the intermediate host. For example, Oncomelania spp., the snail intermediate hosts of S. japonicum, favour shallow water as found in lake edges and marshes. They are unable to withstand desiccation. Equally, Bulinus spp. snails, intermediate hosts of S. haematobium and S. intercalatum, prefer slow flowing water and a depth of less than 2 m and are often found associated with dams and irrigation channels. Lastly, Biomphalaria spp., the snail hosts of S. mansoni, are commonly found in sluggish waters. Both Bulinus and Biomphalaria are able to aestivate (assume a dormant state in hot, dry conditions) to withstand periods of drought while all three groups need well oxygenated water and the presence of vegetation. Knowing about such factors can inform any decisions about how pathogens or their vectors could be influenced in order to reduce the risk of infection to humans. Control is also hampered by the presence of reservoir hosts. S. japonicum has been identified in over 40 different species of wild and domestic animals in South East Asia and these play an important role in maintaining the parasite, especially in areas where there are also high populations of the snail host. Dogs help to maintain the transmission of S. mekongi to humans by providing a suitable reservoir of infection. Reservoir hosts are less important in Africa. Although both S. haematobium and S. intercalatum infect some animal species these are not thought to play a significant role in maintaining the parasites. In South America, S. mansoni infects rodents, opossums and cattle but the role of these animals in maintaining the parasite is unknown. Drug treatments Three types of antihelminitic drug are used against flukes: ivermectin, oxamnaquine and praziquantel (PZQ); although only PZQ is currently recommended by the WHO (2006). PZQ has been used successfully since the 1990s to control schistosomiasis. It is a cost-effective control approach as it costs around US$0.30 per person for a full course of treatment (WHO, 2011). However, PZQ is only effective against the blood stages of flukes and reinfection is not prevented by a single drug dose. It appears to have three distinct modes of action against flukes. 1. Disrupting calcium homeostasis by increasing the permeability of schistosome cells to calcium ions. This induces muscle contraction and paralysis of the parasite, preventing it from feeding. 2. Inhibiting adenosine uptake, thereby inhibiting DNA synthesis (schistosomes are unable to synthesize purines and require dietary adenosine). 3. Flukes that are killed by PZQ are attacked and digested by host macrophages, exposing their surface antigens to the immune system. This boosts the levels of schistosome specific antibodies, preventing further reinfection. Thus, three doses of PZQ given during childhood are usually sufficient to prevent disease in adulthood. PZQ is also effective against tapeworms, which are the topic of Section 4. Vaccination strategies Attempts have been made to create vaccines against schistosomes with little success, and at the time of writing (2011) only one candidate has progressed into clinical human trials (a vaccine against the 28 kDa GST (glutathione S transferase) from S. haematobium). Other promising candidates being tested in mouse models are antibodies against various transmembrane tegument proteins. It is hoped that since the publication of the genome of both S. mansoni and S. haematobium some kind of recombinant subunit vaccine or DNA vaccine can be produced in the near future. The situation with S. japonicum is slightly different. As transmission is zoonotic it is possible that a transmission-blocking vaccine could be produced for use in livestock animals, particularly bovines, to reduce transmission to humans (McManus and Loukas, 2008). In the 150 years or so since these fascinating, though dangerous, parasites have been studied scientifically, much has been revealed about their life cycles, epidemiology and pathobiology, yet they still remain a scourge of their human host. The challenge for future parasitologists is to convert growing molecular understanding into effective prevention measures.

Risk factors

Variable associated with an increase in the risk of a specified event.

HPV

A century ago, epidemiologists noted that cervical cancer was common in female sex workers but very rare in nuns - except for nuns who had been sexually active before entering the convent. They also observed high risks of cervical cancer in women married to men whose first wives had died of cervical cancer. From this epidemiological evidence, it could be deduced that cervical cancer was caused by a sexually transmitted infectious agent. In the 1970s, this agent was identified as human papillomavirus (HPV), which was already known to cause cutaneous and genital warts. It has since been demonstrated that HPV is implicated in over 90% of cervical cancers. Despite advances in methods for detecting and treating cervical cancer, this disease remains a leading cause of cancer death. Globally, approximately 16 per 100 000 women are affected with about 9 per 100 000 dying. Developing a safe and effective vaccine against HPV could thus have an enormous public health impact.

SIR model

A compartmental model with three compartments: susceptible, infected and recovered.

Concomitant immunity

A condition in which hosts appear to be able to resist further parasitic infections, even though adult forms survive in their blood vessels. The immune response acts against new invading parasites.

Biovars

A group of bacterial strains that are distinguishable physiologically or biochemically from other strains of the same species.

Biotypes

A group of organisms with the same genetic constitution.

Mutual aggression

A model of coevolution in which the host and the pathogen are engaged in an 'arms race' where each evolves in a way that is likely to be detrimental to the other.

Sentinel surveillance system

A surveillance system based on a selected number of committed participants.

Active surveillance system

A surveillance system in which information is actively sought from clinicians or patients.

Oocyst

A zygote produced by fusion of gametes in the apicomplexans (protists), which produces infective haploid spores by meiosis (sporogony).

American Trypanosomiasis u

Chagas' disease affects around 10-12 million people. American trypanosomiasis is caused by Trypanosoma cruzi, which is transmitted by various species of reduviid bug. This insect is also called the 'kissing bug' because it is attracted to carbon dioxide exhaled in the breath and often bites around the mouth. The kissing bug defecates in the wound from which it has just fed. The metacyclic trypomastigotes are shed in the bug's faeces and introduced into the body through the wound. Alternatively, the faeces remain on the skin until, in response to the irritation the bite causes, the victim rubs them into the bite wound. The first sign of infection is a painless, purplish swelling at the site of the bite. This is often called a skin chancre, chagoma, or sometimes Romana's sign. The parasites enter various cells of the body including macrophages and tissue cells (particularly heart muscle and oesophageal muscle cells). Once inside a cell they shed their outer membranes along with their flagella and round up, forming an amastigote: the morphological form found in intracellular environments. This is the reproductive stage of the parasite's life cycle, and they replicate by binary fission typically in macrophages in the bloodstream but also in other cells particularly in muscle cells. The infected cells eventually rupture releasing the progeny into the blood as trypomastigotes (the infective stage). The cycle of the parasite existing as trypomastigote and then dividing amastigote continues for several weeks. Any reduviid bug that feeds on this person may now pick up the trypomastigotes. The acute phase of the infection lasts for 4-8 weeks after which time the disease becomes chronic with the parasite continuing to multiply in the tissue cells although more slowly. The most common cause of death in Chagas' disease is heart failure occurring many years after initial infection and due to the considerable damage caused by the amastigotes. In the midgut of the insect vector the trypomastigotes transform into epimastigotes - the dividing stage. After repeated binary fission, the epimastigotes move to the hindgut and attach to the gut wall. Here they change into the infective metacyclic trypomastigotes, detach from the hindgut wall and are passed out in the faeces. The full life cycle is shown in Figure 6.6b. If the disease is in the acute phase, diagnosis is possible by detection of the trypomastigote stage in a wet blood smear. A stained thin smear may also detect parasites if there is high parasitaemia (see Box 6.1) during this phase. In the chronic phase when the parasites are hidden in tissue cells serological techniques such as ELISA and indirect haemagglutination assay are used to detect antibodies. (The different types of haemagglutination assay are detailed in Unit 9.) Treatment with benzimidazole and nifurtimox will kill the parasites in the acute phase but the treatment lasts 60 days, has frequent side-effects and has a cure rate of only 80%. Efficacy of both drugs decreases as the disease progresses making it important to identify an infection as early as possible. T cruzi, chagas disease, vectors are Rhodnius prolixus, Triatoma spp(Reduviid bugs), Contamination with infected bug faeces (classified as a stercoraria parasite). Also by blood transfusion and congenitally. Intracellular in many cell types but preference for macrophages, spleen and non-skeletal muscle such as cardiac muscle. Amatigote form in host cells. Trypomastigote moves between host cells. Trypomastigote and epimastigote in bug vector. Romana sign and/or Chagoma at site of bite. Most are asymptomatic. Some have flu-like symptoms. Lasts 3-4 months. No detectable parasitemia (the number of parasites in the blood). Lasts decades. Cardiomyopathy (progressive destruction of the myocardium). Lifetime Inflammatory response. Later on, immune suppression. Domestic animals, especially dogs. Wild animals including opossums, raccooons, armadilloes, guinea pigs. Nifurtimox and benznidazole. Insecticide treatment of houses with deltamethrin. Blood screening to prevent transmission by blood transfusion

Cestodes

Cestodes are the other class of parasitic flatworms within the phylum Platyhelminthes. They inhabit the intestines of vertebrates as adults and the bodies of various animals as juveniles. Amongst helminth parasites, tapeworms are perhaps the least pathogenic to humans, and they occur in high- as well as low- and middle-income countries. Adult tapeworms live attached between gut villi in the intestine, taking up digested food through a specialised body wall, very much in competition with the host's intestinal cells. Tapeworms have become so well adapted to their environment that they have lost their mouth and digestive tract. Indeed, in late Victorian times it was considered fashionable to harbour a tapeworm as an aid to slimming! Taenia solium (pork tapeworm) and Taenia saginata (beef tapeworm) both fall into this category. Humans are the only definitive host of these two species and become infected by eating undercooked pork or beef in which the cysticercus larva (Figure 8.3b) has developed. • How would swine and cattle have come to harbour the cysticercus stage? • By picking up eggs shed in human faeces and perhaps spread on pasture or soil as fertiliser (see Figure 8.1b and Figure 8.3 for a reminder of tapeworm life cycles). The eggs hatch in the small intestine of pigs or cattle, releasing the hexacanth, which penetrates the intestinal wall using its six hooks, enters the bloodstream and migrates to muscle tissue. Here the larva encysts, becoming a small, pea-sized cysticerus. Cysticerci can survive for many years in the animal tissue, and still be infective to humans when the meat is eaten. If the meat has not been cooked sufficiently before human consumption then the cysticercus ruptures in the person's intestine and the small tapeworm head (scolex) evaginates and anchors itself bewteen the gut villi either by suckers (as in T. saginata) or hooks, or sometimes both. The neck region just behind the scolex is a growth area containing germinative cells, which continually give rise to the segments (or proglottids) that are characteristic of adult tapeworms. The body (or strobila) of T. saginata reaches lengths of 3-5 m, while T. solium grows a little longer: up to 7 m in length. Each proglottid contains both male and female reproductive organs (Figure 8.16b) that mature as the proglottids progressively move away from the neck (pushed along by the formation of new proglottids). After fertilisation the proglottids become egg sacs containing hundreds of eggs. Eventually the fertilised proglottids drop off the end of the tapeworms and are passed out with the host faeces. The eggs are a resistant stage and can survive several months in the external environment while adults can live inside their host for up to 20 years. Tapeworms can be killed by treatment with drugs like praziquantel (PZQ, see Section 3.1.3), niclosamide or albendazole. Niclosamide uncouples oxidative phosphorylation, poisoning the tapeworm's metabolism. However, it does not affect the eggs, which are released in a viable form as the parent proglottids disintegrate. Albendazole is effective against a wide variety of adult, juvenile and egg stage helminth parasites and works by destabilising cellular microtubules, weakening the tegument and impairing glucose uptake. 4.1 More harmful tapeworm infections Taenia solium infestations in humans usually produce mild, non-lethal symptoms. However, humans can also act as intermediate hosts when they ingest eggs of the tapeworm. In these cases, hexacanths hatching from the eggs encyst in various tissues other than muscle and can grow from pea-sized objects up to 200 mm long. Many people harbouring cysticerci have them in the central nervous system, which causes a disease known as neurocysticercosis. In the brain, the presence of the cyst causes epilepsy, a common occurrence in many LMICs. A more devastating infection, though, is caused by the larval stage of Echinococcus granulosus. This tiny tapeworm is 3-7 mm long and consists of only three proglottids: an immature one, a mature one and a gravid one full of fertilised eggs. Carnivorous animals, commonly dogs, foxes, wolves and coyotes, are the definitive hosts and harbour the adult tapeworm, which attaches to the intestinal wall using both hooks and suckers. Eggs pass out in the original host's faeces and are typically picked up by intermediate hosts (sheep, cattle, goats, camels, horses, pigs) The hexacnths then hatch in the small intestine and after boring through the intestinal wall they either remain in the body cavity or get carried to the liver where they are filtered out of the blood and lodge there. In these locations a cyst develops and grows by 10-20 mm per year. The inner lining of the cyst is a germinative layer, which buds off hundreds of small invaginated tapeworm heads and even daughter cysts, which then bud off many more heads. The whole structure is called a hydatid cyst (see Figure 8.3c). The definitive host becomes infected by ingesting offal or viscera containing cysts. If a human is unfortunate enough to become infected from dog faeces these cysts can occur in the same organs as in the natural intermediate host. They tend to grow larger in the human host, sometimes reaching 300 mm in diameter, and can contain many litres of fluid. The fluid in the cyst houses parasite antigens, so if it ruptures, the host is suddenly exposed to huge quantities of these antigens, to which it is strongly hypersensitive. In most cases, the host suffers anaphylactic shock and death (a process that is explained in Block 2). Although the Echinococcus granulosus parasite is found all over the world it is particularly common in areas where there is a close relationship between dogs, sheep or goats and humans.

Amoebiasis

Diarrhoeal disease caused by several types of amoeboid protists.

Transmissible spongiform encephalopathies (TSEs)

Diseases that produce characteristic lesions in the brain, giving the tissue a sponge-like appearance upon microscopical examination.

Entamoeba histolytica amoebiasis

Entamoeba histolytica infections are endemic in warm climates with inadequate sanitation, and affect around 50 million people each year, resulting in 40 000-100 000 deaths. Life cycle Infections arise when the protist's cysts are ingested in faecally contaminated water or food. Once the cysts reach the far end of the small intestine excystment takes place, producing metacysts that subsequently divide to give eight trophozoites or amoebae (10-60 μm, see Figure 6.8). These amoebae then travel to the large intestine where they multiply by binary fission. The amoebae may then simply colonise the intestine, and persist as harmless commensals, producing no symptoms in the carrier. Alternatively, they may invade the colonic mucosa, causing inflammation and giving rise to characteristic ulcers (Figure 6.9). This inflammation, known as dysentery, produces profuse bloody diarrhoea that may contain pus and mucus, and be accompanied by abdominal cramps. Systemic spread of the amoebae can occur, sometimes without any obvious signs of dysentery, resulting in abscesses in the liver, lungs and brain. At some stage during an infection, amoebae in the large intestine encyst and the nucleus divides twice (Figure 6.8a, route B). The highly resistant spherical cysts, 10-20 µm in diameter, each containing four nuclei are then shed in the faeces to infect other hosts. Asymptomatic carriers, in particular, are important sources of infection within their communities. When studied in vitro, E. histolytica produces an impressive range of virulence factors, including proteases, pore-forming proteins and collagenases; however, its pathogenicity in vivo is poorly understood. Like Giardia, Entamoeba histolytica has no mitochondria but does possess a mitosome which is again thought to be an adaptation to a mostly anaerobic way of life. Diagnosis is by identification of cysts in a stool sample. Treatment for E. histolytica infection depends on how the disease develops. Metronidazole and tinidazole are used for amoebic dysentery and patients with liver abscesses. Asymptomatic carriers can be treated with iodoquinol or paromomycin. The parasite can also be removed from the water supply by increasing the concentration of chlorine used for disinfection, or by using iodine instead.

Defining periods of an infection

Following its entry into a host (i.e. infection), the pathogen finds its way to its preferred home within the body. There are three important variables related to this phase of the pathogen's life cycle: the latent period, the incubation period and the infectious period. The latent period This is the time period from infection until the host begins to release the pathogen's progeny, i.e. until the host becomes infectious. • What is the distinction between latent period, as described here, and latent infection? (Latent infections were described in Block 1 Unit 1.) • A latent infection is one in which a pathogen may be hidden in the host without causing disease symptoms. We refer to this situation as the 'dormant period' below, to avoid confusion with the latent period as defined here. The incubation period This is the time period from infection until the host begins to show symptoms. The latent period is usually shorter than the incubation period; in other words, the host can pass on the disease before it has developed symptoms (Figure 3.4). The infectious period Another variable that is important to know for preventing the spread of a disease is its infectious period (Figure 3.4). This is the time for which, following infection and the latent period, a host can pass on the disease. For measles-infected individuals, it is six to seven days; for HIV/AIDS-infected individuals, it is several years. Other ways of quantifying life cycles There are a number of other ways in which the life cycles of pathogens can be measured and differentiated. Generation time Every organism, whether it is a host or a pathogen, has a characteristic generation time. This is defined, for a sexually reproducing species such as humans, as the average span of time between the birth of parents and the birth of their offspring. For humans, this is around 20 years; for our pathogens, it can be much shorter. For example, bacteria can vary in fission time from 12 minutes, through to 30 minutes for Escherichia coli in the gut, to days for the leprosy-causing bacterium (Mycobacterium leprae). In other pathogens, such as the protists or helminths, the life cycle is indirect, i.e. it occurs in more than one host or host and vector. In these cases, the generation time has to be measured across all the hosts from one stage to the completion of the same stage in the next generation. Infectious dose Whether or not an infection leads to illness may depend on the number of pathogens that enter the body at the time of infection. For hepatitis B, one virion can cause illness; for other diseases, a large number may be required. For example: • E. coli O157 requires about 10-500 individual bacteria • Salmonella requires anywhere between 100 and 107 (depending on the strain) • the typical number of Vibrio bacteria per host required to cause cholera is about 108, i.e. 100 million • for Giardia, 10 cysts are required for infection. In discussing this issue, it is important to distinguish between the actual number required for infection and the higher probability of infection due to high numbers of pathogens. This variable is called the infectious dose (sometimes expressed as 'ID50'), which is the number of pathogens required to make 50% of infected hosts ill. ID50 values are determined experimentally, by injecting different doses of pathogen into susceptible animals. Knowing the infectious dose of an organism is important in treatment and control, e.g. determining the level of hygiene that is required to prevent its spread. For a disease like norovirus (winter vomiting virus), with a low infectious dose of 10 to 100 virions, very high standards of disinfection are required to help prevent its spread. Table 3.1 presents data on the range of variation in these variables in pathogen life cycles and host response. Table 3.1 Variables in the life cycle and host response of some key pathogens in humans Variable Influenza Malaria Cholera Syphilis TB AIDS Pathogen orthomyxo-viruses Plasmodium sp. (protist) Vibrio cholerae (bacterium) Treponema pallidum (bacterium) Mycobacterium (bacterium) HIV (virus) Latent period 1+ days 21 days 1+ days up to 4 months - 1-2 months Infectious period 3-4 days 1 to 3 years less than 2 weeks up to 4 years months or years* - Incubation period 1+ days 12-30 days 1-3 days 9-90 days (average 3 weeks) 4-12 weeks up to 10 years Duration of symptoms 7-10 days usually maximum of several years from dormant pathogens in liver cells about 48 hours months or years months or years* several years Infectious dose (no. of organisms) 800 10 108 60 not known uncertain * Depends on reactivation. • Briefly describe the range of variation in the life cycle parameters in Table 3.1. • In all cases there is enormous variation, especially in the infectious period and duration of symptoms, where the range is from a few days to years - this probably represents up to three orders of magnitude variation (less than 10 to more than 1000 days).

Protists

Protists are eukaryotic organisms that are either single-celled or have only simple multicellular structures. The Protista include organisms of diverse origins, such as those that make up most of the oceans' plankton, the amoebae and the algae (including seaweeds). This kingdom originally gave rise to plants, animals and fungi, but today's protists probably bear little resemblance to these first eukaryotes, since they have had millions of years in which to evolve. The protists now comprise those eukaryotes that cannot be classified as plants, animals or fungi, and not surprisingly therefore this kingdom exhibits extraordinary diversity, having 30 or so phyla (major groups). In comparison, the plant kingdom has only 12 phyla, and although the number of animal phyla (29) is about the same as the number of protist phyla, there is more genetic diversity in the single protist phylum Ciliophora than is found in the whole of the animal kingdom! The extraordinary variety of structures found in the Protista ranges from microscopic organisms to sea kelps up to 60 m long. Water is an absolute requirement for survival and growth of protists, and many are found in freshwater or saltwater. Others inhabit the soil, and some live as symbionts in plants, animals and fungi. Protists vary widely in their modes of nutrition, with some able to photosynthesise like green plants, while others are heterotrophs and feed on other organisms or dead material. The reproductive strategies of the protists are also very varied and may involve alternating sexual and asexual generations. If an organism has both a sexual and an asexual life cycle, it is often the environmental conditions that determine which type of reproduction predominates. If conditions are favourable for growth and replication, the asexual cycle frequently predominates, whereas in less favourable conditions, the sexual cycle may be more common. Finally, there are some protists that have not been observed to reproduce sexually at all. Many parasitic protists form special, thick-walled, highly resistant structures called cysts, in a process called encystment. Such cysts serve a number of purposes, such as: • protecting the organism from adverse environmental conditions • providing shelter for reproductive processes • allowing transmission of the parasite from one host to another. Some protists can survive as cysts for a very long time, so transmission of the parasite to a new host can occur long after it has left the previous one. At some point, excystment occurs. This is when the cyst releases the vegetative (i.e. actively dividing) form of the parasite, known as a trophozoite. Ingesting cysts is a common way in which humans become infected with protist parasites, and excystment frequently takes place in the intestine. Other parasitic protists use a vector to provide protection and to allow transmission from one host to another. Their survival is dependent on the survival of the vector, which can vary from a few weeks (sandflies) to several months (mosquitoes and tsetse flies). Despite the variety of forms and modes of life, there is a structure that is found in nearly all protists for at least one stage of their life cycle, and that is one or more flagella. The flagella are used in feeding and locomotion, and are distinct from the flagella found on bacteria, which have a different structure and function by a different mechanism. Protists that possess flagella during the 'adult' stage of their life cycles are colloquially known as flagellates. These animal-like cells have one or more flagella and include Giardia spp., which cause the diarrhoeal disease giardiasis, and the trypanosomes, which cause sleeping sickness. However, even though these groups both have flagella, they are not related to one another. Single-celled protists that have animal-like cells and are unable to photosynthesise are sometimes referred to as protozoa. However, this is not a discrete classification, since (like the flagellates) its members have diverse evolutionary origins. Protozoan cells have one or more nuclei and are usually classified according to their cell structures: for example, members of the phylum Apicomplexa all have an arrangement of organelles known as the apical complex. It is largely the animal-like protists, found in both free-living and parasitic forms, that are human pathogens. Although few in number, they cause some of our most serious diseases.

Disease in Ethiopia

Slide 1 (00:00) This slide cast will introduce you to the realities of infectious disease in rural Ethiopia and how it is being tackled in every village by local health workers with basic training in disease prevention, diagnosis, treatment and control. My aim in giving you an insight into this astonishingly beautiful and hospitable country is to illustrate infectious disease and public health at local level and the impact it can make on a national scale. But I also hope it will challenge the negative image of Ethiopia often presented in the media. Slide 2 (00:40) Ethiopia is in the horn of East Africa. Its largely mountainous terrain can be clearly seen in the aerial view on the right. Slide 3 (00:54) Notice the huge expanse of Lake Tana in the north and the chain of lakes that mark the start of the Great Rift Valley, beginning in central Ethiopia - not far from the capital Addis Ababa - and extending south into Kenya. Slide 4 (01:08) This slide cast focuses on infectious disease in rural Ethiopia. But before we go there I want to make the point that Ethiopia has several large and expanding cities in addition to the capital Addis Ababa. There are grand buildings, modern hotels, busy traffic and many Orthodox Christian Churches, mosques and other places of worship in this intensely religious country. The shanty settlements in the foreground of this photo are being rapidly replaced by apartment blocks. But Ethiopia remains a largely rural country - only about 20% of the population live in urban conurbations. Slide 5 (01:52) Most of Ethiopia's 83 million people live by farming in distributed rural communities, called kebeles in the Amharic language. The average kebele has about 1000 households and roughly 5000 inhabitants, led by a council of elected representatives. The round thatched houses in this photo are called tukul and are found all over Ethiopia. Slide 6 (02:22) As you can see, the Ethiopian countryside can be green and lush in the rainy season, supporting some of the country's major exports: coffee, cut flowers and leather goods from animal hides. But the soil is poor after decades of subsistence farming and it soon dries out when the long rains end. Slide 7 (02:45) Animals are vital to the rural economy, but close proximity to their domestic animals exposes the rural population to infectious diseases with reservoirs in animal hosts, particularly intestinal parasites and the bacteria causing diarrhoeal diseases. Slide 8 (03:05) Most farming is done by hand or with simple ploughs pulled by oxen. Here the staple cereal crop, called tef is being harvested.The tiny grains are ground to make Ethiopia's national dish - a thin pancake called injeera, which is unique to the country. I'll come back to injeera when I talk about malnutrition in Ethiopia in a moment. First, let's look more closely at a particular kebele and the infectious diseases that affect its inhabitants. Slide 9 (03:40) This is a map of Fura kebele, a rural area with a population of approximately 5000 people in the Southern Nations, Nationalities and Peoples Region of Ethiopia. The households are distributed across a wooded area, roughly five by ten kilometres, but some kebeles are larger than this. The inhabitants of Fura have carved out small fields in which to grow tef and graze their animals. Slide 10 (04:11) There are no paved roads, no piped water, and no electricity supply to the houses, so they are dark inside and difficult to photograph. Grass partitions divide the living and sleeping areas and the furniture is simple and stands on the mud floor. Slide 11 (04:29) This road in Fura is typical of how people get around in rural kebeles - mostly on foot, walking long distances, sometimes on a bicycle or in a donkey cart. The deep channels in the earth are caused by flood water during the rainy season, when the roads are often impassable, even with a four-wheeled-drive vehicle. The nearest health centre is 15 kilometres down this road, so if people need urgent medical help they may have to be carried there on a stretcher. Slide 12 (05:03) Many of the causes of infection in Ethiopia are related to the lack of clean water, which is a constant problem for many rural communities. Only 38% of the population has access to improved drinking water and many of these are in towns and cities - in the countryside, most people get their water from lakes, rivers and streams. Slide 12 (05:28) Some rural communities have deep communal wells, like this one in Fura. The water table is a very long way down in the dry season. The bucket is made from an old rubber tyre. Slide 13 (05:43) Unsafe drinking water presents a major health hazard to a large proportion of Ethiopia's population. Here a horse, man and boy drink from the same muddy stream flowing from pasture where cattle were grazing. Shallow streams and pools like these make ideal breeding grounds for malaria mosquitoes and other vectors of infectious diseases such as onchocerciasis and schistosomiasis, as well as harbouring many pathogens causing diarrhoeal diseases. Slide 14 (06:17) People and their animals competing for scarce drinking water also increases the risk of water contamination by human and animal urine and faeces. Slide 15 (06:29) This chart shows the main causes of death in Ethiopia in children aged under five years. The high rate of diarrhoeal disease is one indication of the lack of access to clean water. As you can see from the four bars on the left, diarrhoeal disease, pneumonia, neonatal sepsis (which is mostly due to tetatnus) and HIV/AIDS account for the majority of these avoidable deaths. The 21% of deaths labelled 'all other causes' include many important infections such as measles, malaria and tuberculosis. Slide 16 (07:11) An idea of the challenges in rural communities can be gauged from this wall chart from a Health Centre in the small town of Modjo. It shows the top ten causes of serious morbidity (illness) among male children aged under five years who were sent from the surrounding rural kebeles because their condition could not be managed locally. The chart for girls of this age is very similar, but my photo wasn't as clear. The prevalence of infection among the under fives is striking: the top ten causes of serious illness are non-bloody diarrhoea, pneumonia, acute urinary tract infections, dysentery (that's bloody diarrhoea), infestation with helminths, infections of the skin and subcutaneous tissue, all other respiratory diseases, malaria, all other skin diseases, and typhoid fever. Slide 17 (08:18) Inadequate food hygiene standards and the practice of eating kitfo - raw beef or ox meat - at important festivals is another major source of infection. Beef tapeworm is prevalent and infestation with other intestinal worms is widespread, affecting over 50% of children, with major effects on their growth and development. Slide 18 (08:40) Ethiopia is commonly associated with famine, but such deep crises of absolute food insecurity are relatively rare. The more consistent problem is generalised protein energy malnutrition. The staple diet is injeera made from tef, which you saw being harvested earlier. Injeera is typically eaten from a communal plate with a hot lentil sauce and perhaps some stewed vegetables. Meat is only for special occasions in most families. Injeera is highly nutritious, but there often isn't enough food to go around. Children in particular suffer from a persistent shortage of calories in their diet and a lack of vitamin A and iron because of the scarcity of vegetables. Slide 19 (09:32) These data come from the last national survey in Ethiopia conducted in 2005 and published in 2007. At that time, 38% of children under 5 years were underweight and 47% were stunted (short for their age). The extent of iodine and iron deficiency is demonstrated by the high rates of goitre and also of anaemia among young children. These conditions also affect many women of childbearing age. Malnutrition is a major contributor to susceptibility to infectious disease and premature death, and not only among children - over a quarter of women are chronically malnourished, which makes them particularly vulnerable during pregnancy, childbirth and breastfeeding. Slide 20 (10:30) Despite its many health problems, some of Ethiopia's indicators of public health are better than the average for Sub-Saharan Africa, and steady progress is being made towards achieving the Millennium Development Goals for reducing child and maternal deaths. But there is a long way to go. For example, for every 1000 live births in Ethiopia, more than 4 women die from causes related to pregnancy, childbirth, postnatal infection or haemorrhage. This is roughly 50 times higher than the maternal mortality rate in the UK. Slide 21 (11:11) Faced with all these challenges to public health, in 2005 the government of Ethiopia began an ambitious programme to bring basic disease prevention and health promotion services to the entire population, now totalling over 83 million people. They have built a small Health Post in every kebele staffed by two full-time Health Extension Workers trained and paid by the Ministry of Health. There are now over 12 500 rural Health Posts like these and over 33 000 Health Extension Workers deployed in rural areas. Slide 22 (11:55) Here are two of them - Asafesh on the left and Almaz on the right. The Health Extension Workers are all young women - an innovation in Ethiopia where the majority of nurses and midwives, as well as doctors, are male. Asafesh and Almaz have graduated from the one-year residential training programme and now work in rural Health Posts in different regions of Ethiopia. They each work with another Health Extension Practitioner to take care of the 5000 inhabitants of their communities. Slide 22 (12:33) The vital importance of the Health Posts and their Health Extension Workers to the delivery of health services in Ethiopia is well illustrated by the pie chart on the left. Each Health Post is in a ring of five 'satellite' Health Posts, at a distance of 10-15 kilometres from the nearest Health Centre, which supplies and supervises the Health Extension Workers. The provision of health services in Ethiopia relies very heavily on this organisation of primary care, partly because of the remoteness of the rural population from the hospitals, which are all in the cities, but also because of the shortage of medical expertise: there are just 20 hospital beds and 2 doctors for every 100 000 population. Slide 23 (13:26) So what health services do Asafesh and Almaz provide for their communities? This Health Post wall chart lists the main components. Take a moment to read the list. Click on the 'Pause' button while you do this. How many of the service areas are directly or indirectly concerned with preventing or controlling infectious disease? Immunisation and all of the environmental health and disease prevention and control packages are obvious examples, but infection control is also an important aspect of health care during pregnancy, labour and delivery and the postnatal period. It also figures prominently in many of the health education activities that Health Extension Workers conduct, for example, on handwashing, food hygiene and digging latrines. Slide 24 (14:34) Here is some of the basic equipment in Asafesh's Health Post. It doesn't have running water or a sink, so Asafesh carries water in a bucket from a nearby well and washes her hands with soap between patients, using a bowl in a corner of the room. She has a pressure cooker for sterilising instruments, mainly used when she is attending births; there are syringes for giving immunisations and injectable contraception, and a stethoscope and blood pressure cuff. Slide 25 (15:06) The scales are for weighing adults (mainly pregnant women attending for antenatal care) and there is a delivery couch in a side room. However, around 90% of rural women choose to give birth in their own homes with a Health Extension Worker present or a traditional birth attendant. Postnatal sepsis and haemorrhage are the main reasons for the high mortality rate: remember it was 4.4 maternal deaths per 1000 live births in the bar chart you saw earlier. Slide 26 (15:41) Asafesh also a spirit burner, boxes of disposable syringes, packets of oral rehydration salts to treat diarrhoea, and sachets of PlumpyNut (an energy-rich paste given to malnourished children). The scales made from a large bowl attached to a spring balance are to check children's weight and growth rate. There is also a conventional weighing scale for young babies. Identifying children who are underweight is an important responsibility for Health Extension Workers. Supplementing the diet and giving vitamin A capsules help to protect malnourished children from infection. Slide 27 (16:26) Health Extension Workers have a very small stock of medicines for the 5000 people they serve. At the Health Post in Fura, Asafesh has only paracetamol syrup for pain, and mebendazole and trimethoprim sulphamethoxazole to treat intestinal worms. She has anti-malaria tablets (Coartem) and contraceptive pills and the injectable contraceptive suspension Depo-Provera. There are iron and folic acid tablets for anaemia, tetracycline eye ointment for eye infections in newborns - mainly due to chlamydia transmitted from their mother's birth canal - a first aid kit, cotton wool swabs, antiseptic, a thermometer, a torch and surgical gloves. That's all. Slide 28 (17:20) Every Health Post has a kerosene, gas or electric refrigerator for storing vaccines and an insulated vaccine carrier lined with ice packs. The Health Extension Workers use this to collect their monthly supply of vaccines from the Health Centre 10 to 15 kilometres away. Remember that rutted mud road that Asafesh has to walk along to collect her supplies, unless she can get a lift from someone in a donkey cart. She also uses the vaccine carrier to keep vaccines cold when she provides outreach immunisation sessions in remote parts of the kebele. Slide 29 (17:59) Ethiopia follows the World Health Organization's recommended Expanded Programme on Immunization (or EPI) for low resource countries. A fully immunised infant should receive all the vaccines listed on this slide before its first birthday. BCG vaccine protects them against the most serious forms of tuberculosis. The DPT-HepB-Hib vaccine is also known as pentavalent vaccine because it immunises against the pathogens causing five infectious diseases - diphtheria, pertussis (or whooping cough), tetanus, liver disease caused by hepatitis B viruses, and bacterial pneumonia and meningitis caused by Haemophilus influenzae type b. Ethiopia recently added PCV10 vaccine to the routine immunisation schedule to protect infants against pneumococcal pneumonia. Rotavirus vaccine to prevent the most common cause of diarrhoeal disease becomes routine in 2012. Slide 30 (19:13) Inside the Health Post, wall charts help Asafesh to keep track of her public health targets for the local population, including immunisations. The inside of all Health Posts are covered with monitoring charts like these. Slide 31 (19:28) This wall chart shows how Almaz has been achieving her targets for childhood immunisation coverage in her kebele, which is called Shera Dibandiba. 'Protected at birth' refers to neonatal tetanus protection induced by immunisation of the mother with at least two doses of tetanus toxoid during her childbearing years and one during the pregnancy. Neonatal sepsis is a major cause of newborn deaths; note that less than 80% of newborns in this kebele received this protection. Pentavalent-3 refers to infants receiving all three doses of this combined vaccine to protect them against diphtheria, pertussis, tetanus, hepatitis B virus and Haemophilus influenzae type b bacteria. Measles vaccine is given at 9 months in Ethiopia. Almaz is achieving around 90% coverage with pentavalent-3 and measles vaccines. Slide 32 (20:36) Immunisation clinics are mainly held at the Health Post, but all Health Extension Workers spend four out of five days visiting families in the surrounding area and teaching them about disease prevention and health promotion, for example in food preparation, breastfeeding, hand washing, waste disposal and digging latrines. When Asafesh and her colleague are away from the Health Post they use this cardboard dial to indicate the direction they have gone in that day, so they can be found in an emergency. Slide 33 (21:11) One of Asafesh's most successful campaigns has been to persuade and support the inhabitants of Fura to build a latrine for every household. Until everyone has access to a latrine, open defaecation in the fields remains very common. This is a major transmission source for parasitic worms and for the bacteria and viruses causing diarrhoeal diseases. Fura kebele, where Asafesh works, was the first village in the southern region to be declared 'open defaecation free' because every household had its own latrine. This one is in the yard behind the Health Post. Notice the yellow can of water for washing hands. Slide 34 (21:58) Handwashing is one of the simplest, cheapest and most effective of all public health measures against infectious diseases. Health Extension Workers have promoted the installation of handwashing sites like this one on latrines all over rural Ethiopia. A 30% reduction of diarrhoeal diseases can be achieved by washing hands with clean water after defaecation and before food handling. Using soap reduces diarrhoeal illness by over 40%. Slide 35 (22:36) Another of the main activities of Health Extension Workers in malarious areas is to mobilise the community to keep their water tanks covered and to drain water collections where mosquitoes breed, like those on the left of this slide. Only 33% of Ethiopia's children were sleeping under insecticide-treated nets at night in 2007, but the coverage is rising rapidly, thanks to the distribution of nets to every household by the Health Extension Workers. However, there are still over three million reported cases of malaria in Ethiopia every year. Asafesh and Almaz can diagnose malaria with the rapid test kit shown on the right and treat non-urgent cases themselves with simple drugs. But patients with malaria complications have to be referred to the nearest Health Centre, 10-15 kilometres away, for more specialised treatment. Slide 36 (23:39) This is Godino Jitu Health Centre, where Almaz refers patients that she and her colleague can't manage in the rural community. It serves the surrounding population of 36 700 people and sees up to 300 clients every day. There is no operating theatre and it doesn't have a fully qualified doctor. Slide 37 (24:05) Godino Jitu Health Centre has a staff of about 24 health workers, including 8 Health Officers with a shorter medical training than doctors, 10 nurses and 2 healthcare assistants, 2 laboratory technicians, a pharmacist and a druggist. It also has clerical staff to register patients and keep records. Slide 38 (24:30) Health Centres deliver services that cannot be provided at Health Posts, including HIV-testing and prescription of anti-retroviral drugs and medication to treat tuberculosis. The pharmacy at Godino Jitu is well stocked, mainly with drugs to treat infectious diseases. Slide 39 (24:52) One of the diagnostic tests that the Health Centre can provide is microscopic examination of blood films from patients with suspected malaria. The sink on the left is where the slides are stained for confirmation of a malaria diagnosis, but more important is the identification of the Plasmodium species so that the correct treatment can be given - for example, to children and pregnant women with acute malaria crises. This is beyond the scope of what Health Extension Workers can currently provide. Slide 40 (25:27) However, Health Extension Workers like Almaz and Asafesh are keen to increase their knowledge and skills, so they can offer even more health services to their local communities. The Open University is proud to have been involved in a unique programme to support the upgrading of Ethiopia's rural Health Extension Workers, as Vice Chancellor Martin Bean saw for himself when he visited Almaz at her Health Post in 2011. Slide 41 (25:58) 1000 Health Extension Workers began studying 13 upgrading modules in 2011, which were produced by Ethiopian health experts with the support of The Open University's HEAT (Health Education and Training) team. If you are interested, you can see all the modules on the HEAT website. The focus in the modules is on infectious disease and its prevention, diagnosis, treatment and control, but the curriculum also includes modules on non-communicable diseases and mental health. Slide 42 (26:34) There is an increasing focus in many low and middle-income countries on the rising rates of chronic conditions such as hypertension, diabetes and heart disease. Ethiopia is ahead of the curve in planning to train all its rural Health Extension Workers in these new areas of health concern, which The Open University has helped to develop. Slide 43 (26:58) But Almaz and Asafesh - like the other 33 000 rural Health Extension Workers - will still be focusing most of their attention on improving the life chances of beautiful children like these, through immunisation, nutritional support and education on domestic and environmental hygiene. Their efforts are steadily raising the health indicators in Ethiopia, despite the challenges of an often harsh environment and a widely distributed rural population. Ethiopia's local approach to public health provided on a national scale is leading the way in tackling infectious disease in Africa. Slide 44 (27:42) Amehseghinalehu - thank you!

Filarial worms and bacteria

The filarial nematodes (except Loa loa) carry an endobacterium of Wolbachia species. It appears that there has been a long period of coevolution such that the worms and the bacteria now have a mutualistic relationship, with the bacteria playing an important role in viability and fertility of the nematodes. It has also become apparent that these bacteria are a source of pathology of filarial disease triggering an adverse inflammatory reaction when microfilariae die either due to the immune response of the host or following drug therapy.

Encystment

The process that occurs when a parasitic protists forms a cyst.

Relative risk

The risk posed by a risk factor for one category of individuals divided by the risk for another category. Also sometimes called the 'risk ratio'.

Escape from host immunity

The strategies that pathogens have evolved to evade the host's immune responses can be grouped broadly into: • antigenic disguise or antigenic variation, which prevents or reduces the host's ability to recognise pathogen antigens - thereby evading the recognition required to trigger an immune response • mechanisms that obstruct or reduce the effectiveness of the host's immune response after this has begun.

Infectious period

The time for which, following infection and the latent period, a host can pass on the disease.

Serial interval

The time interval between an individual becoming infected and infecting others.

Toxomplasmosis

Toxoplasmosis is the name given to any infection with the parasite Toxoplasma gondii. This organism is found throughout the world and is unusual among parasites as it can survive and replicate asexually in almost any warm-blooded vertebrate including humans. However, members of the cat family, Felidae, are the only definitive hosts, since sexual reproduction of T. gondii is confined to them. As it is a zoonotic infection for humans, humans are a 'dead end' host. Cats cannot pick up the parasite from humans so in that sense humans are not a true intermediate host; rather, they are an accidental host. Cats infected with T. gondii shed oocysts in their faeces, and these oocysts can then infect intermediate hosts when ingested Sporozoites are released from the oocyst in the intestine of the intermediate host. They pass through the intestinal wall, penetrate a variety of cell types and transform into tachyzoites. These divide rapidly by nuclear division known as endodyogeny. Soon the cell lyses and tachyzoites are released. Some are taken up by macrophages and others disseminate around the body invading almost any cell but with a preference for muscle and neuronal cells. If the parasite is taken up into the macrophage, it prevents the phagocytic vacuole from fusing with lysosomes, and can therefore multiply and persist in the phagocyte. This is the acute phase of infection, which lasts 7-10 days. After this time the immune response takes effect, and tachyzoites differentiate into bradyzoites. These divide only slowly and form tissue cysts predominately in muscle tissue and in the central nervous system (CNS). This is the chronic phase and because the cysts have thicker walls, the parasites are protected from the immune system and can persist for the life of the host. Cats become infected by eating birds or small mammals with cysts in their tissues, or by ingesting oocytes shed in cat faeces. T. gondii infections are common in humans (up to 30% of the global population carries a Toxoplasma infection) but are mostly asymptomatic, and if symptoms do occur they may be limited to a mild flu-like illness. Acute toxoplasmosis is more unpleasant and can involve painful swelling of the lymph glands and necrosis (tissue death) of the lung, heart and liver. Humans may become infected by ingestion of oocysts from faecally contaminated hands or food, or more commonly by eating undercooked meat containing tissue cysts from infected intermediate hosts such as pigs or lambs. If tissue cysts are ingested the bradyzoites are released in the intestine as the muscle is digested, penetrate the intestinal wall, transform into tachyzoites and travel around the body as described above, ending up as cysts in the tissues. For most people this is the end of the story, with infection resulting in lifelong immunity. However, if the immune system is compromised in any way, for example by drugs or infection, then the parasites begin rapid multiplication again which can be fatal. This explains why toxoplasmosis has become an important cause of death in immunosuppressed individuals, such as patients with AIDS, in whom it causes encephalitis. Just as important is the devastating effect the parasite has on the human fetus. The tachyzoite can, and does, cross the placenta during pregnancy leading to infection of the fetus. This can cause congenital abnormalities of the eyes or brain, or even stillbirth. Pregnant women in the UK are made aware of the dangers of toxoplasmosis, and are encouraged to take the appropriate preventative measures. These include washing the hands after time in the garden, cooking all food thoroughly and avoiding exposure to cat faeces in the home or garden. Interestingly, Toxoplasma is known to affect behaviour of the intermediate host. This is presumably because of its presence in cells of the CNS, especially in the brain. Mice and rats, for instance, fail to show fear of their predator, the cat, such that successful transmission of the parasite to the cat is more likely. It has also been shown that reaction times of infected people are impaired and a possible link to schizophrenia is being investigated. The parasite secretes an enzyme that affects dopamine production (a neurotransmitter in the brain), so it is therefore conceivable that the parasite can also affect human behaviour. Diagnosis is by serological testing for the presence of patient antibodies against the Toxoplasma pathogen, or by PCR detection of parasite DNA. Toxoplasmosis may be treated with a combination of pyrimethamine and sulfadiazine. In pregnancy, spiramycin may be used and clindamycin has been used to treat infections involving the brain.

Brugia malayi

Vector is principally Masonia. Hosts are humans, monkeys, cats and dogs. Found in regional lymph vessels. Causes lymphatic or Malayan fliariasis. Found in China, Korea, Indonesia, Malaysia, Philippines, Sri Lanka

Variation and reproduction

When a pathogen infects a human population, it is immediately apparent that there is a great deal of variation among individual hosts. Some do not become ill at all; some are only mildly affected; some are very ill indeed; and others may die. The causes of this variation are many and varied. Some individuals are just lucky and do not get infected; some receive a larger dose of pathogen than others; some are more susceptible than others to the pathogen. Variation in susceptibility may reflect variation in exposure to the disease earlier in life; it may be due to overload of an individual's immune system as caused by a number of superimposed infections compounded by poor diet; or it may have a genetic basis. The primary focus in this section is genetic variation, both among hosts and among pathogens. It is the interplay between host and pathogen genotypes that provides the basis of their coevolution. Genetic variation is the fundamental basis for evolution by natural selection. If there is no genetic variation in a character, it cannot evolve. So you will explore the nature and sources of genetic variation in hosts and pathogens next. A major source of variation is reproduction and so you will also consider the nature of reproduction in the two partners in their coevolved relationship. The development of new techniques for analysing genomes has made it possible to quantify the genetic basis of variable susceptibility to infectious diseases. For example, 12 genes have been identified that contribute to susceptibility or resistance to malaria in humans. For large, complex organisms such as humans, the major source of genetic variation is sexual reproduction. Many organisms reproduce asexually, either in response to particular conditions or continuously, and the significance of this in relation to infectious disease will be considered later. During sexual reproduction, meiosis and syngamy (fusion of gametes from different parents) involve genetic reassortment and recombination, as a result of which offspring have unique genotypes. Other sources of genetic variation are mutation and, at the level of populations, genetic drift. Genetic mutations Genetic mutations occur spontaneously, but mutation rates are accelerated by a variety of environmental factors such as chemical pollutants and increased radiation, e.g. radioactivity and UV-B. Few mutations are expressed phenotypically; many are eliminated by DNA repair mechanisms; others are eliminated by natural selection during development; while others are recessive and not expressed in diploid or polyploid organisms. Nevertheless, in organisms with large genomes, such as humans, every individual carries a small number of novel mutations. Genetic drift Genetic drift is a factor in small, relatively isolated populations of a species. Such populations contain a random sub-set of all the alleles in the genome of that species. Genetic drift is an important factor in many aspects of host-pathogen coevolution. For example, small, isolated populations of humans and other vertebrate animals are often more susceptible to a particular disease because, by chance, they lack alleles protecting that species against that disease. It may also be important, as you will see later, in the evolution of pathogens. Genetic variation in host populations Contemporary host populations show enormous genetic variation that is the result not only of mutation, drift and reproduction, but also of natural selection. The genome of a species is the result of generations of natural selection and reflects its evolutionary history. This is very apparent in the context of infectious disease. In humans, a genetic basis for variation in susceptibility to infectious diseases has been revealed by a variety of techniques. The frequency of some susceptibility alleles shows a very strong association with disease. For example, alleles associated with immunity to malaria are much more common in Africa, which has long been exposed to this disease, than in other parts of the world that have no history of exposure to it. In contrast, alleles that influence the infectivity of HIV, a very recent disease in an evolutionary sense, are expected to be much less common in the human population. The allele called CCR5∆32 provides low or high protection against AIDS, depending on whether it is in the heterozygous or the homozygous state. These alleles are predicted to increase in frequency, because they prolong survival during peak reproductive years. In fact, this allele frequency is about 10% in Caucasian people, which is more common than most alleles that affect malaria susceptibility. It is not known why CCR5∆32 should occur at such a high frequency. The mutation itself does not seem to be deleterious to the immune system, since other linked genes can carry out the functions of this particular chemokine receptor. It has been speculated that: • either it is a duplicate gene that has arisen by chance and until the present it has been evolutionarily neutral and has coincidentally been common in the gene pool • or it is linked to a gene that has a definite evolutionary advantage - possibly another chemokine receptor. 1.2.2 The MHC and genetic polymorphism The most highly variable part of the genome of humans and other mammals is the major histocompatibility complex (MHC, see Block 2 Unit 1, Section 2.1.1), of which HLA-B35 is one variant. The MHC consists of a small number of linked genetic loci that code for human leucocyte antigens (HLAs). Six loci are commonly recognised, called HLA-A, -B, -C, DP, DQ and DR. These loci are highly polymorphic and an ever-increasing number of alleles are being identified. The number of alleles known by late 2002 was 206 at HLA-A, 403 at HLA-B, 92 at HLA-C and 400+ at HLA-D. Such levels of genetic polymorphism are far greater than anything known for other parts of the human genome. At most loci in the human genome, the number of alleles is in single figures. Different human populations have different MHC allele frequencies, presumably reflecting the history of diseases to which that population has been exposed. The figure is a bar chart showing the frequencies of four MHC alleles at the HLA-B locus in three different human populations: European Caucasian, African Black and Japanese. The horizontal axis is labelled MHC allele and is marked HLA-B5, HLA-B17, HLA-B18 and HLA-B37. The vertical axis is labelled allele frequency and is marked from zero to 25 per cent at intervals of 5 per cent. Japanese populations have a very high frequency (21 per cent) of the HLA-B5 allele, while frequencies of HLA-B17 and HLA-B37 are below one per cent and HLA-B18 is not present at all. African Black populations have a high frequency (about 16 per cent) of the HLA-B17 allele, while frequencies of HLA-B5 and HLA-B18 are below 3 per cent and HLA-B37 is not present at all. Finally, European Caucasians have moderate frequencies (about 6 per cent) of HLA-B5, HLA-B17 and HLA-B18, but less than a 2 per cent frequency of HLA-B37. The MHC was originally discovered in the context of graft rejection. Tissue grafts are much more likely to be successful when donor and recipient have very similar MHC genotypes, and this is more likely when they are close relatives. Genetic variation in the MHC is interpreted as an adaptation against pathogens; over a number of progeny, it allows recognition by the immune system of a wide range of pathogens. A consequence of the very high levels of variation that exist at MHC loci is that there is a great deal of variation in the effectiveness with which individuals respond to specific pathogens. Consequences of reduced genetic variation If genetic variation in the MHC is an adaptation against pathogens, we would expect to find hosts to be more susceptible to pathogens if, for some reason, that genetic variation is reduced. • Identify a process that tends to reduce genetic variation. Answer Inbreeding: frequent matings among close relatives lead to a reduction in heterozygosity, i.e. reduced genetic variation. This hypothesis has been tested among free-ranging Soay sheep living on the Scottish island of St Kilda. Because the size of the sheep population is restricted by the small size of the island, the level of inbreeding is quite high. The study revealed that, in winters when particularly severe weather caused high mortality, the sheep that died had higher levels of intestinal nematode pathogens and were also more inbred than those that survived. What is unclear is the extent to which the homozygosity or the increased parasite burden (possibly caused by the former) contributed to the lower survival. Immune defence loci on host genomes The very high levels of variation that exist at loci associated with immune defence are paradoxical. Infectious disease exerts strong selection on host populations and the usual outcome of strong selection is very low levels of genetic variation. However, in this case you should not consider 'infectious disease' as a single entity. Each infection is different and requires a different means of recognition (antigen presentation) and response. Consequently, different pathogens select for different genes and therefore promote diversity. The severity of specific disease outbreaks also varies from generation to generation. As a result, strong selection by one pathogen is not sustained for long enough to drive alleles related to that disease to fixation. The human genome is in a constant state of flux The 'selective environment' presented by disease is constantly shifting and much of the genetic variation observed among hosts reflects the exposure of past generations to infections. There are a number of other factors that may promote high genetic variation in hosts, including those listed below. • Pathogens selectively infect commoner genotypes so that rare genotypes escape infection and thus tend to increase in frequency. For example, rare genotypes of the New Zealand snail Potamopyrgus antipodarum largely escape infection by the trematode Microphallus. • A number of species, including humans, have been found to mate preferentially with partners who are different from them at MHC loci (a mating pattern called disassortative mating). This increases heterozygosity at MHC loci. Experimental studies using mice provide evidence that MHC heterozygosity is adaptive. For example, when mice were exposed to multiple strains of Salmonella and a single strain of Listeria in large population enclosures. MHC heterozygous mice had greater survival and higher body weight than homozygous mice. • Selection may act against homozygous individuals, as in the case of the Soay sheep described above. • Migration of individuals from one part of a host's range to another introduces novel alleles into local populations. People are particularly mobile hosts and levels of genetic variation in human populations due to migration are thought to be very high. As a result of all these factors, the human genome is far from static over time; indeed, in terms of allele frequencies in local populations, it is in a constant state of flux. Genomes change. Different versions of genes rise and fall in popularity driven by the rise and fall of diseases. ... The genome that we decipher in this generation is but a snapshot of an ever-changing document. There is no definitive edition. (Ridley, 1999) 1.2.3 The role of sexual reproduction Because humans reproduce sexually, it is usually assumed that sexual reproduction is the 'normal' way to reproduce. It is, however, only one of many reproductive mechanisms found among animals and plants and its evolution is a matter of continuing debate among evolutionary biologists. For example, consider the following quotation. Sex must be important, simply because reproducing in this eccentric way is so expensive. By becoming involved with a male, a female dilutes her genes with those of someone else who does rather little to ensure that they survive. Even worse, she produces sons who go in for the same selfish behaviour. To balance this enormous cost, sex must have some hefty advantages for genes if not for their products - and it does; for a sexual world has conquered death. (Jones, 1996, p. 272) Jones is referring to the capacity of sexual reproduction to act as a filter of germ cells that eliminates the majority of mutations before they can be passed on to progeny. But this is only one reason why sexual reproduction is advantageous. It is widely accepted that another important advantage is that the genetic recombination that results from this sort of reproduction enables a species to counteract pathogens that cause infectious disease. Evidence that pathogens may play a role in determining the reproductive mode of hosts comes from studies of the freshwater snail P. antipodarum in New Zealand lakes. Individual snails are either male or female, unlike many snails that are hermaphrodites, but females are capable of a form of asexual reproduction called parthenogenesis (meaning 'virgin birth'). Some populations consist entirely of females and so must reproduce asexually; others contain as many as 40% males, and so have the potential for sexual reproduction. A comparison of 66 snail populations revealed a strong tendency for males to be more frequent in locations that were heavily infected by parasitic trematode flatworms (Figure 5.2). This suggests that sexual reproduction is favoured in snail populations where pathogens are abundant. The figure is a bar chart showing the frequencies of four MHC alleles at the HLA-B locus in three different human populations: European Caucasian, African Black and Japanese. The horizontal axis is labelled MHC allele and is marked HLA-B5, HLA-B17, HLA-B18 and HLA-B37. The vertical axis is labelled allele frequency and is marked from zero to 25 per cent at intervals of 5 per cent. Japanese populations have a very high frequency (21 per cent) of the HLA-B5 allele, while frequencies of HLA-B17 and HLA-B37 are below one per cent and HLA-B18 is not present at all. African Black populations have a high frequency (about 16 per cent) of the HLA-B17 allele, while frequencies of HLA-B5 and HLA-B18 are below 3 per cent and HLA-B37 is not present at all. Finally, European Caucasians have moderate frequencies (about 6 per cent) of HLA-B5, HLA-B17 and HLA-B18, but less than a 2 per cent frequency of HLA-B37. Sexual reproduction is a complex process that confers many diverse benefits on individuals, but which also incurs many diverse costs. Its role in generating genetic variation as a defence against disease is only one benefit. An important cost is that it provides a very reliable and efficient means of transmission for pathogens, a topic that will be discussed in Section 4. An interesting benefit, relevant to infectious disease, is the opportunity it provides for individuals to mate preferentially with apparently disease-free partners. 1.2.4 Variation and reproduction in pathogens Like hosts, pathogens show genetic variation that results from mutation and drift, but they vary in terms of the amount of variation resulting from reproduction. Typically, viruses and bacteria reproduce asexually, but protists and invertebrate pathogens possess various forms of sexual reproduction that increase genetic variation, as it does for hosts. For example, many tapeworms, which have a very small chance of encountering a mate, are self-fertilising hermaphrodites. A survey of pathogens found that sexual reproduction is more common in parasitic species than it is in closely related non-parasitic species and models of coevolution suggest that, under certain conditions, sexual reproduction is adaptive for pathogens. Genetic recombination in bacteria Although most bacteria do not engage in sexual reproduction, there is increasing evidence that they achieve higher levels of genetic recombination than has generally been assumed. A review of reproductive mechanisms among pathogenic prokaryotes suggests that most can reproduce sexually, thereby rapidly generating genetic variation, and clonally (i.e. asexually), thereby stabilising successful genotypes. The mechanisms that cause a prokaryote pathogen to switch from one reproductive mode to another are not known. Horizontal gene transfer in bacteria and viruses A source of genetic variation that sets viruses and bacteria apart from other pathogens is the horizontal transfer of genes from a variety of sources. Viruses can acquire genes from other viral strains and from host cells. Bacteria can incorporate into their genomes a variety of genetic elements, including plasmids, phages and transposons, which may be derived from their environment or from bacteria of other strains or 'species'. The details of these mechanisms are beyond the scope of this unit; the significant point is that viruses and bacteria have many ways of acquiring genes from a variety of sources, and thus have the capacity for generating a very high level of genetic variation. Pathogenicity islands Of particular interest in the context of host-pathogen relationships are the genetic elements called pathogenicity islands (PAIs). As you may remember from Block 1 Unit 5 (Section 5), PAIs are large regions of a bacterial genome that are present in the genomes of pathogenic strains but absent from the genomes of the same or related non-pathogenic bacterial species. They are transferred horizontally among bacteria and identical PAIs have been found in bacteria that cause different diseases. They can influence the virulence of a bacterium in a variety of ways, for example, by coding for toxin production. (The topic of virulence is discussed further in Section 2 of this unit.) PAIs were only recently discovered and important questions about them are still to be answered, such as: where do they come from and what kind of selective processes are involved in their transfer? Within-host variation Largely as a result of the development of modern molecular techniques, biologists are now much more aware of the extent, and the causes, of genetic variation in pathogens. Previously, pathogens were assumed to conform to a 'clonal model' in which a successful strain could spread through a host population, reproducing asexually, and remaining little changed genetically. It is now realised that, in many pathogens, new strains can appear very frequently because they have ways of generating genetic variation that had not previously been appreciated (e.g. PAIs). One consequence of this is that a single host individual may be host to more than one pathogen strain at the same time. Such within-host variation has important consequences for the severity of illness experienced by hosts. Experimental studies have compared the effect of injecting mice with Plasmodium cultures of single and of mixed genotypes. The immune response that mice have to mount against mixed genotype pathogens is more costly and they suffer more severe sickness than when injected with a single genotype. More significantly, the occurrence of within-host variation has completely altered the way in which biologists now view the evolution of a very important aspect of pathogens - their virulence (Section 2). Genetic variability in different pathogens As the number of pathogens whose genomes have been sequenced increases, it becomes apparent that some microbes are more variable than others. Very high levels of genetic variation have been found among samples of E. coli, Helicobacter pylori and Staphylococcus species, but very little variation has been found among samples of Mycobacterium tuberculosis. In contrast to the other species, the genome of M. tuberculosis shows very little evidence for the acquisition of new alleles by recombination or by horizontal transfer. High variation is related to pathogen versatility; some Staphylococcus species, for example, cause many kinds of disease in many different host species. 1.2.5 HIV and genetic variation The capacity of pathogens to generate very high levels of genetic variation is strikingly illustrated by HIV. You should recall that HIV is an RNA virus. RNA is intrinsically less stable than double-stranded DNA, both because it breaks more often and because DNA can be repaired using a complementary strand that is obviously absent in single-stranded RNA. • Recall the peculiar feature of RNA replication in HIV. Answer HIV is a retrovirus, which means that DNA can be produced from its RNA template using reverse transcriptase. Reverse transcriptase has a high error rate, leading to base substitutions, insertions or deletions. The consequence is that, on average, there is about one mutation per genome in every replication cycle. Furthermore, recombination may occur between the two RNA strands in the HIV virus. • What do you predict is the consequence of the high mutation rate and the recombination between strands? Answer It will lead to high genetic variation in the virus. Because HIV is able to replicate very rapidly, the genetic diversity is apparent even within a single host. • Identify another feature of HIV that is particularly relevant to this fact. Answer It has an extremely long period of infectivity, which can be as much as 10 to 15 years. Therefore, HIV has a very long time within a host in which to reproduce and generate variation. HIV sub-populations and genetic drift When HIV infects a person, it reproduces and migrates to different parts of the body, especially to the lymph nodes. As a result, several sub-populations are formed that are largely physically isolated from one another. This situation is very similar to the distribution of many animals and plants in fragmented landscapes and, in the language of ecology, is called a metapopulation (a population of populations). Each sub-population may differ genetically from the others from the start, because of genetic drift, and, as reproduction proceeds, sub-populations tend to diverge more and more over time. A genetic analysis of HIV samples taken from different parts of one dead person's body revealed very large genetic differences between different sub-populations. In fact, the genetic differences revealed are greater than those that separate humans from chimpanzees; the latter having evolved over 5 to 10 million years. Genetic variation is much greater in HIV-1, which also has a much wider geographic distribution, than in HIV-2. This example raises an important general point about the nature of the environment that hosts provide for pathogens. It has been suggested that, to many pathogens, 'we and other mammals ... are little more than soft, thin-walled flasks of culture media' (Levin and Antia, 2001). However, in reality, the host environment is much more diverse. A large, complex organism like a human provides a very varied environment for a pathogen, which can be regarded as a number of 'habitats' or habitat patches. Most pathogens are specialised to colonise only one habitat (e.g. the influenza virus colonises the respiratory tract), but some can colonise several habitats. The colonisation of different parts of a host's body by a pathogen has two important potential consequences. • As in the example of HIV described above, it can lead to the pathogen population forming a metapopulation structure with consequent considerable genetic variation. • A particular pathogen may have very different effects on its host, depending on which part it has colonised. For example, Neisseria meningitidis is typically a commensal inhabitant of the upper respiratory tract, but it is a highly virulent pathogen when it crosses into the brain, causing meningitis. 1.2.6 Overview of variation and reproduction You have seen that coevolution between hosts and pathogens is a process in which each partner is adapted to counteract the adverse effects on it caused by the other. This is a process that may never reach a stable or static outcome. • What is the hypothesis that suggests that coevolved organisms are in a state of continuous evolution? (Hint: you read about this briefly in Block 1 Unit 8.) Answer The Red Queen hypothesis (see Block 1 Unit 8, Section 6). Thus, both partners are constantly generating genetic variation. For either partner to continue to exist, it must continually counteract the genetic changes that arise in the other. An important feature of the coevolutionary relationship between hosts and pathogens is that it involves partners that differ, sometimes markedly, in the mechanisms that each possesses for generating genetic variation. They also differ markedly in another important respect - generation time. Because genetic variation largely arises from, and can only be passed on during, reproduction, it follows that the rate at which an organism can generate new genetic variants depends on its rate of reproduction, which is determined by its generation time. Humans have a generation time measured in years; many pathogens have generation times measured in hours or a few days. It follows that pathogens can, potentially , generate new genetic variants at a much higher rate than their hosts.

Extrinsic factors affecting dynamics

Why might some diseases show cycles or predictable fluctuations in abundance? These fluctuations may be linked to the seasons, e.g. the higher incidence of influenza in winter or the onset of yellow fever with the rainy season, due to the hatching of eggs of previously infected mosquitoes. In these cases, there are clear reasons why the pathogen population should change in abundance - survival, fecundity and/or transmission of the pathogens can all be affected by the environmental conditions. Another more contentious example is the outbreaks of Vibrio cholerae infection that may be associated with climatic cycles and ocean plankton blooms. • With reference to Block 2, what else might be affected by the environmental conditions for a given pathogen species? • The host's defences may be impaired (e.g. immunodeficiency). These explanations of pathogen abundance driven by external environmental patterns are examples of factors extrinsic to the host-pathogen system. The same factors may have been responsible for shifts in the range of the pathogens over time. In some cases, there appear to be no obvious extrinsic factors. In these examples, cycles or predictable fluctuations of outbreak years may occur over tens of years. A set of data on epidemics recorded in the USA from 1657 to 1918 by present-day genealogists is given in Figure 3.8. • How reliable are these data? • They depend on correct diagnosis at the time - this may be incorrect for influenza and possibly yellow fever (which may have been confused with hepatitis). They are also not a systematic sample - there is not a standard definition of an epidemic and they have not been taken from a fixed number of people in a standardised manner. (See Unit 2, Section 4 of this block for examples of how epidemiological data can be sampled.) Despite concerns over reliability, these sets of data are indicative of the likely frequency of major outbreaks of three given diseases. Note that they are not records of pathogen abundance but indications of disease prevalence. More reliable sets of data are provided in the returns of general practitioners and other health workers from the mid-twentieth century to the present day (see Unit 2). • What conclusions can be reached about the frequency of occurrence of the epidemics of the three diseases shown in Figure 3.8? • Some of the yellow fever epidemics are widely spaced, while others appear to be tightly clustered. There is no obvious pattern with the influenza epidemics. The measles epidemics may be increasing in frequency (but are restricted to examples prior to 1800). Indeed, the frequency of occurrence of these measles epidemics (every ten years or more) can be contrasted with those every two years in England and Wales from 1948 to 1982. As you will see in Unit 4, the analysis of measles data has been particularly fruitful for its insights into cycles of pathogen abundance. For example, Grenfell and co-workers (1994) used weekly statistics on measles data from 1944 to 1994 in England and Wales to illustrate waves of infection originating in large cities and then spreading to surrounding towns and villages. This gives the cycles of abundance a spatial dimension, i.e. the peaks of abundance are not restricted to one particular locality but are able to spread between localities. It may be that there are some correlations of the outbreaks with climate patterns, but this is not always the case.

Enzyme-linked immunosorbent assay (ELISA)

Diagnostic technique in which serum samples containing antibodies are placed in microtitre wells that are precoated with specific microbial antigens. If the patient has antibodies against that antigen, specific binding occurs, and unbound material is then washed out of the wells. Next, an anti-human antibody is added, and this binds to any of the patient's antibodies that remain in the well. Again, unbound material is washed out. The second antibody is attached to an enzyme that catalyses a colour change when its substrate is added. Thus colour changes in the microtitre wells indicate that the patient's serum contained specific antibody and the intensity of the colour reaction can be used for relative quantification.

Cercaria

(Plural, cercariae.) The larval form of the parasite, developed within the germinal cells of the sporocyst or redia. A cercaria has a tapering head with large penetration glands. It may or may not have a long swimming 'tail', depending on the species. The motile cercaria finds and settles in a second intermediate host where it will become either an adult or a metacercaria, according to species.

Direct life cycle

A parasite life cycle in which the parasite passes from one definitive host to the next, without any intermediate hosts or vector.

Incipient mutualism

A model of coevolution in which the pathogen and the host come to benefit each other.

Prudent pathogen

A model of coevolution in which the pathogen evolves towards doing less harm to the host.

Vector

An agent such as a bacterial plasmid or disabled virus that is used to carry DNA into cells. The same term is applied to an organism that carries a parasite or pathogen from one host to another.

Homogeneous mixing

Characterises a population in which effective contacts occur at random between individuals in the population.

Schistosomiasis

An important human parasitic disease caused by infection by blood flukes of a number of species of Schistosoma (schistosomes). The disease was originally known as bilharzia.

Parasitaemia

Any condition when parasites are present in the blood. The extent of parasitaemia can be quantified by counting the number of parasites per millilitre of blood or, for parasites that infect red blood cells (e.g. Plasmodium), by calculating the percentage of red blood cells that are infected.

The global burden of infectious disease

As a reminder of the scope of the infectious disease burden already summarised earlier in this module (e.g. Block 1 Unit 1 and the case studies on influenza, cholera, tuberculosis and malaria), consider the following facts. In 2008, a total of 8.8 million children died before their fifth birthday - half of them in Sub-Saharan Africa. Pneumonia, diarrhoea, malaria, HIV/AIDS and measles caused 44% of these deaths in children under five years (Figure 1.1). Newborn deaths from sepsis and tetanus in the first four weeks of life accounted for a further 7%, and many of the 17% of deaths attributed to 'other causes' were also due to infection, including TB and meningitis exacerbated by malnutrition, which contributes to over one-third of all child deaths. (Note that the 'non-infectious neonatal causes' depicted in the figure refer to deaths resulting from prematurity, birth asphyxia, trauma and congenital abnormalities.) Although the impact of infectious disease episodes and deaths is disproportionately felt among the populations of low- and middle-income countries (LMICs), the richer nations of the world have not escaped. For example, infectious and parasitic diseases were responsible for 201 751 admissions to National Health Service (NHS) hospitals in England in the budget year 2010-2011, an increase of 10% on the previous year - the biggest percentage rise of any disease admission group (Hospital Episode Statistics, 2011). It was largely due to a 29% increase in admissions of children aged under 15 with non-hepatitis, non-influenza viral infections, and it confirms a rising trend. This category of hospital admissions has risen by 84% in England since 2006-2007. 2.1 Acute respiratory infections Influenza, pneumonia and other acute respiratory infections (ARIs) are often forgotten in the focus on TB, HIV/AIDS and malaria, but they rank first among infectious causes of disease worldwide: • The average global burden of seasonal influenza alone is about 600 million cases per year, of which 3 million result in severe illness, causing between 0.25 and 0.5 million deaths (WHO, 2009a). As you know from the Influenza Case Study, pandemic influenza occugrs at unpredictable intervals with the emergence of new variant influenza viruses. • The 'swine flu' epidemic in 2009 caused by an H1N1 virus spread rapidly to 208 countries; the speed of transmission overwhelmed the capacity for laboratory confirmation of cases, especially in low-resource countries, but clinical diagnosis suggests that several million people were symptomatically infected and at least 12 220 died (WHO, 2009b). • Pneumonia is the largest infectious cause of death among young people globally (see Figure 1.1), accounting for an estimated 1.5 million deaths in children under 5 years - more than TB, HIV/AIDS and malaria combined in this age group. Pneumonia is mainly due to the bacteria Streptococcus pneumoniae and Haemophilus influenzae type b (Hib), the respiratory syncytial virus (RSV) and parainfluenza viruses, all of which also affect elderly people worldwide. A major underlying cause of susceptibility to these pathogens is inflammation of the lungs due to atmospheric pollution; this is a major problem particularly among women, children and elderly people who are persistently exposed to indoor smoke from cooking fires in poor rural communities. • The World Health Organization (WHO) estimates that up to 18 million episodes of pneumococcal disease and around 16 million episodes of RSV disease occur globally every year. 2.2 Diarrhoeal diseases Worldwide, there are about 2 billion cases of diarrhoeal disease every year, including cholera, dysentery (bloody diarrhoea), giardiasis and a long list of bacterial and viral causes. They are the second largest infectious cause of death among young children (Figure 1.1), killing about 1.4 million annually, 80% of them under two years of age (WHO, 2009c). Even in the USA, the richest nation on Earth, diarrhoeal diseases cause an estimated 38.4 million episodes annually and around 1700 deaths (Scallan et al., 2011). The pathogens causing diarrhoeal diseases are transmitted in contaminated food and water and from hand to mouth (the faecal-oral route). A major underlying cause is the shaming fact that in 2010 around 780 million people (11% of the global population) lacked access to improved sources of drinking water (Figure 1.2), and 2.5 billion - 37% of the world's population - had no access to basic sanitation (UNICEF and WHO, 2012). However, steady progress is being made: in the 20 years from 1990 to 2010, an estimated 2 billion people gained access to improved drinking water and 1.8 billion gained access to improved sanitation (i.e. a covered pit latrine or better). These changes are gradually reducing the impact of diarrhoeal diseases on children's health. 2.3 Tuberculosis As the Tuberculosis Case Study demonstrates, TB has overtaken HIV/AIDS globally as the largest infectious cause of death by a single pathogen. Kaufmann reports that in 2011: • Every minute of every day, nearly 20 people were infected with Mycobacterium tuberculosis and four people died from TB. • One-third of the global population (well over 2 billion people) were carriers of TB bacteria. • There were over 9 million new or relapsed active cases. • Over 2 million people with chronic TB died. Although the prevalence of TB is highest in the poorer countries of South-East Asia and Sub-Saharan Africa (Figure 1.3), it is resurging in Eastern Europe and increasing in the richest parts of the world. For example, there were over 9000 new cases of TB in England and Wales in 2011, most of them in London (HPA, 2012). 2.4 HIV/AIDS Worldwide, HIV incidence (the number of new cases occurring in a given period, usually one calendar year) has stabilised and deaths have been declining in recent years (UN, 2010). However, the impact on the global burden of HIV-related disease is still huge. Global HIV incidence stabilised at around 2.7 million new HIV infections annually between 2007 and 2010 (the most recent year for which data are available at the time of writing). AIDS-related deaths fell from 2 million in 2008 to 1.8 million in 2010 due to the expansion of access to effective antiretroviral therapy. One outcome of this success is that HIV prevalence (the number of people living with HIV infection) is steadily increasing - to 34 million in 2010. Just over 2 million of those people were children under 15 years infected via mother-to-child transmission (WHO, UNAIDS and UNICEF, 2011). People with HIV are primarily in LMICs, but richer parts of the world are also affected. For example, according to the Health Protection Agency, 91 500 people in the UK were estimated to be living with HIV at the end of 2010, of whom 6660 were newly diagnosed in that year and around 24% were unaware of their infection status (HPA, 2011a). 2.5 Malaria The prevalence and incidence of malaria has also been steadily decreasing worldwide, but 216 million new cases still occurred in 2010, with an estimated 655 000 deaths in that year, mainly among young children and pregnant women. It is a shocking truth that an African child dies from malaria every 60 seconds. Although there were 126 000 fewer deaths globally in 2010 than in 2009, malaria still accounted for 22% of the deaths of African children (WHO, 2011a). In October 2011, some progress was announced from large-scale clinical trials of the RTS,S/AS01 malaria vaccine, which reduced the incidence of new infections among young African children by 50% (RTS,S Clinical Trials Partnership, 2011). This is a significant breakthrough, representing the first vaccine with established clinical effectiveness in preventing a human infectious disease caused by a parasite. However, it is not yet known how long the protection will last, and 50% efficacy is well below the desired 95% achieved by some well-established vaccines, e.g. against measles or diphtheria. But the RTS,S/AS01 vaccine also includes surface antigens from hepatitis B viruses and gives good protection against the latter. Combined (or combination) vaccines such as this one are highly effective because they protect children against two or more diseases at the same time. • Which other combined vaccines can you remember from Block 2, or from your own experience? • You may have suggested the MMR vaccine against measles, mumps and rubella; or the DTP vaccine against diphtheria, tetanus and pertussis (whooping cough). 2.6 Other infectious disease headlines Here are some other estimates from the WHO (accessed in 2011). • Over 350 million people are chronically infected with hepatitis B virus (HBV) and 130-170 million with hepatitis C virus (HCV), causing over 1 million deaths from liver disease and hepatic cancer annually (WHO, 2008, 2011b). • Excluding HIV and HBV, there are around 350 million new cases of the major sexually transmitted infections, including gonorrhoea, chlamydia (the subject of a case study later in this block) and syphilis, which has a disproportionate effect on infants. Approximately 12 million new infections with Treponema pallidum pallidum occur every year, including 1.5 million babies born with congenital syphilis - the most common infection passed from mother to newborn baby. This is more prevalent even than congenital HIV, but far less publicised. In Africa alone, congenital syphilis causes the death of almost 500 000 babies every year (WHO, 2007). • Around 1.5 billion people are infected with intestinal worms. A typical child in a poor rural environment in a low-income country commonly carries around 1000 hookworms, roundworms and whipworms, causing anaemia, stunted growth and increased vulnerability to other infectious diseases (WHO, 2012a). • The so-called 'neglected tropical diseases' include Schistosoma parasites, which infect 207 million people worldwide - 85% of them in Africa, causing an estimated 200 000 deaths annually. Around 12 million people in 88 countries are chronically infected with Leishmania parasites, which cause severe cutaneous (Figure 1.4a) or visceral (Figure 1.4b) disease. • Around 500 000 people are blinded by the microscopic parasitic worm Onchocerca volvulus (WHO, 2012c) and over 120 million people worldwide are infected with Wuchereria bancrofti, the parasite causing lymphatic filariasis (WHO, 2012d). • The viral haemorrhagic fevers are epidemic-prone diseases, which are increasing their geographical range and the number of people affected. Annually, it is estimated there are: • 200 000 cases of yellow fever, causing 30 000 deaths • 300 000-500 000 cases of Lassa fever, causing 5000 deaths • up to 50 million dengue virus infections, including at least 250 000 cases of haemorrhagic fever and 24 000 deaths (WHO, 2012e). With this daunting background in mind, the next section illustrates the diversity of public health approaches to controlling the huge burden of infectious diseases. An additional aim is to demonstrate the importance of addressing local needs and engaging local communities in interventions to improve public health, as you will see particularly in a slidecast on 'Infectious disease and public health in rural Ethiopia' in Section 5 (Video 1.1), and when you read about the guinea worm eradication campaign in Section 6.

The force of infection

As indicated earlier, transmission rates, although fundamental to any understanding of infectious diseases, are difficult to estimate. The difficulty is due to the fact that it is not usually possible to document all contacts. For some less-common yet serious infectious diseases, such as tuberculosis and AIDS in HICs, much effort is devoted to contact tracing, namely identifying the contacts of cases in order to control the spread of infection. For most infections, however, such an exercise is impractical. 8.1 Incidence rates of infection It is considerably easier, however, to calculate the incidence rates of infections, that is, the rate at which individuals become infected, rather than trace specific contacts. For example, if in a given city of, say, 100 000 inhabitants, 800 cases of measles are observed over a six-month period (i.e. 0.5 years), an estimate of the (annual) incidence rate of measles is simply: Clearly, as discussed earlier in this unit, there are all kinds of practical difficulties with such calculations, since not all cases are reported and not all infections produce clinical symptoms. Because clinical disease is usually the trigger for a case to consult a doctor, subclinical or mild cases will not be included in official statistics. • Suppose that 50% of infections with Bordetella pertussis produce typical symptoms, and that on average doctors notify only 25% of the typical whooping cough cases that they see. a. Calculate the proportion of infections that are notified as whooping cough cases. b. What other major source of inaccuracy might affect notification data? c. How will this affect the incidence rate? a. The proportion of infections included in the whooping cough notifications is 0.5 × 0.25 = 0.125, or 12.5%. b. Another source of inaccuracy in notification data stems from the fact that diagnoses are often based on clinical criteria, which may be non-specific. For example, other infectious agents may produce similar clinical symptoms to Bordetella pertussis. c. This non-specificity of clinical diagnoses will tend to exaggerate incidence. Incidence rates do not reflect transmission rates However, a more fundamental shortcoming of incidence rates is that they do not reflect transmission rates; this is because they take no account of susceptibility. Consider, for instance, the following example. A family includes five people: two adults and three children aged 6 months, 2 years and 5 years. None of the family is vaccinated against measles. The 5-year-old develops measles. The other four members of the family can be assumed to be equally exposed to the measles virus in the home, from contacts with the infectious 5-year-old. However, secondary cases can occur only among susceptibles. The two adults are likely already to have had measles, and hence to be immune: if so, they are not susceptible. The two children, on the other hand, are quite likely not to have had measles, and hence will be susceptible. Thus, if infection spreads in the family, the children are more likely than the adults to become infected, even though the contact rates are the same for all members of the family. • Table 2.6 and Figure 2.14 show the age distribution of notified measles cases in England and Wales for a period prior to the introduction of measles vaccination in 1968. In which age groups is the annual incidence of measles highest? What else might you want to know in order to compare transmission rates in different age groups? Table 2.6 Measles notifications by age, England and Wales, 1956-1965 Age group/years Notifications 0 221 267 1 424 616 2 544 251 3 580 855 4 595 364 5-9 1 868 340 10-14 109 937 15-24 26 790 25+ 15 855 All 4 387 275 Figure 2.14 Histogram of measles notifications in individuals aged less than 25 years, England and Wales, 1956-1965 View description • Note that the histogram in Figure 2.14 has varying bar widths in order to allow for the different age group widths. The incidence increases with age to a peak in 4-year-olds, then declines. In comparing transmission rates in different age groups it is important to remember that data are based on notifications, and that completeness of notifications might vary with age. However, it is perhaps unlikely that it would vary quite enough to explain the big differences in incidence shown in Figure 2.14. Even if notification rates do not vary, we cannot conclude that the transmission rate is higher among 5-9-year-olds than among 10-14-year-olds, say, since many more 10-14-year-olds than 5-9-year-olds are likely to be immune. 8.1.1 Infection rate of susceptible individuals Some elaboration of the notion of incidence rate is required to take account of susceptibility, and so allow comparisons to be made between different subgroups of the population, and in particular between different age groups. The concept needed for this is the force of infection. The force of infection is the rate at which susceptible individuals become infected and is denoted here by λ (lambda). It can be calculated in much the same way as an incidence rate, except that the denominator population used in the calculation (i.e. the number of people exposed to the infection) includes only susceptible individuals: immunes are excluded from the denominator. • Cytomegalovirus (CMV) infection in pregnancy can cause hearing loss, visual impairment, and neurological problems in the child. These problems appear to occur primarily in the babies of women not previously infected. Data from the USA suggests that about 70% of women of childbearing age have been infected. It is estimated that about 2% of pregnant women first become infected with CMV during pregnancy. Calculate: a. the annual incidence rate b. the annual force of CMV infection in pregnant women in the USA. • a. Remember that Since pregnancy lasts approximately 9/12 = 0.75 years, and 2% = 2/100, the incidence rate is: b. Out of 100 pregnant women, about 30 have not had a prior CMV infection and are thus susceptible (since 70% have been infected). So the average annual force of infection, λ, is: 8.1.2 Force of infection proportional to transmission rate Unlike the incidence rate, the force of infection, λ, is directly proportional to the effective transmission rate, β. Hence it makes sense to compare forces of infection in different subgroups of a population, in different populations and even between infections. This may be done directly using serological surveys: the higher the force of infection, the greater the rate at which susceptibles become infected and, hence, the steeper the rise in the proportion of the population that has specific antibodies against the infectious agent in question. The proportion that is seropositive (i.e. with antibody) in each age group (and, hence, with evidence of previous infection) is plotted as separate columns for people from large and small families. Figure 2.15a shows that the proportions infected grow more rapidly for children in large families, reflecting the higher force of infection in this group. This, in turn, is the result of more frequent exposure, and hence higher effective contact rates. Caution is required in interpreting serological data in the presence of vaccination, since both naturally occurring infection and vaccination elicit an antibody response. Figure 2.15b, for example, shows the proportion of people with rubella antibody by age and sex in the UK in 1980-1984. The small excess of seropositives in females aged 10-25 years is most probably the result of the selective rubella vaccination programme for girls, which was in operation at the time. Force of infection and the SIR model In terms of the SIR model in Section 5, the force of infection is the transition rate at which individuals move from the susceptible compartment to the infected compartment. To fully describe the SIR model, it only remains to specify δ (the Greek symbol delta), which is the transition rate between the infected and the recovered classes. Since the infectious period is D, an individual experiences 1 recovery in D units of time. Hence the rate at which infectious individuals recover is δ = 1/D. The fully specified SIR model, complete with transition rates, is shown in Figure 2.16.

R0

As you will see later, R0 plays a central role in infectious disease epidemiology. One reason is that if R0 is less than (or equal to) one, the infection will die out (assuming no reservoir species), whereas if R0 is greater than one, there may be a large epidemic and the infection may become endemic, that is, ever-present. To understand this, it is useful to think in terms of generations of infected individuals. An outbreak begins with one or more introductory cases. These form generation zero. The introductory cases may directly infect other people, who form the first generation of spread. These in turn infect a second generation of cases, and so on. For example, Figure 2.2 shows the dates of onset of symptoms in cases of measles during a school outbreak in the USA in 1985. The outbreak is thought to have originated from a visitor to the school, who unknowingly infected two children in the first generation of spread. During the early stages of the outbreak, the cases are clumped together in generations 10-14 days apart (i.e. from the start of one outbreak to the start of the next outbreak). Later on, the generations begin to overlap and become more difficult to distinguish. The outbreak depicted in Figure 2.2 occurred in an immunised population, so it reduced to zero after a few generations. But consider what would happen in a large, totally susceptible population: • One infection causes, on average, R0 secondary infections, which constitutes the first generation of cases. • Each one of these cases will also cause, on average, R0 infections, so that there will be, on average, R0 × R0 = R02 (i.e. 'R0 squared') infections at the second generation. • At the third generation there will be, on average, R0 × R0 × R0 = R03 (i.e. 'R0 cubed') infections, and so on. The average number of infections in successive generations will begin to increase very quickly if R0 > 1 (i.e. R0 is greater than 1), and reduce to zero if R0 < 1 (i.e. R0 is less than 1), as illustrated in Figure 2.3, where R0 = 2. (The average number of infections will also go to zero if R0 = 1, due to chance fluctuations.) • Suppose that R0 = 3. Calculate the number of infections, on average, in successive generations, following the introduction of a single infection. Can this trend continue indefinitely? • The average numbers in successive generations are 1, 3, 9, 27, 81, 243, 729 ... The trend cannot continue indefinitely, as eventually the number of susceptibles will be depleted, thereby limiting the spread of infection. But by then a large epidemic will have occurred. 3.3 R0 as a measure of scale and severity of epidemics A second reason why R0 is important is that the larger its value, the more effort is needed to control the infection. This makes sense intuitively: the larger the value of R0, the steeper the growth of the epidemic, and hence the more difficult it is to control it. • Table 2.4 shows some typical values of R0 for selected infections. a. Which infection is likely to require most effort to control? b. Which is likely to prove easiest to control? c. Is the magnitude of R0 the only relevant factor in controlling an infection? Table 2.4 Values of R0 in different locations and time periods for selected infectious diseases Infectious disease Location and time period R0 Measles England and Wales, 1947-50 Ghana, 1960-68 Eastern Nigeria, 1960-68 13-14 14-15 16-17 Whooping cough (pertussis) England and Wales, 1944-78 USA, 1943 16-18 16-17 Chickenpox England and Wales, 1944-68 USA, 1943 10-12 10-11 Mumps England and Wales, 1960-80 Netherlands, 1970-80 11-14 11-14 Rubella England and Wales, 1960-70 Poland, 1970-77 Gambia, 1976 6-7 11-12 15-16 Polio Netherlands, 1960 6-7 HIV Uganda, 1985-87 10-11 Smallpox West Africa, 1968-73 3-5 • a. Measles and whooping cough both have the greatest R0 value ranges of 13-18 and so are the most difficult to control of the infections listed. b. Note that the infection with the lowest value of R0 in Table 2.4 is smallpox, which has now been eradicated globally. c. The magnitude of R0 is not the only factor that determines how difficult it is to control an infection. Other relevant factors include, for example, the availability of effective vaccines, and the existence of a healthcare network to carry out large-scale vaccination programmes.

Manipulation by pathogen

Crickets are terrestrial animals that normally avoid water. However, those insects that are infected with hairworms (genus Nematomorpha) hurl themselves into ponds and streams, where they are eaten by fish. This is but one example of manipulation of host behaviour by a pathogen. The host's behaviour is maladaptive for itself but is adaptive for the pathogen; in this example, fish are the final hosts and crickets an intermediate host for the pathogen. Many examples of host manipulation involve pathogens with complex life cycles, in which it is the behaviour of an intermediate host that is manipulated in such a way that the chances of the pathogen's life cycle being completed are increased (see Table 5.1 and Figure 5.4). • Recall from Block 1 Unit 8 an example of the manipulation of a host by a pathogen. Answer The liver fluke Dicrocoelium dendriticum affects the behaviour of its intermediate ant host such that it maximises the chance of the infected ant being eaten by sheep or cattle, thus allowing the fluke to reach its definitive host. Beetle, Tenebrio monitor (final host is rat) tapeworm, Hymenolepis diminuta Infected beetles invest less in reproduction and consequently survive longer (40% increase in survival time in females, 25% in males) Mammals rabies virus Infected host develops rabid behaviour, which includes intense salivation and tendency to bite other animals Snail, Succinea (final host is bird) trematode, Leucochloridium macrostomum Parasite sporocysts invade snail's tentacles, causing them to swell, making snail very conspicuous to birds and thus more likely to be eaten (see Figure 5.4) Banded killifish, Fundulus diaphanus (final host is bird) digenean trematode, Crassiphiala bulboglossa Parasitised fish are hungry, move to periphery of shoal to find food, and are more likely to be eaten by birds Mosquito malaria pathogen, Plasmodium sp. Infected mosquitoes continue feeding throughout the night, whereas uninfected ones do not; as a result, infected mosquitoes bite more hosts Another remarkable form of host manipulation by symbiotic microbes is shown by rickettsial bacteria of the genus Wolbachia (Figure 5.5). These organisms infect arthropods and nematode worms (and may infect other invertebrate species) and are primarily transmitted vertically in the cytoplasm of host eggs, but can also be transmitted horizontally, sometimes between host species. In line with the general prediction that vertically transmitted symbionts will be benign or mutualistic, they are generally not virulent and, indeed, there are cases where they enhance host fecundity. However, the fact that they are transmitted exclusively through host females means that they are very hostile to the males of host species. Some Wolbachia species kill male host embryos, others render sperm incompatible with uninfected eggs, and others cause males to become females. In some host species, normally sexually reproducing females become parthenogenetic as a result of Wolbachia infection, rendering host males redundant. It seems that Wolbachia is very widely distributed. In 1997 it was estimated to be present in 15 to 20% of all insect species, but it is now thought to be present in about 90% of all insects! Although usually benign, at least to female hosts, some forms of Wolbachia are highly virulent. A strain found in Drosophila flies, charmingly called 'popcorn', becomes active as its host reaches maturity and kills it before it reproduces. Research into Wolbachia has increased greatly in recent years, much of it driven by the possibility that Wolbachia could be used as a means to control insect pests, including disease vectors. For example, research is currently under way to see if Wolbachia can be used to control tsetse flies. The human disease river blindness (onchocerciasis) is caused by the insect-borne nematode Onchocerca volvulus. It has been discovered that the immune reaction in the cornea that leads to blindness is not due to the nematode itself, but to endosymbiotic Wolbachia in the nematode. Treating the nematode with antibiotics eliminates the Wolbachia; it no longer causes blindness and, furthermore, becomes sterile. 3.2 Pathogenic manipulation of humans Pathogens affect the physiology and behaviour of their hosts in a variety of ways. In the context of human diseases, these changes, such as sneezing or diarrhoea, are generally regarded as symptoms. There are three ways of looking at these behaviours: 1. they are incidental consequences of infection, and are of no functional significance to either host or pathogen 2. they are adaptive, defensive reactions to infection by the host 3. they are changes in the host caused by pathogen manipulation and are adaptive for the pathogen, e.g. in aiding transmission. Categories 2 and 3 are not mutually exclusive - it is possible for a change in a host to be adaptive for both the host and the pathogen. Evolutionary medicine The possibility that symptoms of disease may be adaptive responses on the part of the host belongs to an area of biology called evolutionary medicine, which was pioneered by Randolph Nesse and George Williams (1995). This school of thought takes the view that symptoms of disease such as pain, coughing, nausea, vomiting, diarrhoea and fever are defensive responses by the host that reduce the impact of infection. Vomiting and diarrhoea are mechanisms that eliminate pathogens from the body quickly; fever changes the host's body temperature in a way that favours the host's defence systems over the pathogen (see Section 3.2.1). This way of viewing the world of infectious diseases raises a lot of interesting ideas but is fraught with pitfalls. A problem with evolutionary theory is that it can be used to explain everything by applying the general argument that 'if it exists, it must be adaptive'. This does not advance our scientific understanding. Explanations that invoke adaptation should be regarded as hypotheses that can be rigorously tested. If we wanted to test the hypothesis that vomiting is an adaptive response to infection, for example, we would need to find a way of preventing it and then seeing if subjects become more ill as a result. There have been very few rigorous tests of ideas put forward under the heading of evolutionary medicine and so, for the present, it is best regarded as an interesting alternative way of looking at some aspects of infectious disease. 3.2.1 The role of fever Increased body temperature is a common symptom of infectious disease. There has long been a debate as to whether fever is an example of host manipulation by pathogens or is an adaptive response by the host. Fever as a survival mechanism This question has been addressed experimentally in reptiles, and mammals, such as rabbits. Reptiles are ectotherms, meaning that they acquire body heat from their external environment, moving to warmer places to raise their body temperature; mammals are endotherms, meaning that they generate heat internally. With reptiles, it is relatively easy to experimentally change their body temperature, by simply controlling the temperature of their environment. Several studies of this kind have shown that the survival of infected animals is increased if their body temperature is raised. With mammals, the most commonly used experimental approach is to suppress the physiological responses that cause fever; many such studies have shown that this procedure reduces the survival of infected animals. Fever and stimulation of the host's immune response At one time, it was thought that fever acts as a defence against infection by suppressing bacterial or viral reproduction, but this is not usually the case. Rather, it seems to enhance certain aspects of the host's immune response, e.g. increasing the rate of lymphocyte division. In particular, it acts synergistically with the host's physiological mechanisms that reduce the amount of iron circulating in the body. Iron is vital for microbe reproduction and denying them access to it reduces their reproductive rate. Host manipulation Views on the importance of host manipulation as an outcome of the coevolution of hosts and pathogens are varied, and have shifted over the years. Examples like those listed in Table 5.1, which mostly relate to the manipulation of intermediate rather than final hosts, lead some researchers to see manipulation as a general characteristic of pathogens. However, a detailed analysis of the literature by Poulin, published in 2000, showed that the concept of host manipulation was a 'weakening paradigm'. On the other hand, it appears that the concept of host manipulation is staging a revival, primarily as the result of new discoveries about the very intimate ways in which pathogens interact, at a subcellular level, with their hosts. These ways include the process of apoptosis (Block 1 Unit 3, Section 6.1) in which the response of hosts to infection is to selectively kill off infected cells. As might be predicted from coevolution, some pathogens can prevent this threat to their survival by switching off apoptosis, for example, cytomegalovirus (CMV).

Population dynamics of pathogens in the host

At several points in this unit, and in other parts of the module, the potentially high rates of population increase of different pathogens have been discussed. In this section, some of these rates are quantified with reference to bacteria and viruses, and their implications for survival of the pathogen in the host are discussed. Note that this section focuses on population dynamics within the host - the parallel story of the spread of disease between hosts, drawing on similar methods, is given in the compartmental models of this and the previous unit.

Disadvantages of case studies and laboratory reports

Both case reports and laboratory reports have a major shortcoming, at least for the purposes of studying the dynamics of transmission - lack of completeness. In other words, only a fraction of cases are reported or tested in the laboratory. As a result, many infections are never documented. For example, studies of notified cases of whooping cough in the UK suggest that less than 20% of cases are notified. On the other hand, lack of completeness need not be a hindrance to the study of trends (as shown in the graphs in the previous section), provided that the proportions of cases reported remain constant over time. A further difficulty stems from the fact that not all infections result in disease states sufficiently serious to warrant seeking medical advice. For example, only about 60% of mumps infections produce clinical symptoms. The remaining 40%, which are called subclinical infections, will never come to the attention of a GP, and so will never be notified or sampled for laboratory analysis. Nevertheless, individuals with subclinical infections can pass on the pathogen and so play an important role in the transmission of the infection. For example, less than 10% of poliovirus infections are symptomatic (and less than 1% cause paralysis), but asymptomatic infected individuals can still transmit poliovirus.

Common cold

Common colds account for over one-third of all acute respiratory infections in humans, and carry substantial economic costs from days off work. Rhinoviruses are responsible for about 30% of common colds, and coronaviruses for 10%. Many other infectious agents cause symptoms indistinguishable from those caused by rhinoviruses and coronaviruses. In 1946, the Medical Research Council in the UK set up the Common Cold Unit on Salisbury Plain to undertake laboratory and epidemiological research. Volunteers were recruited to take part in experiments on the transmission and treatment of the common cold. In 1965, human coronaviruses were first isolated from volunteers at the Unit. Advertisements for volunteers were placed in newspapers and magazines. They were paid a small amount, and a stay at the Unit was presented as an unusual holiday opportunity. Volunteers usually stayed for ten days, and were housed in twos and threes, but were strictly isolated from others during their stay. The last batch of volunteers left the Common Cold Unit in 1989, ending a unique chapter in the study of infectious diseases. Unfortunately, a cure for the common cold remains elusive. One of the few opportunities for a more direct assessment of transmission risks is provided by contact studies, sometimes carried out in the context of an infection outbreak. Primary and secondary cases The first case of infection (the primary case) within a defined group such as a family unit is identified. Cases infected by the primary case (these are called secondary cases) are then documented. If the number of susceptibles within the group is denoted n, and the number of secondary cases is x, an estimate of the transmission risk in conditions of close contact is represented by p = x/n. This value is sometimes called the secondary attack rate. (It is really a risk, not a rate, but the term is commonly used, nonetheless.) Note that, even if the whole group were susceptible, x is generally less than the reproduction number, R0. This is because R0 includes all secondary cases, not just those occurring within the group surveyed. For example, secondary attack rates might be calculated within families, and thus ignore any secondary cases outside the family group. Group data aggregation Usually, data are aggregated over several groups to provide a more representative value for the infection's transmission risk. For example, suppose that in a measles outbreak, primary cases are identified in five families. The families have, respectively, 3, 1, 1, 2 and 2 children who have not had measles, and the numbers of secondary cases within these five families are, respectively, 2, 0, 1, 1 and 0. The total number of susceptible children exposed is 3 + 1 + 1 + 2 + 2 = 9, and the total number of secondary cases is 2 + 0 + 1 + 1 + 0 = 4. Hence the secondary attack rate is: There are inevitably many complications with such calculations. For example, determining who is susceptible often relies on tenuous evidence; distinguishing secondary cases from subsequent cases (for example tertiary cases, namely those infected by the secondaries) can be difficult; and it has to be assumed that infections from outside the group can be ignored. Nevertheless, family contact studies can be used for comparing the transmission risks of different infectious agents. They are also used to compare secondary attack rates in vaccinated and unvaccinated family members, and hence in assessing the efficacy of vaccines in reducing transmission risks. • The data in Table 2.5 is from a study of the efficacy of whooping cough vaccine in conditions of household exposure. Calculate the secondary attack rates in the diphtheria-tetanus-pertussis (DTP) vaccinated group (pv), and in the DTP-unvaccinated group (pu), assuming that all exposed children were susceptible. Compare the rates in the two groups. What do you conclude about the effect of vaccination? Table 2.5 Number of children exposed to a primary case of whooping cough and number of secondary cases by DTP vaccination status Group status Secondary cases Exposed children DTP-vaccinated 452 1937 DTP-unvaccinated 1528 2479 • The secondary attack rate in the DTP-vaccinated group is pv = 452/1937 ≈ 0.233. In the DTP-unvaccinated group it is pu = 1528/2479 ≈ 0.616. The risk value for the unvaccinated group (pu) is much higher than in the vaccinated group (pv), so DTP vaccination appears to confer protection against whooping cough in conditions of household exposure.

Oncosphere

Egg sac containing the six-hooked (hexacanth) larva of a tapeworm.

Introduction

Having highlighted some of the key concepts of epidemiology and modelling techniques, the next section of this unit considers a simple modelling framework for the transmission of infections, in which only pathogenic microbes transmitted directly from person to person are included. Thus, for example, it does not include infectious agents for which the presence of an animal reservoir plays a major role, like toxoplasma or many salmonellas, or vector-borne infections, like malaria or yellow fever, in which the pathogen is indirectly transmitted between humans. Multicellular parasites such as helminths will also be excluded. The reason for restricting the scope of the models in this way is to keep them simple, and to focus on key concepts. However, many of the concepts that apply to directly transmitted microbial infections can also be extended to larger parasites or to indirectly transmitted pathogens.

Case reports

Instances of disease in individuals reported to a surveillance system, usually diagnosed by medical examination on the basis of clinical signs and symptoms.

Kuru

Kuru is a fatal neurodegenerative disease. It was prevalent among the South Fore people of New Guinea in Melanesia until the last century, particularly, although not exclusively, among women. Early theories that the disease was genetic in origin were discounted because of its high mortality. In the 1960s, it was demonstrated that kuru was infectious and transmitted by contact with, and consumption of, the human tissue of deceased relatives during mortuary cannibalistic rites. The higher prevalence of disease in women was due to their greater contact with infectious material during these rites. Kuru is related to Creutzfeldt-Jakob disease (CJD) and other transmissible spongiform encephalopathies (TSEs). Originally, it was assumed that the infectious agents in TSEs were viruses. In the 1980s, it was suggested that a new kind of infectious agent, a prion, was responsible. Prions are proteins and do not have a nucleic acid genome (see Block 1 Unit 4). The discovery that proteins alone can transmit infectious diseases came as a considerable surprise to the scientific community.

Host-pathogen communities

Many other species, in addition to human hosts and pathogens, are frequently involved in host-pathogen interactions. Indeed, in many cases it is not a simple host-pathogen interaction but a three-way (or more) interaction. This extends the area of study into community ecology that addresses the dynamics and other properties (such as relative abundance of species) arising from the interactions of many species. An ecological community can be defined as a set of species, populations of which may interact with each other over a given area. You have already seen many ways in which hosts and pathogens may interact, not only with each other, but also with alternative or intermediate hosts and vectors. These interactions are summarised in Figure 3.14. This module refers to these as 'host-pathogen communities' (which may include intermediate or secondary hosts and vectors). The interactions within the host-pathogen community are generally played out over small areas, e.g. in the case of malaria it is limited by the flight of the female mosquito that, in turn, is linked to the availability of habitat for egg laying. This is not always the case, especially in human examples where infected milk or meat, or the hosts themselves, may travel hundreds of miles, thereby potentially spreading the range of the community. These host-pathogen communities, and ecological communities in general, may be described as more or less diverse, depending on the number of interacting species and their relative abundance in the community (Box 3.3).

ate

The number of events occurring per unit time within a defined population.

Latent period

The time period from infection until the host begins to release the pathogen's progeny, i.e. the host becomes infectious.

Incubation period

The time period from infection until the host begins to show symptoms.

Infectious disease data

The three main sources of epidemiological data on infectious diseases that you will learn about here are case reports, laboratory reports and serological surveys. 4.1 Case reports Case reports are counts of disease, usually diagnosed by medical examination on the basis of signs and symptoms, and notified through passive or active surveillance systems. A passive surveillance system is one that relies entirely on the initiative of the individual making the report, usually a physician or public health official. In an active surveillance system, on the other hand, cases are sought, for example by means of questionnaires. Some surveillance systems do not aim to capture all events, but instead seek comprehensive reporting from a limited number of committed participants, for example general practitioners (GPs) or paediatricians; this sort of mechanism is called a sentinel surveillance system. Data from sentinel systems are often converted into population rates rather than raw counts, since the numbers of participants in the scheme may vary over time. A prerequisite for most surveillance systems is the existence of a robust public health infrastructure. This exists in many, but not all, HICs. Some LMICs, for example Cuba, also have good quality surveillance systems. In many poorer countries, however, the infrastructure required for the systematic collection of disease surveillance data is not available. 4.1.1 Notification system for infectious diseases The statutory notification system for infectious diseases in the UK is a passive reporting system (see Box 2.3). The USA also uses a national reporting system, coordinated by the Centers for Disease Control and Prevention (CDC). France has a sentinel reporting system, the Réseau Sentinelles, based on a subset of general practitioners. The Royal College of General Practitioners (RCGP) in the UK also runs a sentinel surveillance system (you saw examples of data from this source in Block 1 Unit 1). The British Paediatric Association has an active surveillance programme for selected, generally rare, paediatric infections.

Measuring population susceptibility to infection

This 'average' steady-state level can be quantified by the average proportion of the population that is immune. This is called the herd immunity level (or population immunity level). As you will see, the concept of herd immunity plays a key role in strategies to control infections, particularly through vaccination programmes. Let S denote the average proportion of the population that is susceptible. This is roughly equal to 1 minus the proportion that is immune, i.e. 1 minus the herd immunity level. The 'roughly' in the previous sentence comes from the fact that those with active infection are not included. However, this is reasonable since they account for only a small proportion of the population. In terms of the SIR model of Unit 2 (Section 5), the average proportion susceptible, S, is the average value of S(t) over a long interval of time, t. There now follows a little logic, from which an important relationship between R0 and S will emerge. The argument is expressed through mathematical equations that are discussed in more detail in Box 4.2 below. In a wholly susceptible population, an infectious individual makes on average R0 effective contacts. Now consider what happens if only a proportion, S, of the population is susceptible. If contacts are made at random (i.e. homogeneous mixing is taking place), a proportion S of the effective contacts are made with susceptible individuals, resulting in R0 × S secondary infections. However, since the infection is in endemic steady state, this number must be equal to one. We thus have the important relationship: R0 × S = 1 for an infection in endemic steady state for which contacts occur at random. The importance of this relationship may be grasped from the fact that it can also be written: In other words, provided that it is possible to determine the proportion susceptible, it is also possible to calculate the basic reproduction number (provided that contacts occur at random). This 'average' steady-state level can be quantified by the average proportion of the population that is immune. This is called the herd immunity level (or population immunity level). As you will see, the concept of herd immunity plays a key role in strategies to control infections, particularly through vaccination programmes. Let S denote the average proportion of the population that is susceptible. This is roughly equal to 1 minus the proportion that is immune, i.e. 1 minus the herd immunity level. The 'roughly' in the previous sentence comes from the fact that those with active infection are not included. However, this is reasonable since they account for only a small proportion of the population. In terms of the SIR model of Unit 2 (Section 5), the average proportion susceptible, S, is the average value of S(t) over a long interval of time, t. There now follows a little logic, from which an important relationship between R0 and S will emerge. The argument is expressed through mathematical equations that are discussed in more detail in Box 4.2 below. In a wholly susceptible population, an infectious individual makes on average R0 effective contacts. Now consider what happens if only a proportion, S, of the population is susceptible. If contacts are made at random (i.e. homogeneous mixing is taking place), a proportion S of the effective contacts are made with susceptible individuals, resulting in R0 × S secondary infections. However, since the infection is in endemic steady state, this number must be equal to one. We thus have the important relationship: R0 × S = 1 for an infection in endemic steady state for which contacts occur at random. The importance of this relationship may be grasped from the fact that it can also be written: In other words, provided that it is possible to determine the proportion susceptible, it is also possible to calculate the basic reproduction number (provided that contacts occur at random).

Host-pathogen dynamics and the wider ecological debate

This discussion of conditions under which the pathogen population will increase or decrease is an example of the wider ecological debate about the nature of stability of population dynamics and conditions for extinction and colonisation. The malaria scenario described in this unit is a situation analogous to a ball balancing on a very fine point (Figure 3.7a). It is mathematically possible for the ball to stay on the peak, but if there is even the slightest change in conditions (e.g. as represented in our example by alterations in daily bite rate), it will roll one way or the other. Thus, it is possible to have a population persisting with one infected host, but it requires all the variables to stay at the same value over a long period of time. This is highly unlikely! Much more likely is the situation in which the pathogen population increases or decreases. This then raises new questions. Will the decline in the pathogen be continued, ultimately resulting in its local extinction? Will the increase in the pathogen population continue until all susceptible hosts have been infected? In reality, it is likely that neither of these will occur. Instead, there may be either a high increase followed by (possibly dramatic) reductions in population size (Figure 3.7b) or reductions to a few infected hosts. Over longer periods of time, this may be seen as cycles of pathogen population abundance or at least fluctuations in pathogen population size. This sort of host-pathogen dynamic is the subject of the next section.

Quantifying pathogen life cycles

To model infectious disease effectively you need to be able to quantify the life cycle of the pathogen. The life cycles of pathogens are very different from those of their hosts. But, as you will be aware from other parts of this module, there is a bewildering diversity of pathogen life cycles. In some cases, longevity may be very short, perhaps less than an hour. In these organisms, e.g. some bacteria, growth may be very rapid. Furthermore, bacteria do not die under normal circumstances unless by accident or infection by phages - they just divide! In contrast, some individual pathogens such as guinea worm may live for months or years in their host. There is also a wide range in the numbers of offspring produced: for example, a single female roundworm (Ascaris lumbricoides) is estimated to produce over 20 million eggs in her lifetime.

Schizongy

Type of mitosis used by apicomplexans to produce many infective organisms.

Pathogen range and abundance

Understanding the interactions between hosts and pathogens is critical for the successful control of disease. In Unit 2 you saw how simple assumptions about the status of host populations can be used to model disease growth or decline. This unit explores the ecology of hosts and their pathogens to help develop your understanding of interactions. Ecology is the study of the interactions of organisms with their non-living or living (other organisms) environment. A major set of questions asked by ecologists is how populations of organisms change in time and space. Changes in population abundance in time and space are known as population dynamics. Once such patterns are described, you can begin to elucidate the underlying processes that give rise to these patterns. This section introduces some of the patterns of range and abundance of pathogens and their hosts. You have already encountered variation in the geographical range relevant to infectious disease, defined as the full geographical extent of the organism under consideration. For example, the geographical range of the three major species of Schistosoma that cause schistosomiasis extends from some coastal parts of South America and the Caribbean to large areas of Africa and small pockets of the Far East. As you saw in Block 1, certain species are generally found in specific locations (see Block 1 Unit 8, Figure 8.6). While it is obvious that a pathogen cannot have a larger range than its host, in fact, in many cases, it has a much smaller range. For example, contrast the geographical range of the malaria pathogen (Figure 3.1) with the distribution of the human hosts, who live in all but the most extreme environments on Earth. Not only is the pathogen's geographical range much smaller than that of the host, it may also be appreciably smaller than that of the vector. For example, there are several species of Anopheles mosquito living in the UK. They share the habitat characteristics of their tropical relatives: one species breeds in brackish waters, one in inland freshwater and one in water-filled tree holes. The difference in range size between pathogen and vector may not always have been so great. Up until the end of the nineteenth century, indigenous malaria (ague) could still be found in the coastal regions of the UK. Thus range areas are themselves dynamic (Figure 3.1). • With reference to Figure 3.1, what are the current limits of the range of the pathogens that cause malaria? • The limits of the pathogen range are roughly between the Tropics of Cancer and Capricorn, i.e. malaria is now primarily a tropical disease with some areas of high risk in northern subtropical areas, e.g. northern India. • How did the range of the malaria pathogens change from 1946 to the present time? • The range contracted, especially from the northern limits in the southern USA, Mediterranean Europe and parts of the former Soviet Union. This contraction in range has left areas of the world with the nuisance of biting mosquitoes or related flies, but no longer the threat of malaria (such as in the coastal regions of the UK). 1.2 Patterns of pathogen abundance It should already be apparent to you that assessing pathogen abundance is extremely difficult. • Are measures of abundance of cases of disease, such as prevalence or incidence, also measures of pathogen abundance? • Yes, indirectly, as they are measures of the number of human hosts affected by a particular pathogen. Numbers of infected human hosts may be closely correlated with abundance of pathogens. However, they may also be grossly misleading: for example, if the numbers of pathogens per human host are highly variable, or if many pathogens live outside the human host such as in the external environment, in vectors or in intermediate hosts. Consider the problem of variation of pathogen abundance in the host. A useful way of expressing the abundance of the pathogen is with respect to its relative abundance in different host individuals; in other words, its distribution among the host individuals. This is especially useful for larger pathogens like roundworms, where the number of individual pathogens per host shows a characteristic pattern (Figure 3.2). • How is the abundance per host represented in Figure 3.2? • As the worm burden, i.e. the number of worms per host. • From the data presented in Figure 3.2, what conclusion can you draw about the distribution of Ascaris lumbricoides in the host population? • The majority of the host population is uninfected with the pathogen, while a few hosts carry the majority of the worms. You have already encountered an example of this distribution in Block 1 with reference to hookworms (see Unit 8, Figure 8.18). This shows that a few individuals carry a heavy worm burden, while many individuals have no or very few worms. This kind of distribution pattern is referred to as a clumped or an overdispersed distribution. It is also referred to as an aggregated distribution. All three names refer to the large number of pathogens in a small number of hosts. 1.2.1 Patterns of aggregated distribution The aggregated distribution or 'pattern of relative abundance' is one of three types of pattern you will see in ecological studies (Box 3.1). To consider the patterns of malaria pathogen abundance, then, as with the geographical range, it is necessary to consider the abundance of the pathogen in the vector. Indeed, for a full appreciation of the abundance of pathogens, you need to consider patterns of abundance within the host, within the intermediate hosts or vectors and, in some cases, within the abiotic (non-living) environment. Fortunately, understanding the dynamics of disease does not usually depend on a complete assessment of pathogen abundance. Box 3.1 Regular, random and aggregated patterns of abundance Organisms can be distributed in various ways in space. This may include the distribution (pattern of abundance) of organisms in different habitats or different parts of the same habitat. With pathogens, we consider the patterns of abundance among hosts. In other studies we might consider, for example, the patterns of abundance of cells on a microscope slide. The three patterns of abundance are random, regular and aggregated. These patterns can be visualised in two ways. First, as the distribution of organisms (such as pathogens) in a grid of habitat patches (such as host organisms), and second, as a histogram of the frequency of habitat patches or hosts occupied by different numbers of organisms (e.g. Figure 3.2). • A regular pattern is characterised by an equal number of pathogens in each host. • In an aggregated pattern, there is a relatively high frequency of hosts with few pathogens and a low frequency of hosts with a high number of pathogens. In this case, the mean number of pathogens per host is not helpful in describing the abundance of a pathogen. • A random pattern lies somewhere between regular and aggregated. Impact of population density on transmission rates An important distinction is between the aggregation of the pathogen in the host and the distribution (and possible aggregation) of the host. The latter can be described in the same way as aggregation of pathogens (Box 3.1) except that now the grid is composed of units of geographical space among which the humans (or other hosts) are divided. This means that some areas may have a much higher human population density than other areas. Population density, defined as numbers of individuals per unit area or, in the case of pathogens, numbers of organism per host, is an important variable in population dynamics. Humans are often highly aggregated in space, especially in urban areas (Figure 3.3). A feature of some of the major historical and contemporary diseases, such as plague, cholera, HIV and TB, is their high prevalence among aggregated host populations due to enhanced rates of transmission.

Adaptable pathogens

Changes generally occur under the selection pressure exerted by two general categories of threat to their existence: • the host's immune response to pathogens entering the body, including the enhanced immunity elicited by the use of vaccines • chemical attacks on the pathogen or vector species. Other factors like climate change and altered agricultural practices also affect pathogen evolution.

Excystment

Process in which a protist cyst releases the vegetative form of the parasite.

PrP proteins

Prion proteins are rogue versions of a group of proteins, known as PrPc (pronounced 'pee-are-pee-see'), found in mammals and birds. These proteins are small cell surface glycoproteins (relative molar mass (Mr) = 33 000-35 000), and while they are expressed on most cells, they are found in greatest abundance in the brain. The normal functions of PrPc in the brain are thought to include: • neuroprotection against oxidative stress and apoptosis • cellular uptake of copper ions • formation and maintenance of synapses • cell adhesion to the extracellular matrix An unusual property of PrPc protein is that molecules with identical amino acid sequences can fold up in different ways, to give proteins with different conformations. One of these conformations (PrPc) represents the regular functional protein, and is usually the only form present, while a group of unusual conformations (PrP scrapie, normally abbreviated to PrPsc) are rogue versions, or prions. The interaction of PrPc with certain phospholipids and RNA has been shown to favour folding into the infectious PrPsc prion conformations Once PrPsc is present in a normal host, it can recruit the host's PrPc protein and catalyse its conversion into the same misfolded conformation as PrPsc, which then acts as a template for yet more conversion of host PrPc. This autocatalytic amplification of PrPsc from the host's normal PrPc protein is essential for TSEs to develop. For instance, mice that lack the PrPc gene do not develop TSE infection after inoculation with prions, whilst inoculation of the same prions into mice with a functional PrPc gene does lead to TSE. (Note that, conventionally, genes are written in italics (as in PrPc), but the protein they encode is written in roman font (PrPc).) In this way, the autocatalytic amplification of PrPsc mimics a conventional infectious agent in its ability to invade and colonise susceptible hosts. This ability of protein to replicate itself without the need for DNA has been called 'protein inheritance'. The subtly different misfolded conformations of PrPsc can be considered individual 'strains' of prion. Recent studies have shown that differences exist in the efficiency of prion strains to replicate in different susceptible species. Thus, over time, those prions that are best able to amplify in a particular species become the dominant strain in that species. In this way, the 'lifeless' prion protein is capable of exhibiting Darwinian evolutionary traits. One might expect a proteinaceous infectious agent to be easily inactivated, but, surprisingly, prion proteins are incredibly robust compared to the normally folded protein. Whereas PrPc proteins are degraded by protease enzymes and are soluble in mild detergents, prion proteins (PrPsc) resist degradation by proteases and are insoluble in detergents. Resistance to the bacterial protease Proteinase K is a feature of PrPsc, and can be used as a marker for TSEs. Prion protein is actually notoriously difficult to inactivate, even using sterilisation treatments that kill, or inactivate, all other known organisms and pathogens. Interestingly, there is no immune response in TSE diseases, even though prion proteins, and certainly the aggregates that they form, are large enough to qualify as antigens. Prion proteins are either derived from the host itself, or resemble the host PrPc protein so closely that they cannot be identified as non-self.

TSE transmission

Scrapie, kuru, BSE and vCJD appear to be infectious diseases. Unlike CJD, vCJD affects predominantly younger people (typically in their late twenties), has a longer duration of illness (median span of 14 months), and is correlated with exposure to BSE, most likely via food. The scrapie agent can be transmitted from ewe to lamb, but this is thought to occur by routes other than in utero. Infection in utero would be vertical transmission, so most ewe-to-lamb infections are probably horizontal. However, a placenta from a ewe with scrapie can be a source of infection on farms. Within flocks, scrapie seems to affect related animals, possibly because of natural selection for the PrPsc conformation best able to amplify in these families. The scrapie agent appears to have been transmitted to mink fed on infected sheep heads, as mink fed in this way went on to develop so-called transmissible mink encephalopathy. However, although humans have eaten scrapie-infected meat - including neural tissues - for at least 200 years, there are no records of any TSEs developing in humans as a result. Although TSEs can all be transmitted experimentally, not all of them are naturally infectious. For example, Gerstmann-Straussler-Scheinker syndrome and fatal familial insomnia are inherited disorders in which mutations exist in the PrPc gene that greatly increase the person's susceptibility to developing TSEs. Also, Creutzfeldt-Jakob disease has several different aetiologies (causes). About 90% of CJD cases are described as sporadic, which means they are isolated, unrelated cases. Sporadic cases affect one in a million of the population per year. Around 10% of CJD cases are inherited and less than 1% are acquired iatrogenically (as a result of medical treatment). Iatrogenic CJD results from the receipt of material derived from a CJD sufferer. This material may be in the form of grafts, e.g. corneal grafts, or hormones, e.g. growth hormone purified from the brains of cadavers. Prion proteins are not destroyed by standard hospital sterilisation procedures, so CJD and other prion diseases could be contracted from fomites such as surgical instruments. This has led to the use of disposable instruments for some procedures. The transmission of kuru is probably the best understood of the TSEs. This disease was found solely in the Fore tribe of New Guinea, and was a major cause of death among the tribe's people in the 1960s, when more than 10% of the population succumbed. It was passed from person to person by cannibalistic practices. In this culture, it was customary to eat deceased relatives, with women and children consuming brain and nervous tissues. It has been speculated that at one such ritual, the deceased had sporadic CJD. Those who consumed tissues from this person contracted kuru, though why such individuals should develop kuru and not CJD itself is unclear. The name kuru means shivering or trembling, which was the main symptom of the disease. Once cannabalism was abandoned, the disease declined and the last kuru victim died in 1998.

Multiple-strain infection

The existence of more than one genetic variant (strain, serotype) of a pathogen in an infected host.

Antigenic variation

The following examples illustrate different ways in which antigenic variation is achieved, and the effect it has on the host's immune response. HIV and influenza viruses Mutation in genes encoding the surface antigens of these dangerous viruses generates multiple new variants at unpredictable rates - a process known as antigenic drift. It enables new variants to evade recognition by memory cells developed during the host's immune responses to previous encounters with earlier virus strains, which would have slightly different surface antigens. For over 30 years this process has defied the efforts of vaccine research laboratories worldwide to produce an effective vaccine against HIV, and it explains the requirement for annual revaccination against influenza with the most recently circulating flu viruses. • Influenza viruses have another method of varying their antigens, which can lead to global pandemics of influenza. These viruses that infect birds or pigs can recombine with genetic elements from human influenza viruses to generate a variant that may never have existed previously. This is termed antigenic shift. Recombinant virus strains can spread rapidly between susceptible contacts, often resulting in high mortality, because there is no pre-existing natural immunity in any human population and none of the currently available vaccines has any protective effect. You may also have noted that the speed of international travel and the extent of trade between countries promote the global distribution of pandemic influenza. Some bacteria can generate so many different variants that they evade attempts to control them with vaccines. For example, in 2011 there were 93 known serotypes of pneumococcal bacteria. Within a year of birth, the nasopharynx of every child has been colonised by at least one serotype, creating a huge natural reservoir of infection. The original pneumococcal vaccines contained at most seven serotypes selected because of their prevalence in particular populations. However, their effectiveness declined within a few years of their introduction because other serotypes began circulating in vaccinated populations and causing pneumonia and bronchitis, meningitis, septicaemia and middle-ear infections, mainly in young children or frail elderly adults. The pneumococcal vaccines introduced in some African countries in 2011-2012 contained 10 to 13 serotypes, but experts believe it is only a matter of time before the huge variation in circulating pneumococcal bacteria begins to undermine the effectiveness of these vaccination programmes. Pneumococcal bacteria are not alone in producing a multiple-strain infection, i.e. the existence of more than one genetic variant (strain, serotype) of a pathogen in an infected host. A review by Balmer and Tanner (2011) reports the unambiguous detection of this phenomenon in 51 different human pathogens and concludes that this pattern of infection is the norm, not the exception. The authors highlight these far-reaching implications: Multiple-strain infections can affect host immune responses and our ability to prevent and treat infection efficiently. Competition and mutualism between strains change pathogen and disease dynamics and promote pathogen evolution. Co-infection enables gene transfer among strains. The causative agent of sleeping sickness, the African trypanosomes (Trypanosoma bruceii), have a unique coating - variable surface glycoprotein (VSG). This not only provides a physical barrier against host antibodies, but also the glycoprotein displays variation over time due to 'gene switching' between about one thousand VSG gene variants. Only one VSG gene is active at a time, but random switching between them changes the surface antigen presented on the parasite's surface, so the immune response is always 'one step behind'. Many parasites achieve a similar outcome by presenting the human host with a succession of life stages, each with their own unique antigens. As an immune response develops against the antigens of the first life stage, the parasite progresses to the next stage of its development, and presents the host with different antigens, and so on. The sequence of life stages that occurs during the development of malaria parasites (Plasmodium species) in the human host are sporozoites, schizonts, gametocytes and early trophozoites. They each have their own unique antigens. Before the immune response directed against one stage can eliminate it, another stage develops with different antigens.

Trophozoite

The vegetative form of a protist parasite, released from a cyst.

Sporogony

Type of meiosis in apicomplexans that forms infective haploid spores.

Wuchereria bancrofti

Vector is Aedes spp (common in Asia and the Pacific) Culex spp. (common in The Americas, Europe, Russia, Saudi Arabia and Japan), Anopheles spp. (common in Africa, especially in rural areas). Definitive hosts are human regional lymph vessels. Causes lymphatic or bancroftian filariasis and elephantiasis. Found in many parts of the tropics and subtropics.

Passive immunity

A type of immunity in which antibodies are transferred from one individual to another. Occurs naturally from mother to fetus, or artificially when antibodies are injected into a non-immune individual.

Intermediate host

One or more organisms parasitised by juvenile stages of a parasite with a complex life cycle.

Surgical site infections

Post-operative infections at the site of a surgical wound; a common type of healthcare-associated infection (HCAI).

Flagellates

Protists that possess flagella during the adult stage of their life cycles.

Chemical resistance in disease vectors

The Malaria Case Study refers to the growing problem of resistance to the most widely used antimalarial drugs against Plasmodium species. The sale of counterfeit antimalarials, including to health facilities, is also a contributing factor. A further challenge to controlling malaria and other vector-borne infectious diseases is the evolution of resistance to insecticides, larvicides and other chemicals used for vector control in the environment. The insect vectors of malaria, yellow fever, African sleeping sickness, dengue fever and several other vector-borne diseases are increasing in some parts of the world, in part due to the gradual failure of once-effective chemical controls. One reason for increasing chemical resistance is the widespread agricultural use of many insecticides and some other vector-control compounds, for example, the chemicals directed against the intermediate hosts (freshwater snails) of Schistosoma parasites. The small size of the available chemical arsenal for use in public health programmes is another major factor, in part due to the banning of some effective insecticides on the basis of their environmental toxicity. For example, at the time of writing (2012), the WHO approves only four groups of chemical compound for indoor residual spraying (IRS) to kill mosquitoes roosting in houses (Figure 6.6), and only one for insecticide-treated nets (ITNs). (For interest, the approved compounds are an organochlorine (DDT, see below), six pyrethroids, three organophosphates and two carbamates.) The rapid replication rates of most vector species means that resistance can begin to emerge in a relatively few generations of repeated exposure to the same chemical. In a trial in Mexican villages, resistance in the local mosquito population rose from zero to 20% within three years of pyrethroid chemical controls being introduced (cited in Read et al., 2009). In 2010, Ethiopia abandoned spraying with DDT (dichlorodiphenyltrichloroethane) because resistance to it had become so widespread in mosquito populations in malarial areas. A problem with all the available chemicals is that they affect mosquitoes at all stages of their development, including the non-infective immature insects. Individuals in this life stage have the greatest potential to evolve resistance under the selection pressure of repeated suboptimal exposure to the same chemical. Read et al (2009) report early research on the development of 'almost evolution proof' insecticides that target only the mature egg-laying females that transmit the parasites to humans when they take a blood meal. • Can you deduce how such an insecticide could combat malaria and reduce the problem of resistance evolving in the mosquitoes? (Hint: you will need to think back to the Malaria Case Study and the maturation cycle of the parasite in the mosquito.) Answer It takes an average of 10-14 days (depending on the Plasmodium species) in the gut of a mature female mosquito for the parasite larvae to develop to the infective stage and migrate to its salivary glands. Killing these older females directly targets the vectors of malaria. The insecticide would be 'almost evolution proof' because mature egg-laying females will not have time to evolve resistance to the insecticide before it kills them, so they cannot pass on resistant characteristics to their offspring.

Environmental challenges to pathogen control

The development of novel chemical agents such as the one described above holds out some hope for progress in tackling at least some infectious diseases in the future. The question remains whether there will be sufficient financial investment for costly research that will largely benefit the populations of the world's poorer nations and may not generate much profit for pharmaceutical companies. This section gives some brief insights into the problems for infection control programmes posed by features of the natural and built environments. We can only touch on a few of the huge range of environmental challenges, but they illustrate the scope of the issues involved. 3.1 Geology and the natural environment The UNICEF and WHO (2012) Joint Monitoring Programme for Water and Sanitation estimates that, in 2010, approximately: • 780 million people (11% of the global population) lacked access to improved sources of drinking water • 2.5 billion people (37% of the world's population) had no access to basic sanitation. There are many reasons for this shaming fact, chief among them the astronomical cost of financing water and sewage treatment plants, piped water, domestic toilets and latrines for the world's population of over seven billion. However, it should not be forgotten that there are geological as well as financial barriers to installing the necessary infrastructure and, of course, the two factors are linked. Digging water reservoirs in deserts, drilling sewage pipes through mountain ranges, and installing water and sewage treatment facilities in the vast areas of river deltas prone to flooding are hugely expensive and may be technically impossible to achieve or maintain. The following example illustrates the unintended consequences of obtaining water from geologically unsafe sources. 3.1.1 Arsenic contamination in Bangladesh Surface water is often contaminated with pathogens that cause high levels of morbidity and mortality from diarrhoeal diseases, especially among young children. From the 1970s onwards, charitable and private organisations funded the drilling of millions of simple tube-wells in Bangladesh and some other low-income countries to pump groundwater to the surface from aquifers where pathogen contamination is very low. In 1997, UNICEF reported that 80% of the Bangladeshi population had access to 'safe' groundwater and that the incidence of diarrhoeal diseases and other waterborne infections was declining. However, evidence was already beginning to emerge of a growing public health emergency in Bangladesh due to arsenic contamination of groundwater from tube-wells (Smith et al., 2000). Arsenic occurs naturally in the Earth's crust and leaches into underground water in some parts of the world at dangerous concentrations. The health effects of drinking arsenic-contaminated water over a prolonged period include extensive skin lesions and cancers of the skin, bladder and lungs. Arsenic poisoning has also affected people in West Bengal, Taiwan and a few other locations where tube-wells were installed in geologically unsuitable locations. The laudable aim of these projects was to increase access to improved sources of water and reduce exposure to water-borne pathogens. Attempts to mitigate the problem of arsenic contamination in groundwater are costly and complex. They include sinking tube-wells to depths below 200 metres where arsenic contamination is minimal, sharing water from wells with low levels of arsenic, chemical removal of arsenic from well water, and using sand filters to reduce contamination by bacteria and other pathogens in surface and rainwater collections. This example illustrates the geological problems facing attempts to install piped water and sanitation in many parts of the developing world. 3.1.2 Forests, agriculture and fishing Factors that increase human exposure to natural reservoirs of pathogens, or make control very difficult, include the following. • Large parts of the Earth's surface are covered with forests and tropical jungles, which form ideal breeding grounds for zoonotic pathogens, such as African trypanosomes and the viruses causing Ebola and Lassa fever, and for common vector species such as tsetse flies and the mosquitoes transmitting malaria and yellow fever. • Animal husbandry has brought humans into close contact with domesticated species that transmit pathogens ranging from the largest bovine tapeworms to the smallest viruses causing Rift Valley haemorrhagic fever. • The burning of biofuels (coal, peat, charcoal, wood from forests, dried animal dung) not only contributes to climate change effects on the distribution of pathogens and their vectors (Section 3.2), but also persistent exposure to atmospheric pollution - particularly indoor smoke from cooking fires - causes inflammation of the lungs and increases susceptibility to respiratory infections (e.g. pneumonia, influenza). • The irrigation of agricultural land and paddy fields creates habitats for waterborne pathogens and water-breeding vectors, such as the snails that transmit Schistosoma parasites and the blackfly larvae that emerge as adults to transmit 'river blindness' (onchocerciasis). • Wading in streams and estuaries to fish, wash or collect water exposes people to naturally occurring pathogens such as Schistosoma and Vibrio cholerae bacteria (Figure 6.8). These vast natural reservoirs of pathogens make it impossible to eradicate the infectious diseases affecting humans living and working in these habitats. You have encountered many other examples in Block 1, so this short section serves as a brief reminder of the key point: the natural world and its adaptation by human populations present insuperable challenges to eradicating many infectious diseases. However, the progress of the guinea worm eradication campaign (Block 3 Unit 1, Section 6) demonstrates that some simple low-cost strategies can sometimes be surprisingly successful. • How does this campaign illustrate the importance of scientific knowledge, despite major limitations in what scientific research has been able to achieve? Answer Despite the lack of any effective drugs to kill the worms in human hosts, or a vaccine to protect against infection, scientific understanding of the life cycle of the guinea worm - including the role of its intermediate aquatic host (cyclops, see Figure 1.19 in Unit 1, Section 6.3) - has proved vital to the progress of the eradication campaign. Low-cost strategies to filter drinking water and protect water sources from contamination by infected people during the period when worms are emerging are proving effective at reducing guinea worms to the brink of global extinction.

Cryptosporidiosis

The protists that cause the human diarrhoeal disease known as cryptosporidiosis belong to the genus Cryptosporidium, and were first recognised as human pathogens in 1976. Since then, they have been identified as the cause of a number of large-scale outbreaks of diarrhoea, the most well-known of which occurred in Milwaukee, Wisconsin, USA in 1993. Cryptosporidia are often isolated from cattle, and their resistant oocysts (Figure 6.11) are shed in the faeces. In Milwaukee the water supply became contaminated with cattle faeces containing such oocysts, and the water purification system failed to remove them. The infectious oocysts are only 4-6 µm in diameter and are too small to be removed by ordinary filtration. To make matters worse, the oocysts are also resistant to chlorine, and in Milwaukee the result of the contamination episode was 400 000 cases of diarrhoea. Unlike Toxoplasma and Plasmodium (see Section 4.3) this apicomplexan does not use an intermediate host or vector. Cryptosporidium parvum, which is frequently found in the small intestines of many animals and birds, is the parasite most often associated with cryptosporidiosis. Ingesting only 10-100 of its oocysts in faecally contaminated food or water is enough to cause the infection, and the oocysts are common in sewage, reservoirs and river water. After ingestion, the oocysts excyst in the small intestine and release haploid infective organisms called sporozoites. These sporozoites invade the epithelial cells of the host's intestine and divide asexually by schizogony (sometimes called merogony) to produce merozoites. Merozoites enter uninfected epithelial cells and undergo further schizogony repeating the cycle (Figure 6.12). Eventually, some meronts become either male or female gamonts. The male gamont divides to give 16 microgametes. These locate female macrogametes with which they fuse to form a zygote, so reproducing sexually. A thick wall forms round the zygote and sporogony takes place, resulting in the formation of a resistant oocyst containing four sporozoites. This mature oocyst is then shed in the faeces to infect other hosts. Some of the oocysts produced in the intestine do not have thick walls and have only a few membranes around the sporozoites, so these are called 'thin-walled oocysts'. These are able to excyst in the gut infecting new cells and beginning the whole cycle again - an example of autoinfection. The symptoms of cryptosporidiosis begin 2-10 days after infection and consist mainly of profuse, watery diarrhoea, although there may be abdominal pain, nausea and fatigue as well. Despite autoinfection as described above in healthy individuals, the disease is self-limiting, and recovery occurs after 2-4 weeks. In the immunocompromised, it can be far more serious, leading to chronic diarrhoea. In AIDS patients, the presence of chronic, severe cryptosporidial diarrhoea often marks the transition to full-blown AIDS. These patients may also suffer from a rare form of cryptosporidial meningitis. Diagnosis is by microscopical detection of oocysts in a stool sample using the acid-fast stain. ELISA and immunofluorescence (explained in Unit 9) can be used to detect antobodies but these do not provide a direct diagnosis. There is no specific treatment for cryptosporidiosis; the patients are usually just rehydrated, but paromomycin can be used if the infection is life-threatening.

Human African Trypanosomiasis (HAT)

There are two types of African sleeping sickness (human African trypanosomiasis, or HAT) caused by different subspecies of Trypanosoma brucei. In the East African savanna, the disease is caused by Trypanosoma brucei rhodesiense (about 10% of all sleeping sickness cases), while in the rainforests of West and Central Africa, Trypanosoma brucei gambiense is responsible (about 90% of all cases). The metacyclic (infective) trypomastigotes are injected into humans with saliva when the host is bitten by infected tsetse flies of the genus Glossina (Figure 6.3). The parasites soon reach the bloodstream where the trypomastigotes develop into two forms: • a long slender form that multiplies by binary fission • a stumpy form that is non-dividing but is the only form that is infective for a biting tsetse fly. The change from slender form to stumpy form is thought to be triggered by parasite cell density in the host blood and lymph. The two trypanosome subspecies responsible for HAT attack the body in different ways. Subspecies T. b. rhodesiense causes inflammation of the lymph nodes and kills the cells in the small blood vessels that supply the heart and brain. This infection develops rapidly and can kill a person within a year. In contrast, subspecies T. b. gambiense is a chronic disease taking around three years to cause death. This chronic phase is often called the second stage of the disease and is when parasites enter the central nervous system (CNS), causing the typical lethargy that gives the disease its name. At this stage small numbers of parasites can be found in the cerebrospinal fluid (CSF). There are also differences in the subspecies' route of transmission. T. b. gambiense is essentially passed from human to human by the tsetse vector, whereas T. b. rhodesinese is passed to humans by tsetse flies from a reservoir in both wild and domestic animals. The tsetse fly is a true vector because the parasites go through several developmental stages in the fly (Figure 6.4). The ingested stumpy trypomastigotes differentiate into long slender procyclic forms in the fly midgut from where they migrate to the salivary glands and transform into the epimastigote form. Multiplication takes place in the salivary glands and then the final transformation to metacyclic (infective) trypomastigotes occurs. The whole process in the fly takes about three weeks to complete. The African trypanosomes have an outer protein coat, the composition of which they can vary, so their antigenic properties change. This leaves the vertebrate host immune system continually one step behind, and rules out any hopes of a vaccine. Diagnosis is by the presence of parasites in a thick blood smear or by detection of specific antibodies against the parasites in host blood serum using an enzyme-linked immunosorbent assay (ELISA) or haemagglutination inhibition assay (HAI). Both of these tests are explained in Unit 9. The CSF may contain parasites and will usually have raised numbers of mononuclear cells and above normal CSF protein levels. African sleeping sickness can be treated with the drugs suramin and pentamidine, or melarsoprol if the CNS is involved. Recently, a combination of eflornithine and nifurtinox has been used to treat this second or chronic stage of the disease, and to successfully overcome melarsopol-resistant strains. Now complete the activity below to find out more about the diseases caused by the trypanosomes discussed above. T.b gambiense, T. b. rhodiesiense Sleeping sickness Glossina spp. (Tsetse fly) Injection of infected insect saliva (classified as a salivaria parasite). Extracellular in blood and later in cerebrospinal fluid (CSF). Slender and stumpy trypomastigote forms in host blood. Trypomastigote and epimastigote in Tsetse fly. Chancre at site of bite. Irregular periods of fever and headaches. None, symptoms are progressive. Lymphadenopathy (infection and swelling of the lymph nodes), also weakness, anorexia and sleep disturbances. Coma and death if untreated. T. b. gambinese: months to years. T. b. rhodesiense: weeks to months. Repeated IgM and IgG antibody responses leading to immune exhaustion. Domestic livestock such as cattle. Wild animals, especially antelopes such as bushbuck and hartbeest for T. b. rhodiesiense. Suramin and pentamidine for acute phase. Eflornithine and melarsoprol if central nervous system involvement. Vector habitat alteration. Traps.

Helminths

Type of worm, some species of which are parasitic. Parasites occur as members of virtually all invertebrate groups, and although some may have devastating effects on their hosts, many hosts tolerate a parasite burden with little obvious effect. Although we concentrate here on some of those parasites that cause debilitating diseases in humans, it is important to recognise the ubiquitous nature of invertebrate parasites, since it has been suggested that over half of all individual animals and plants are infected, at least at some stage of their life cycle. Some of the more important parasitic infections of humans are caused by various worms known generally as helminths. These include the majority of animal parasites found living on or within vertebrates and are represented in two phyla: • Platyhelminthes (flatworms) • Nematoda (roundworms). Although these groups incorporate the 'parasitic worms', in fact many parasites are worm-like (recall from Unit 6 and the Malaria Case Study, the worm-like shape of the invasive stages of the protists Trypanosoma and Plasmodium). • Why do you think that parasites have evolved this worm-like morphological characteristic? • Such a shape is useful for taking up residence in tubular cavities of hosts, such as the gut and blood vessels, and would also favour penetration into or through host tissue. Table 8.1 shows the estimated occurrence of some important helminth parasites that affect humans, and their distribution throughout the world. Table 8.1 Estimate of current human infection by some helminthic parasites Phylum Parasite Location in host Disease Numbers infected (millions) Distribution Platyhelminthes (flukes) Schistosoma spp. blood vessels (around gut or bladder) Schistosomiasis (bilharzia) >240 Africa, Asia, South America Clonorchis sinensis bile duct, gall bladder, liver clonorchiasis (Chinese liver fluke) 30 Far East Fasciolopsis buski intestine fascioliasis 10 Eastern Asia, South West Pacific countries Paragonimus westermani lungs paragonimiasis (human lung fluke) 20 Global Platyhelminthes (tapeworms) Taenia solium adult in gut; larval cysts may infect other tissues, e.g. brain taeniasis cysticercosis 10 Global Echinococcus granulosus cysts in brain, liver, lungs hydatidosis 2.7 Global Diphyllobothrium latum intestine diphyllobothriasis 9 northern USA, Canada, Scandinavia, Balkans, Far East Nematoda (roundworms) Ancylostoma spp. and Necator spp. intestine ancylostomiasis and necatoriasis, respectively 740 southern Europe, Africa, Far East, southern USA, South America Ascaris spp. intestine ascariasis 1000 Global Anisakis simplex larva penetrates through intestine wall anisakiasis up to 0.1 Global Brugia spp., Wucheraria spp. lymphatic system filariasis 120 Asia, South West Pacific countries Onchocerca volvulus eye onchocerciasis (river blindness) 37 central South America, Sub-Saharan Africa Dracunculus medinensis under skin dracunculiasis (guinea worm disease) 0.002 Ethiopia, Ghana, Mali, Sudan • What do you notice about the worldwide distribution of these parasites? • A large number occur in tropical or subtropical areas. • Can you think of reasons for this? • Environmental conditions may favour the continuation of the parasite's life cycle in these areas. Also the helminths tend to be mainly in low- and middle-income countries (LMICs) where poverty, resulting in unhygienic conditions of contaminated soil and water, may favour the survival of parasite stages. Before dealing with the pathobiological effects of helminth parasites and the subsequent impact on humans, you need to examine how the life cycles of flukes, tapeworms and roundworms are beautifully adapted to enhancing transmission from one host to another. This subject is the focus of Section 2.

Loa loa

Vector is Tabanid flies (horse flies, deer flies and mango flies) of the genus Chrysops. Host is human and is found to roam freely in the subcutaneous tissue. Causes loiasis. Found in West and Central Africa. The second nematode species responsible for causing cutaneous filariais is Loa loa, which causes the disease loiasis. This condition is characterised by an allergic type reaction to dead worms or to their metabolic byproducts. Adult worms move freely in the subcutaneous tissues of hosts and can often be seen migrating across the cornea, giving rise to the common name of eye worm. Their life cycle is very similar to that of Onchocerca, with the difference being that Tabanid flies are the vectors, rather than blackflies. Microfilariae of both Onchocerca and Loa loa show periodicity, but the latter show a diurnal pattern with peaks of circulating larvae in the peripheral blood around midday, which once again corresponds to the feeding habits of their vectors.

Lymphatic filariasis

Vectors of W. bancrofti and B. malayi are mosquitoes, with the species in question varying with geographical region. Interestingly they are not limited to one genus, as found with malaria-causing Plasmodium, but several species within four genera are good vectors of disease. In the Americas the mosquito species Culex are the most common carriers, while in Africa it is Anopheles and in Asia and the Pacific it is both Aedes and Mansonia. The parasites spend 10-14 days in the insect vector developing from the microfilarial stage to the infective third larval stage. In the human host it takes from 6-12 months to moult twice more to reach the adult stage. Unlike other nematode species, filarial females do not lay eggs. Instead they shed tiny microfilariae (250-300 μm) which are essentially the first larval stage sheathed in the soft membranous egg shell. Up to 10 000 microfilariae can be shed every day, with the adult worms living 15-20 years. Intriguingly, the microfilariae demonstrate nocturnal periodicity. During the night, when people are sleeping deeply (roughly between the hours of 22:00 and 08:00) the microfilariae move to become concentrated in peripheral blood vessels, whereas during the daytime they retreat to deeper tissues with only a few present in the peripheral blood. • How might this behaviour facilitate transmission? • It is an adaptation to the biting habits of the vector. More vectors bite at night so a higher concentration of microfilariae in the blood at this time means it is more likely that the vector will pick up the parasite and facilitate transmission. Lymphatic filariasis is a chronic and debilitating disease that causes extensive morbidity (illness) but little mortality (death). The most dramatic symptom is caused by the blockage of the lymphatic system by the adult worm, resulting in the condition known as elephantiasis (Figure 8.20) where lower limbs, and often genitals, can swell grotesquely. Enlarged and tender lymph nodes are a signal of the presence of the parasites prior to the appearance of the more severe symptoms. Diagnosis is by observation of microfilariae in a thick blood smear but due to the periodicity shown by the microfilariae it is important that the sample is taken at night. Alternatively, a new antibody-based test that detects parasite antigens is now available, which enables detection of infection within minutes and can be used at any time of the day. Organised strategies against Lymphatic filariasis In 1997 the Global Programme to Eliminate Lymphatic Filariasis (GPELF) was launched by the WHO with the target to eliminate the disease by 2020. The project has two components: 1. stopping the spread of infection by interrupting transmission 2. alleviating the suffering of affected populations (reducing morbidity). A year later the pharmaceutical company GlaxoSmithKline joined forces with the WHO and pledged to donate one of the main vermicidal (worm-killing) drugs, albendazole, free of charge to the effort to combat the disease. Subsequently, a rival company Merck and Co. offered to donate their drug, Ivermectin, in certain geographical areas, and the precedent was then set for other drug companies to join forces and form the Global Alliance to Eliminate Lymphatic Filariasis (GAELF, 2011). The alliance meets every two years. Their strategy involves the annual mass drug administration (MDA) of both albendazole and Ivermectin in areas where infection is endemic. Great strides towards combatting the disease have been made since the organisation was established. 5.2.3 Overview of main filarial worm species The main features of Wucheria, Brugia, Onchocerca, and Loa Loa filarial worms, which are responsible for much of the morbidity due to filariasis, are summarised in Table 8.2. One other important roundworm, which has been the subject of a WHO eradication programme, is Dracunculus medinensis (guinea worm), a large worm that is found under the skin, where it forms blisters. When the blisters burst, larvae are released into water where they enter small crustaceans and mature into infective-stage larvae. These infect new hosts through drinking water, migrating eventually to the skin where they mature into the adult worm.

Population dynamics

Changes in population abundance in time and space.

Passive surveillance system

A surveillance system relying on reporting by clinicians or individual patients.

Compartmental models

A typical human population comprises a large number of different individuals, each of whom has their distinctive characteristics. The first stage in attempting to model such a population for epidemiological purposes is to reduce this diversity to a few key attributes deemed to be of particular relevance to the infection process. For example, for most of the common infections of childhood it makes sense to divide the population at any time into compartments, specifically those who are susceptible to the infection, those who are infected, and those who have recovered from the infection and are immune. For ease of presentation, we shall follow standard conventions and label the three compartments: • S (for 'susceptible') • I ('infected', i.e. with active infection) • R ('recovered', and hence immune to further infection). Accordingly, this model is called the SIR model. The letters also stand for the numbers of individuals in each compartment. To indicate that the numbers might vary over time, represented by the letter t, we write them as S(t) (which, should you wish to read it out loud, is pronounced 'ess of tee'), I(t) and R(t). 5.1 Understanding the SIR model For a given disease, let N denote the total size of the population, which we shall assume is constant, so that the number of births offsets the number of deaths, and the infection has negligible associated mortality. Then: N = S(t) + I(t) + R(t) Note that it is assumed here that, at any point in time, every individual in the population is either susceptible (S), infected (I) or recovered (R). • Which of the three compartments are children born into? Which compartment are most elderly people likely to be in? • According to this model, children are born into the susceptible compartment S. Most elderly people are likely to have experienced infection at some point in the past, and hence will be in the recovered (and immune) compartment R. Fluctuation of compartment numbers As implied by the use of the time variable t, the SIR model is dynamic in that, in general, the numbers in each compartment may fluctuate over time, even though the total population size remains constant.

Epidemiology and modelling infectious diseases

All human diseases involve, to a greater or lesser extent, interactions between individuals and their physical and social environments. For example, some cancers are caused by environmental exposure to sunlight, chemicals or radioactivity; smoking causes lung cancer; and heart disease is related to diet. Infectious diseases are no exception: our social and physical environment, and the ways in which we shape them, have a direct bearing on the spread of infections, and indeed on the emergence of new infections. For example, in the UK in the 1980s and 1990s, contamination of the egg supply with Salmonella enteritidis led to an increase in the incidence of salmonellosis. More recently, in the USA in 2008-2009, an outbreak of Salmonella typhimurium was associated with peanut-butter-based products. And the consumption of meat products infected with BSE led to the emergence of a new variant of Creutzfeldt-Jakob disease (vCJD), as you read in Block 1 Unit 4 (Section 4). War and conflict also invariably result in increased infection rates among civilian populations, for example in the crowded and often insanitary conditions of refugee camps. Lack of access to clean water favours the transmission of waterborne infections such as cholera and other diarrhoeal diseases. Tuberculosis can thrive in conditions of overcrowding and has sometimes been called a disease of poverty. For infectious diseases, as for other types of disease, genetic and environmental factors have a substantial impact on susceptibility to infection, and on the clinical course of the disease. For example, nutritional status influences the clinical course of many infectious diseases of childhood. This goes some way towards explaining the huge disparity in mortality attributable to infectious diseases between low- and middle-income countries (LMICs) and high-income countries (HICs). Such effects are sometimes called risk factors, in that they influence outcomes without necessarily being their sole cause. However, social factors play a uniquely fundamental role in infectious diseases, in that they also determine the transmission of infectious agents. Thus they operate at the level of mechanisms, as well as risk factors. The key concept is that of contact, however defined: without contacts between hosts of some sort, infectious diseases do not spread. For this reason, no microbiological, medical or indeed risk-based epidemiological account of infectious diseases can be deemed in any way complete. To understand infectious diseases, it is necessary to understand the processes that enable the transmission of infection to proceed. As it turns out, a few simple concepts and models go a long way towards explaining how many different infections spread and become established, and hence can help guide strategies for their control and ultimate eradication.

Endemic steady state

An established endemic infection for which infection rates fluctuate around a long-term average that remains constant. An infection is said to be endemic in a given population if the infection is maintained within that population without the need for external inputs. For example, parvovirus B19 infection is endemic in the UK but malaria is not. Each year, there are a few instances of malaria acquired in the UK, but these do not lead to sustained transmission within the UK population. In other words, malaria cannot currently become established in the UK: the value of R0 for malaria in the UK is very much less than 1. This situation could change if, for example, climate change led to conditions in which Anopheles mosquitoes were to flourish. Provided the population is sufficiently large (see Box 4.1 below), endemicity can be maintained by sustained person-to-person transmission, for which the necessary condition is R0 > 1, or by the presence of a suitable animal reservoir. In a large population, infections with R0 > 1 become endemic and coexist in ecological balance with the host population. If the contact rates, transmission probabilities and the size of the host population remain broadly constant over a long period, then the number of infections per unit time will eventually settle down and fluctuate in regular seasonal or epidemic cycles around an average level. Such an infection is said to be in an endemic steady state. For example, infections such as whooping cough, measles, mumps and rubella are endemic infections that were roughly in a steady state in the UK prior to the introduction of mass vaccination. In reality, no infection is ever exactly in a steady state since there are inevitably small fluctuations in the numbers of births and contact patterns. However, the steady-state concept is useful and serves to describe the long-term regularity observed in many time series of infectious diseases. Figure 4.1, for example, shows the incidence rate of chickenpox in France over time. The rate varies seasonally around an average that remains broadly constant: it is reasonable to conclude that varicella zoster infection in France was in a steady state over this period. Consider another example. HIV is endemic both in Europe and in Southern Africa: the transmission of HIV is sustained within both regions, even though the epidemiology of HIV differs. For example, in Sub-Saharan and Southern Africa most transmission is within the heterosexual population, whereas in Europe sustained transmission occurs mainly within male homosexual and intravenous drug-using populations (UNAIDS, 2010). However, in neither region has HIV reached a steady state, owing partly to its long infectious period, often lasting many years.

Stochastic extinction

An infection with R0 > 1 can become extinct if there are insufficient numbers of susceptibles to maintain transmission. This is because the numbers of infectious individuals typically oscillate over time, giving rise to epidemic cycles (epidemic cycles are discussed in more detail in Section 4 of this unit). During a trough, the number of infectious individuals can become zero as a result of chance fluctuations. If this happens, there are no more infectious individuals left to transmit the infection, which then dies out until new infections are imported. This is the phenomenon of stochastic extinction. The likelihood of stochastic extinction occurring increases as the population size gets smaller. Small, isolated island populations provide the ideal setting in which to study this phenomenon. Figure 4.2 shows the pattern of measles epidemics in Iceland prior to mass vaccination. Epidemics are set off by infected visitors to the island; between epidemics, no cases occur.

Epidemiology and modelling

As you have already read in this block, epidemiology is the study of the health of populations. It draws upon knowledge from the fields of anthropology, biology, medicine, sociology and statistics. Its central paradigm is that health and disease have social dimensions, knowledge of which can illuminate the causes, and hence inform the prevention, of ill health in individuals. It is not possible to give a complete account of the epidemiology of infectious diseases here: this is a subject big enough for an entire module. The epidemiology of measles, for example, is totally different from that of HIV. Indeed, the epidemiology of measles in Western Europe is totally different from that in Sub-Saharan Africa, and the same applies to HIV. However, the principles governing the transmission of infectious agents can be treated in a common, and relatively simple, framework, at least with the aid of some simplifying assumptions and models. Given the complexity involved in studying epidemiology, it is inevitable that disease models involve simplification, sometimes oversimplification. However, there are great benefits to be derived from a modelling approach, at both a conceptual and a practical level. Models can help us to focus on those aspects of the process of infection transmission that are in some sense fundamental, in that they help explain some of the main epidemiological features of the infection in a given population. One immediate advantage is that a single modelling framework will often apply to many different infections. For example, a model for measles will have the same structure as a model for other viral infections of childhood such as mumps and rubella, and indeed for bacterial infections such as pertussis. The insights provided by models can also have direct practical spin-offs in suggesting interventions and helping to quantify their likely impact.

Risk and rates

Because epidemiology deals with the health of populations rather than individuals, its methods are statistical. No knowledge of these techniques is required here, other than an acquaintance with the notions of risk and rate. 2.1 Risk In epidemiology, a risk is a probability, that is, a number between 0 and 1, indicating the chance of an event happening. If the risk is zero, the event will never happen. If the risk is 1, it will always happen. You might think, for example, of the risk of becoming infected with the rubella virus during pregnancy, and how a suitable numerical value could be applied to that occurring. Risks are estimated using proportions. For example, imagine that out of 1000 pregnant women, two are infected by the rubella virus. Then an estimate of the risk of rubella virus infection during pregnancy would be: Note that in this calculation no assumption has been made about how many of the 1000 women actually came into contact with the rubella virus. 2.1.1 Risk factors A risk factor is a variable associated with an increased risk of disease or infection. A common use of risks in epidemiology is in computing the probabilities of an outcome event - which could be disease, infection or death - given exposure to one or more potential risk factor. Thus, in that context: For instance, to continue the previous example, having young children might be regarded as a risk factor for rubella in pregnancy, since young children are at high risk of contracting the rubella virus which they might then pass on to their pregnant mother. Much of epidemiology is concerned with identifying risk factors for disease. • Age is a risk factor for measles. Explain why this is the case, and identify the age groups most at risk. • Adults are more likely than children to have developed immunity to measles during previous exposure to the virus. So adults are at lower risk of contracting measles than children. Risk factors vary between infectious diseases. For example, age is a risk factor for both measles and shingles. However, unlike measles, older people are at higher risk of having shingles than children. This is because shingles is caused by the reactivation of varicella zoster virus acquired earlier in life when it produces the disease known as chickenpox. Reactivation results from a decline in immunity levels, and hence occurs more frequently in older people. Note that risk factors are not necessarily causal: for example, being young cannot in any sense be said to cause measles infection. However, some non-causal risk factors can influence outcomes. Identifying such risk factors can help to suggest intervention strategies. For example, in many LMICs, malnutrition and a low level of maternal education are risk factors for infant and child deaths, many of which are caused by infectious diseases. Improving educational opportunities for young mothers, encouraging them to breastfeed their infants for the first 6 months of life and ensuring an equitable distribution of food are thus effective public health interventions to reduce infant and child mortality. It is also important to reflect on the cultural specificity of what might validly be regarded as a 'cause'. A lack of awareness of the cultural context can reduce the effectiveness of prevention strategies, for example against malaria in some parts of Africa. 2.1.2 Evaluation of risk factors Risk factors are evaluated by comparing the risk in individuals exposed to the risk factor with the risk in individuals not exposed to the risk factor. A typical use of this approach is in investigating outbreaks. • An outbreak of salmonellosis has occurred among guests who have all attended a wedding reception. Investigations show that, of 84 guests who report having eaten chicken, 21 were ill, compared with 3 out of the 40 guests who did not eat chicken. Calculate the infection risks in those who ate, and those who did not eat chicken. What might you conclude? What further investigations might you undertake? • The risk in those who ate chicken is 21/84 = 0.25, compared with 3/40 = 0.075 in those who did not. So the risk is over three times higher in those who ate chicken. This might suggest (but does not prove) that contaminated chicken was the cause of the outbreak. Laboratory tests could help confirm this hypothesis, for example if the same salmonella serotype was identified in both the chicken and the infected guests. In the case of numerous infections, such as those caused by measles virus, Bordetella pertussis, rotavirus and hepatitis A virus, becoming infected at some point in life is a near certainty unless an appropriate vaccination programme has been implemented. Thus the lifetime risk of acquiring varicella zoster infection (chickenpox) is close to 1. But this tells us nothing about when the infection is likely to occur. For this we need a different measure, called a rate. 2.2 Rate A rate is a measure of occurrence per unit time. The key difference between a rate and a probability is the 'per unit time'. For example, the lifetime risk of dying is 1 for everybody. But in HICs the mortality rate (or, more precisely, the age-specific mortality rate, calculated for each year of age) is quite low until people reach their seventies, when it rises steeply. A rate is estimated as follows: For example, suppose that over the course of 12 months, 20 colds occur within an office of 100 staff. What is the monthly rate at which colds occur? The relevant exposure here is 'working in the office', so 100 persons are exposed, each for 12 months. The monthly rate is: The symbol '≈' ('approximately equal to') is used here rather than '=' ('equal to') to indicate that the answer has been rounded. Note that, unlike risks, rates have units; in this case, the rate is per month. Also, it is quite possible that some people had more than one cold during the 12 months: the numerator here (i.e. the value displayed in the top half of a fraction) is 'number of events', not 'number of persons experiencing an event'. 2.2.1 Incidence and prevalence measures of rate The rate at which new instances of a disease occur in a population is called the incidence rate, or just incidence. Thus we would say that, in the example of the office population in the previous section, the incidence of common colds was 0.017 per month. • Suppose that among the 350 children attending a primary school, over a period of four weeks, 38 are absent due to measles. Calculate the incidence per week of measles in this population. • The incidence is: As you have read, the proportion of the population with a particular infection at a given time is called the prevalence of infection. For acute infections, which resolve over a short period of days or weeks, there are generally few current cases at any one time (compared with the size of the population), so measures of incidence are more commonly used than measures of prevalence. However, for infections lasting much longer, such as HIV infection, or those with a carrier state such as infection with hepatitis B virus, both incidence and prevalence measures are used. In this unit, we primarily use incidence rates. 2.3 Estimation of risks and rates Estimating risks and rates is perhaps the easiest aspect of epidemiology. The real challenge lies in interpreting them correctly. It is all too easy to jump to conclusions and declare that a risk factor 'causes' the disease, when in fact other, non-causal explanations exist. The fact that two events are associated does not imply that one causes the other. For example, the AIDS epidemic in the USA, the UK and some other western countries began in male homosexual populations, which led some researchers to conclude that AIDS was caused by sex between men. In fact, there is no such causal link, and in most parts of the world AIDS is predominantly transmitted between heterosexual people. Finally, it is worth pointing out that, in the scientific literature (including the epidemiological literature), the distinction between risk (probability of occurrence) and rate (number of occurrences per unit time) is blurred. This is partly due to incoherent terminology: some risks are commonly called rates. For example, the infant mortality 'rate' is in fact the risk (usually expressed per 1000 live births) that a child will die within the first 12 months of life. Another important example is the 'attack rate' of an infection: this is just the risk of infection in a defined group. Generally, such terminological inconsistencies matter little. In this unit they will be pointed out when they arise; otherwise, assume they have been used correctly. 2.4 Odds and risk ratios Odds and risk ratios are a statistical method of assessing the importance of risk factors. Consider the example of malaria, which causes low birth weight, which, in turn, is the greatest risk factor for neonatal and infant mortality (Guyatt and Snow, 2004). Table 2.1 shows data from Malawi, East Africa, where infant mortality rate (IMR) increases with decreasing birth weight. Some of the reduction in birth weight is due to the mothers contracting malaria during pregnancy. The IMR is assessed over the first 12 months following a live birth. Table 2.1 Infant mortality rate (IMR) associated with different birth weight categories Birth weight category/kg IMR per 1000 live births <1.5 0.65 1.5 to 2 0.276 2 to 2.5 0.058 >2.5 0.024 From these data two values can be calculated. The first is the risk of death involved for each of the birth weight categories (as defined in Section 2.1.1) during the first 12 months of life. This value is equivalent to the IMR, in this example: n number of deaths per 1000 infants born alive. The second value is the odds of death in each category. This value is given as the number dead divided by the number alive at the end of the census period. Thus we can see in Table 2.2 that in the birth weight category '1.5 to 2' the risk of death is 276/1000 = 0.276, while the odds of death are 276/724 = 0.381. Table 2.2 Calculation of risk and odds at the end of a census period for the malaria-birth weight example in Table 2.1. You need to complete the missing values* in the cells for Category 1 in the question that follows. Comparison category Birth weight category/kg Number of infants dead Number of infants alive Total number of infants Risk Odds 1 <1.5 650 350 1000 [...]* [...]* 2 1.5 to 2 276 724 1000 0.276 0.381 3 2 to 2.5 58 942 1000 0.058 0.062 4 >2.5 24 976 1000 0.024 0.025 • Calculate the risk and the odds for the lowest birth weight category. • Risk of death is the same as the death rate, i.e. the number of dead divided by the total number born. (The total number of infants is the value of live infants plus those that have died, and happens to be 1000 for each of the birth weight categories in Table 2.2.) Also, remember that the odds are defined as the number of dead infants divided by the number of live infants during the census period. The risk in this scenario is thus 650/1000 = 0.65, and the odds are 650/350 = 1.86. The effect of birth weight on infant mortality can then be assessed by comparing different birth weight categories using two methods. First, the relative risk or risk ratio can be calculated. This is the risk for one birth category divided by the risk for another. In many other epidemiological studies we might be interested in the effect of the treated or exposed group compared with the control or unexposed group. However, in the current example, more comparisons are possible, which makes it more difficult to draw meaningful conclusions. To help resolve this issue we can take birth weight Category 4 (i.e. '>2.5 kg') as the control group because this is considered a healthy 'normal' birth weight, so we can limit the comparisons between categories to three (1 versus 4, 2 versus 4 and 3 versus 4). The second method of comparison is the odds ratio. This is defined in a similar way to the relative risk, but this time it considers the odds of the treated or exposed group divided by the odds for the control or unexposed group. Table 2.3 shows the relative risk and the odds ratio for each of the three comparisons. Table 2.3 Relative risk and odds ratios for comparing birth weight Categories 1-3 with Category 4. You need to complete the missing values* in the cells for Category 2 in the question that follows. Comparison category Birth weight category/kg Relative risk Odds ratio 1 <1.5 27.08 74.4 2 1.5 to 2 [...]* [...]* 3 2 to 2.5 2.42 2.50 • Calculate the relative risk and odds ratios for Category 2 in comparison with Category 4. • Relative risk is 0.276/0.024 = 11.5. Odds ratio is 0.381/0.025 = 15.24. Note that rounding errors have been introduced. If you worked with the raw data, the odds ratio is 15.5. You should see that the risk and the odds (and therefore the ratios) become more similar at lower risk values. However, with higher risks they are quite different. There are various statistical reasons why odds ratios are preferred over relative risk, especially when comparisons are made across many different studies. Regardless of the method of analysis used, the overwhelming conclusion is that low birth weight categories are associated with a much higher level of mortality.

Blood flukes

By any standard, the most important human fluke disease is schistosomiasis, which is caused by infection by blood flukes of a number of species of Schistosoma (schistosomes). Three main species cause disease in humans: • Schistosoma mansoni • S. haematobium • S. japonicum. The disease was originally known as bilharzia, worms having been first discovered at autopsy in a Cairo hospital by the German surgeon Theodor Bilharz in 1851. However, the disease was certainly known many centuries before this. It is possible that hieroglyphics (Figure 8.5) from around 1500 BC depict a disease which was described in medical papyri from those times as 'âaâ' - this was thought to mean 'haematuria' (urinary bleeding; see Colley, 1996), which is depicted by the bottom-right symbol in the hieroglyphic. S. haematobium, is found in the blood vessels around the bladder, and causes this symptom. Subsequently, calcified eggs of blood flukes have been found in mummies and, with molecular techniques, schistosome antigens have been detected in a mummy from 3200 BC! (Miller et al., 1992). Distribution of blood flukes In terms of occurrences of the three main species of blood fluke identified above, urinary schistosomiasis, caused by S. haematobium, is particularly prevalent in the Nile river valley but is also found in many other areas of Africa as well as in parts of the Middle East. S. mansoni - which inhabits the veins around the gut causing intestinal schistosomiasis - has a similar distribution in Africa and the Middle East but is also found along the eastern part of central South America and the Caribbean. Finally, S. japonicum, causing Katayama fever, is restricted to the Far East but is no longer found in Japan as it has been eradicated from there. This distribution pattern of the three major schistosome species is shown in Figure 8.6. Two less important species of blood fluke, both inhabiting blood vessels of the small intestine are S. mekongi (prevalent in South West Asia) and S. intercalatum (commonly occurring in West Africa). The main reservoir of infection for S. mansoni and S. haematobium is man, but for S. japonicum various mammals including dogs, cats, pigs, goats, water buffalo and rodents all serve as sources of infection, meaning that this disease is a zoonosis. The reservoir for infection for S. mekongi is dogs. S. haematobium and S. mansoni - The following section focuses on the two main human parasitic schistosome worms, looking in more detail at their life cycles, the factors that favour their transmission to humans, and the host responses to infection. As mentioned in Section 2.1, there are three requirements for the schistosome life cycle: water, molluscs and vertebrate (human) hosts. The life cycles of S. haematobium and S. mansoni are shown in more detail in Figure 8.7.

Changes in numbers of bacterial pathogen

Consider, first, a bacterium that divides by binary fission. This means that each bacterium produces two new bacteria, which then divide to produce four bacteria, and so on (Figure 3.12). This doubling method of reproduction gives bacteria a characteristic population dynamic, at least in the early stages of increase when there is nothing to limit their increase. For example, E. coli may divide once every 20 minutes (referred to as the doubling time) although for other species it may be much longer. In contrast, the doubling time for lymphocytes (the host's cells mediating the adaptive immune response) is about 6 hours. What are the quantitative implications of this imbalance in doubling time? • Assume a bacterium has a doubling time of 30 minutes and that you start with one individual organism. How many bacteria are present after 6 hours, i.e. the time taken for a lymphocyte to divide? • After 30 minutes there will be two individual bacteria, after one hour there will be four, after 90 minutes there will be eight, and so on. After 6 hours there will be a staggering 4096 bacteria (but only two lymphocytes to help combat them). There are two ways of calculating the number of bacteria after 6 hours. You can either keep doubling the numbers after every 30 minutes, i.e. repeat the calculation 12 times in total. This is rather time-consuming! Or, to speed up the process, you can note that after one 30-minute period there are 21 (2 to the power 1, i.e. 2) individuals, after two 30-minute periods there are 22 (2 to the power 2, i.e. 4), and so on. So, after 12 30-minute periods (6 hours), there are 212 bacteria, which is 4096. Thus, any delay in detection of the bacteria by the host will result in a considerable numerical disadvantage to the host. • Do you imagine that bacteria could maintain that rate of population increase? • No, otherwise the world would be covered with the species of bacterium with the shortest doubling time! Competition between individuals (and between species) will limit availability of resources and will result in the slowing of the population increase. Eventually, interactions with the host defence system will also lead to a slowing down of the increase and even a reduction in numbers of the pathogen.

Effective contact

Contact between two individuals, A and B, such that if A were infective and B susceptible, A would infect B.

Aggregated distribution

The pattern of distribution where individuals are clumped in space or where pathogens are clumped with respect to their host, i.e. a few hosts contain most of the pathogens.

Host-pathogen community studies and disease prediction

Finally in this section, consideration of the host-pathogen community is vital for understanding the reasons for changes in incidence of disease. For example, construction of the Aswan High Dam in Egypt resulted in huge increases in the prevalence of Schistosoma mansoni, from 5% in 1968 to 77% in 1993. This was largely due to an increase in the snail intermediate hosts that benefited from the increased habitat associated with the dam and irrigation channels. Deforestation in Africa has also favoured the Anopheles gambiae mosquito, by increasing its habitat. Understanding the host-pathogen community is also vital for understanding emerging infectious diseases. For example, although Ebola virus disease is a zoonosis, the animals that carry it were unknown for about 30 years. In 2005, the virus was discovered occurring naturally in three species of fruit bat, although it may be present in other species.

Flukes

Flukes develop through five morphological stages during their life cycle that take them from their egg to full maturity, where they can propagate further eggs. To begin, a small swimming ciliated larva, just a few micrometres in size, is released from the egg. Each is called a miracidium (see Figure 8.2a) and they search for a specific mollusc - usually a snail - to which they are chemically attracted, before their energy supply is exhausted (usually within 24 hours). If they successfully locate such a host they penetrate into the soft tissues of the mollusc, where they develop into a sac-like sporocyst larva (Figure 8.2b). In some species of fluke, e.g. Clonorchis sinensis, the eggs embryonate but do not hatch until ingested by the molluscan host. The miracidium then sheds its ciliated coat and transforms into a sporocyst larva. Each sporocyst develops more sporocysts and, within each of these, many individuals of a third type of larva may develop asexually. These are called redia larvae. Each redia larva (or sporocyst if no redia larva is produced) produces asexually either another redia larva or hundreds of cercaria larvae (Figure 8.2c), which are shed from the mollusc. These are sperm-like in appearance and swim by lashing their tails. Many exhibit responses to environmental cues such as geotaxis (gravitationally directed movement - usually upwards motion, or towards a vibration in the water) and phototropism (light directed movement - many move up towards light, although schistosomes seek shade) as well as responding to chemical stimuli. If the cercaria larvae find a specific second intermediate host they penetrate it, lose their tails, and encyst, where they are known as metacercariae (Figure 8.2d). This is the fifth larval type of this strange life cycle. Finally, if the intermediate host that contains the metacercariae is eaten by its predator, which is the definitive host, the metacercaria excysts and develops into an adult fluke with two attachment suckers; one around the mouth and the other on the ventral surface (underside). Of the important human flukes, adult specimens of Fasciolopsis buski remain in the intestine, whereas Clonorchis sinensis migrate into the bile ducts and liver and Paragonimus westermani undergo a migration penetrating the intestinal wall, the diaphragm and the lung to finish in the lung tissue. All use their two suckers to adhere to the walls of the vessels. The schistosomes or blood flukes are exceptions in most aspects of the life cycle, and these are examined in more detail in Section 3.

Homogenous mixing

For many modelling exercises a simplifying assumption is made that contacts occur at random in the population. This is known as homogeneous mixing. The homogeneous mixing assumption is usually an oversimplification: in reality, contact rates typically vary with age, social status, location, gender, and so on. However, assuming homogeneous mixing greatly simplifies technical details and thus brings out the relationships between the various epidemiological parameters. Furthermore, all the main concepts illustrated in the simplified setting of homogeneous mixing also apply in the more general, and realistic, case of heterogeneous mixing. Later you will briefly consider the implications of relaxing this and other simplifying assumptions. Using the homogeneous mixing assumption, a simple relationship can be derived between the effective contact rate, β, and the basic reproduction number, R0. Recall that R0 is the average number of secondary infections resulting from contact with a single typical infective introduced in a totally susceptible population. Assuming homogeneous mixing means that there is no need to worry about how to define what constitutes a 'typical' infective: all infectives are equivalent from the point of view of transmission. Suppose that effective contacts occur randomly at a rate β in this population, and that the average duration of the infectious period is D. Then the total number of effective contacts made by one infectious individual is β × D. But if the entire population is susceptible, each one of these effective contacts results in an infection (by definition of what an effective contact is). Hence the number of secondary infections is also equal to β × D, and we obtain the relationship: R0 = β × D

The basic reproduction number

In addition to risks and rates, another important concept in epidemiology and modelling is the basic reproduction number (R0) of an infection ('R0' is pronounced 'are-nought'). This measure has such a central role in infectious disease epidemiology that you need to know about it early in your studies of the subject. Imagine a population that is totally susceptible to a given infection, i.e. one in which nobody is immune to the infection. Into this population, introduce a single typical infectious individual known as an infective. That individual will make contacts with others and hence will transmit the infection to some number of other persons during the period in which he or she is infectious; these are called secondary infections. The basic reproduction number of the infection is the average number of secondary infections that result from an infective. This definition is represented in Figure 2.1. The basic reproduction number is neither a risk nor a rate: it is just a number. It can take any positive value (or zero). Note the rather theoretical character of its definition: in the case of measles, for instance, there is probably no population on Earth that is totally susceptible. Yet, as it turns out, and as you will see later in this unit, it is possible to estimate R0 for measles, and for other infectious agents. For uncommon infections, on the other hand, R0 may sometimes be calculated directly. • For example, suppose that, in an outbreak of Marburg fever (a viral haemorrhagic fever) in a large susceptible community, a total of 6 people were directly infected by the first case before control measures were introduced. In a second outbreak of Marburg fever in a different but similar community, a total of 9 people were infected by the first case. Estimate the R0 for Marburg fever. • R0 is the average number directly infected by one case in a susceptible population. In the first outbreak, 6 people were infected by 1 initial case. In the second outbreak, 9 people were infected by 1 initial case. So, overall, 6 + 9 = 15 people were infected by 2 initial cases. So the average number directly infected per initial case is (6 + 9)/2 = 7.5. Thus an estimate of R0 for Marburg fever is 7.5. This is only a rough estimate of course, since it is based on only two outbreaks. 3.1 The central role of R0 The basic reproduction number (R0) is important because it encapsulates the relationship between an infection and its physical and social environment. The number of secondary infections depends on the ability of the infectious organism to survive outside the host and to migrate from one host to the next, which in turn is contingent on biological and environmental factors. It depends on the infection-host interaction through, for instance, the duration of the infectious period. It is also affected by the frequency and type of contacts that take place within the population, which vary according to the environmental, social and cultural context (see Box 2.2).

The average of infection

In the previous section, the problem of estimating R0 was reduced to that of estimating the proportion susceptible, S, assuming homogeneous mixing. This section simplifies the problem even further, and shows how S, and hence R0, can sometimes be calculated from an even more straightforward quantity, the average age at infection (A). The advantage of this is that the average age at infection can be calculated directly from case reports. The average age at infection is, as its name indicates, the average age at which individuals in the population become infected. This important parameter is related to the force of infection, λ, by: To understand this equation, think of A as the average time spent in the susceptible compartment. • The rate at which susceptibles are infected is then 1/A • thus, λ = 1/A • and so A = 1/λ. Note that a similar argument was used at the end of Unit 2 Section 8 of this block, to obtain the transition rate, δ, in the SIR model. 2.1 A and control strategies The average age at infection (A) provides a direct way of comparing forces of infection in different subpopulations, or between different infections in the same population. For example, if it is known that the average age at infection of measles is, say, 4 years and that of rubella is, say, 6 years, then you can infer immediately that, on average, the force of infection for measles is higher than that for rubella. The average age at infection is also critical in the design of control strategies, as you saw in Video 4.1. Thus, for a vaccination programme to be effective, it must be targeted at individuals below the average age at infection. For example, if the average age at infection of measles in a given population is 3 years, then relying solely on vaccination at school entry at ages 4-5 years will only have a marginal impact on the transmission of measles virus (although, of course, early vaccination followed by a booster at school entry makes sense). • In a given population, the average age at infection by hepatitis A virus is 22 years. It is proposed to control hepatitis A virus by the mass vaccination of teenagers. Is this likely to work? • A mass vaccination programme for teenagers is a reasonable course of action as it would be targeted at individuals below the average age at infection and who may be within the mandatory school age. 2.2 A and herd immunity Since the force of infection and the average age at infection are related to the endemic steady state, you might expect them to be related also to the herd immunity level, as indeed they are. In order to elucidate the relationship, however, it is necessary to be more specific about the age distribution in the population. Age distributions and mortality rates vary greatly between countries. To help your understanding, it is useful to consider two contrasting hypothetical scenarios that are illustrated in Figure 4.3. Figure 4.3 Rectangular and exponential age distributions with the same life expectancy, L = 75 years View description Scenario 1 - rectangular age distribution Everybody lives to the same age, L, and then dies (in Figure 4.3, deaths occur at 75 years). In this scenario, the age distribution is said to be rectangular: the mortality rate is zero up to age L, and then everybody dies. Scenario 2 - exponential age distribution The mortality rate of the population is the same at each age. This gives rise to an exponential age distribution, with fewer older people than younger people. The average age in the population is L. In both scenarios, L is also known as the life expectancy (or expectation of life) at birth. Both scenarios are unrealistic, and in most cases the true situation will lie somewhere in between them. High-income countries with low infant mortality will lie closer to scenario 1, whereas low- and middle-income countries with high infant mortality will be closer to scenario 2. The point is, however, that for childhood infections, the values of R0 obtained under the two scenarios are quite similar. In other words, the precise details of the age distribution are not critical. Suppose that the age distribution is rectangular, so that everybody lives to age L, the life expectancy, and dies. If the average age at infection is A, then on average individuals aged less than A are susceptible, and those aged more than A are either immune or infectious, as shown in Figure 4.4. Figure 4.4 Relationship between S, A and L in a population with a rectangular age structure View description Thus the proportion of the population who are susceptible is: Thus, in such a population: For a population with an exponential age structure, the argument is a little more complicated and will be omitted here. However, under the exponential age distribution assumption of scenario 2, it turns out that: This differs by only 1 from the value obtained under the assumption of a rectangular age distribution. In other words, provided you know the average age at infection and the life expectancy, you can estimate the basic reproduction number, and need not worry unduly about the age distribution of the population. • In a country with a life expectancy at birth of 60 years, the average age at measles infection is 3 years. Calculate the basic reproduction number assuming (a) a rectangular age distribution and (b) an exponential age distribution. Comment on the impact of mortality on the value of R0. • (a) Under the rectangular age distribution assumption, the basic reproduction number is: (b) Under the exponential age distribution assumption, it is: The two very different scenarios produce roughly the same result. We can conclude that the value of R0 is about 20 in this population.

Laboratory reports

Instances of infection or disease in individuals confirmed by laboratory tests.

Hookworms

Human hookworms have a direct life cycle and include two species, Ancylostoma duodenale and Necator americanus, which are worldwide in distribution in areas with moist, warm climates. Both species occur in Africa, Asia and the Americas, but only Ancylostoma is found in the Middle East and Southern Europe. Hence, A. duodenale is often called the Old World hookworm and N. americanus the New World hookworm. The adult worms live attached, by means of ferocious-looking mouth capsules (Figure 8.17), to the intestinal mucosa, from which they suck blood. This results in chronic blood loss and gradual depletion of the body's iron stores, leading to iron-deficiency anaemia. Intestinal nematodes are estimated to infect one billion people worldwide (Brooker, 2010). Although direct mortality is relatively low (about 7000 deaths per year), the exhaustion and vulnerability to infection resulting from hookworm-related anaemia significantly reduces the lifespan, particularly of children. The hookworm life cycle follows the pattern in Figure 8.1c. Eggs are passed in faeces to develop and hatch in moist, warm soil to a first stage larva. After about 10 days in favourable conditions and two moults the third stage infective larva is reached. These larvae can survive in the environment for about 4 weeks, but on contact with human skin they penetrate into a blood vessel and are carried via the heart to the lungs, where they emerge into the pulmonary cavity. They ascend the bronchi by the mucociliary escalator and are swallowed to eventually end up in the intestine. Two moults take place once the larvae reach the host's intestine, resulting in adult male and female worms. Adults usually survive for 1-2 years inside the intestine, the female passing thousands of eggs in the faeces every day. After this time they are eliminated from the body. An interesting feature of hookworm infection is the pattern of infectivity exhibited in susceptible communities. • What is the most striking feature of the distribution of hookworms among infected individuals? • In both species, a large proportion of the worms present in a host population are carried by a small number of 'infested' individuals. For instance, for A. duodenale two persons carry about 30% of all the worms; in N. americanus four persons carry about 38% of all worms found. • Can you think of any reason for this 'clumped' (also referred to as 'overdispersed') distribution? • Probably host genetic, behavioural, or social factors. Variations in immunity could be an explanation for this overdispersal, but Schad and Anderson (1985) have suggested that heavily infected individuals are predisposed by some undefined genetic, ecological, behavioural or social factors. In fact, human hookworms survive an extremely immunologically hostile environment, given that they are constantly taking in blood containing host antibodies.

Roundworms

Roundworms belong to the phylum Nematoda, which is probably one of the most abundant groups of organisms on Earth. They include many free-living members and, as parasites, they invade almost every kind of animal and many plants. Unlike other helminths, roundworms may penetrate a variety of body tissues, and they have a long history as human parasites. This section looks in closer detail at examples from different groups of these organisms. By comparison with flukes and tapeworms, roundworm life cycles are much more straightforward (Figure 8.4), usually involving both free-living and parasitic stages. The fertilised egg passes out of the host in the faeces, hatches, and progresses through two or three moults (shedding of the outer layers) in the faeces or soil. During this process it becomes an infective stage larva (second-or third-stage larva, depending on the species) that is either ingested or penetrates host tissues and matures into the adult sexual forms. The males are generally much smaller than the females. In some species the first two moults occur in the egg with the third larval stage released from the egg while in other species the egg is ingested and the larva is released in the intestine of the host. Although many eggs are produced by the adult roundworm, there are two important adaptive features that explain why roundworms have not evolved asexual stages. First, the eggs are extremely resistant and can survive in the environment for a considerable length of time. For example, the eggs of the dog roundworm, Toxocara canis, can remain viable in soils in parks and children's playgrounds for a number of years. If accidentally ingested by a child, the egg hatches and the larva migrates to various body tissues. It has been known occasionally to lodge behind the eye where it has encysted and caused a tumour to develop. Secondly, roundworms, including their larvae, have a thick cuticle which again assists survival until a new host may be found. Some roundworms (filarial worms, see Section 5.2) have dispensed with free-living stages in the external environment and have evolved to use a vector or, in one exceptional case (Trichinella, see Section 5.3), are transmitted by eating larvae-contaminated food. With a few exceptions, most adult parasites - including those that infect humans - do not kill their hosts. However, there is accumulating evidence that the parasite can affect the behaviour, especially of intermediate hosts, to enforce transmission. The lancet fluke, Dicrocoelium dendriticum, for example, is a liver fluke found in sheep and cattle worldwide. Its second intermediate host is an ant into which the cercariae penetrate. Within the ant, metacercarial cysts normally occur in the body cavity, but at least one cercaria migrates to, and encysts in, the brain of the ant. Once in the ant's brain the cyst interferes with the neural pathway that controls jaw function, which normally operates as temperature falls, to open the jaw and so allow the ant to cease feeding and retreat into the soil. The result of neural disruption by the encysted cercaria is that the ant remains firmly clamped by its mouth onto a blade of grass, just at a time when sheep and cattle graze extensively, thus ensuring transfer of the metacercaria stage to the definitive host.

Average age at infection (A)

The average age at which individuals in a population acquire infection.

Inter-epidemic period

The average interval between successive epidemics of an infection in an endemic steady state.

Basic reproduction number (R0)

The average number of individuals directly infected by a single typical infective if the population were totally susceptible.

Net reproduction number (Rt)

The average number of secondary infections that result from each infective at time t, taking into account incomplete host susceptibility to infection and any control measures that are in place. If Rt is greater than 1 at time t, the epidemic will (in general) be increasing.

Effect of transmission rates

The average steady-state level reached by an infection may be interpreted in terms of average transmission rates. If an infection has reached an endemic steady state, each infectious individual will infect, on average, exactly one other individual. Clearly, the numbers actually infected depend on the epidemic cycle. But the average number, that is the number of secondary infections averaged over several epidemic cycles, must be equal to one. To see why this must be so, suppose that, on average, an infectious individual infects two other people. Then each of these infects, on average, two people as well, making four newly infected people in total. Each of these people infects two others, making eight, and so on. The implication is that the average incidence must rise exponentially over time, but this contradicts the fact that the incidence fluctuates around a constant average level, since the infection is endemic and in a steady state. A similar argument applies if each infectious individual infects, on average, less than one individual (so that, for instance, two infectious individuals infect, on average, only one other). In this situation, the average incidence should decline, again in contradiction with the observed cyclical steady state.

Serological surveys

The data from surveys of population immunity provide one way of circumventing such difficulties. In such a survey, individuals are tested to establish their immune status. As long as that the survey is representative, it can give a profile of population immunity, which is not biased by non-ascertainment of subclinical infections (although it does rely on the accuracy of the test). Perhaps one of the oldest of such techniques is the intra-dermal tuberculin sensitivity test for tuberculosis (the Heaf test). Today, the most commonly used surveys are serological surveys, usually tested by enzyme-linked immunosorbent assay (ELISA; see Block 2 Unit 9, Section 2.2). Data from a survey of the prevalence of antibodies against the mumps virus in people of different ages in the UK (prior to the introduction of MMR vaccine) are shown in Figure 2.6. It should be mentioned, however, that for some infectious agents such as Bordetella pertussis, the bacterial cause of whooping cough, there is as yet no agreed serological correlate of immunity. In this case, population surveys of antibody prevalence are difficult to interpret in terms of population immunity. • For mumps, the presence of IgG antibodies indicates past infection. In Figure 2.6, the proportion of individuals with IgG antibody to mumps increases with age until about 15 years of age, then reaches a plateau. What could account for this? • The duration of exposure to infection increases with age. Hence the risk of past infection increases with age. Correspondingly, the proportion of individuals with antibodies to mumps increases with age. After about 15 years of age, virtually everyone in this population has been exposed to the mumps virus, so the curve reaches a plateau close to 100%. A survey of antibody levels in a sample of individuals, typically used to estimate the level of herd immunity in the population from which the individuals are drawn.

SIR models

The data reveal large fluctuations in the numbers of reports of measles cases over time, prior to the introduction of measles vaccination in 1968. These recurrent epidemics are typical of many endemic infections with short infectious periods. As you will see in Unit 4, the epidemic waves can be explained by variations in the numbers of susceptibles over time, S(t). • The measles rash typically appears about 10-14 days after infection, but only a proportion of measles infections are reported. Deduce what a graph of I(t) against time, t, might look like. What should a graph of S(t) against t look like? • Figure 2.8 shows an example of what plots of I(t) and S(t) against time might look like for a study of measles in a population. The clinical signs of measles appear soon (on the scale of the graph) after infection, so I(t) mirrors the fluctuations in notified cases, and look like a sequence of sharp spikes. During an epidemic, the number of susceptibles S(t) falls rapidly. Once the epidemic is over, the number of susceptibles gradually builds up again as they are replenished by births. Thus S(t) might be expected to have a saw-tooth profile, with gradual build-ups followed by sudden drops corresponding to epidemics. Individual dynamic aspect of the SIR model The SIR model is dynamic in a second sense (that is, other than in the sense that S(t), I(t) and R(t) vary over time): individuals are born susceptible, then might acquire infection, from which they then recover. Thus each individual typically progresses over time from susceptible, to infected, to recovered. This individual dynamic aspect is often represented by means of a flow diagram in which boxes represent the susceptible, infected and recovered compartments, and arrows between them represent the transitions between the different compartments, as shown in Figure 2.9. For simplicity, the boxes are identified as S, I and R, without their time dependence. In a full specification of the model, it is also necessary to specify transition rates, which quantify how many individuals are transferred between compartments per unit time. These transition rates are described in subsequent sections. For now, we will concentrate on the structure of the model, namely the compartments and their interrelationships. 5.2 Adapting the SIR model The SIR model is a very useful and versatile basic model, which can be used to describe many infections that confer long-lasting immunity, such as measles, mumps and rubella. It can, of course, be elaborated in many ways. For example, in many infections, individuals go through a latent period after they are infected, before becoming infectious. This observation can be accommodated in the model by including an additional compartment, E, denoting the number exposed to infection and infected but not yet infectious; the I box now represents the infectious individuals (Figure 2.10). Figure 2.10 Flow diagram for the SEIR model View description • Develop compartmental models with the following features, and draw their flow diagrams. a. For many infections, such as measles, babies are not born susceptible but with maternal antibodies that protect them for a few months after birth. Introduce a new compartment, M, to represent babies with maternally derived immunity. b. For diseases such as hepatitis B, acute infection can lead either to recovery or to a lifelong carrier state. Introduce a new carrier compartment, C, distinct from the recovered class. c. Some infections, for example those responsible for the common cold, do not confer any long-lasting immunity. Such infections do not possess a recovered state, R; infectious individuals become susceptible again after infection. • Suitable flow diagrams are shown in Figure 2.11. For (c), the model is often referred to as 'SIS', for 'Susceptible-Infected-Susceptible'. Figure 2.11 Flow diagrams for three compartmental models: (a) SIR model with maternally protected class M; (b) SIR model with carrier state C; (c) SIS model

Risk

The probability that an event occurs in a defined population.

Calculation of effective contact rates

The factors influencing the effective contact rates can themselves be explored. Suppose that the effective contact rate for a given infection in a given population is denoted by the Greek letter β (beta), measured in number of contacts per unit time. This may be expressed as the total contact rate, γ (gamma), multiplied by the risk of infection from contact with an infective, p. The quantity p is also called the transmission risk. Thus: β = γ × p The total contact rate, γ, will generally be greater than the effective contact rate, β, since not all contacts between an infective and a susceptible individual result in the transmission of infection. The point of introducing this further complication is to formalise the fact that the effective contact rate depends not just on the social contact patterns, represented by γ, but also on the specific contexts in which contacts occur and the biology of the infective organism, both of which influence the transmission risk, p. For example, it is known that the presence of a concurrent sexually transmitted infection can substantially increase the per-contact infection risk of HIV (and hence the value of p). Hence one way to reduce HIV transmission is to treat other sexually transmitted diseases, and thus reduce the value of p. Accurate measurement of contact rates Measuring contact rates accurately is extremely difficult in general, whatever the route of transmission, not least because contact rates typically vary between individuals and groups, and can seldom be documented accurately. For sexually transmitted infections, large-scale population surveys of sexual behaviour have been conducted to estimate the total contact rate, γ (see Box 2.2). • Researchers have also attempted to measure the total contact rate, γ, relevant to infections transmitted by airborne droplets, like measles and whooping cough, by using the total number of conversations as a measure of the total number of contacts. Is this approach likely to work? • The problem is that other relevant contacts, such as those made with fellow passengers on a bus, colleagues at work or other cinema-goers, will not be documented. A 'relevant contact' here could simply involve breathing in infected droplets from a person sneezing (Figure 2.13). Nevertheless, the method might perhaps provide information on the relative sizes of the contact rates in different age groups. 6.5 Ethical position Ethical considerations generally preclude direct experimentation to establish per-contact infection risks, since they would require individuals to be knowingly exposed to infectious agents. However, there are some notable exceptions, such as the experiments conducted by the Common Cold Unit in the UK (see Box 2.5). Transmission risk, p, can nevertheless be estimated in certain circumstances when exposures to infection are documented, as with needlestick injuries (puncturing the skin with the needle of a syringe) among healthworkers, involving contaminated blood products.

Geographical range

The full geographical extent of a species.

Evolution of virulence

The mechanisms that determine the virulence of a particular pathogen, and the evolutionary factors that determine how virulent it is, are topics of considerable interest. • Recall the meaning of virulence. Answer In Block 1 Unit 1, virulence is defined in terms of the ability of a pathogen to establish an infection. Later, virulence is described as a measure of the ease with which an organism is able to cause damage and disease in host tissues (Block 1 Unit 5). Thus, the term 'virulence' is being used to cover both the ease with which an infection is established and the degree of damage. Elsewhere, the former definition of virulence is sometimes replaced by the term 'infectivity', which refers to the capacity of a pathogen to infect new hosts, rather than the harm that it does to its host. 2.1 Reproductive rate and virulence Once the pathogen has been transmitted, a major factor determining the virulence of a pathogen is its reproductive rate, as it is by reproducing in host cells that pathogens establish infection and cause damage to host tissues. This is an important point about the pathogen-host relationship and is crucial for understanding the evolution of virulence. Unlike predators, pathogens are generally not specifically adapted to kill their hosts, but may be adapted to reproduce rapidly. To understand the evolution of virulence, therefore, we need to consider how natural selection acts, not on virulence itself, but on characteristics such as the reproductive rate of a pathogen. This raises an important point about the analysis of evolutionary processes, which was emphasised by the American evolutionary biologist Stephen Jay Gould (1941-2001). He believed that just because a particular character of an organism, such as the virulence of a pathogen, is of interest to us, it does not follow that it is a character that has been subjected directly to natural selection. It may be either a combination of traits or the result of 'coincidental evolution' resulting from natural selection acting on another character. For example, bacteria such as Clostridium botulinum, which causes botulism, and C. tetani, which causes tetanus, produce powerful toxins that attack the nervous system, making them very dangerous to humans. Neither pathogen is transmitted from human to human; both are better regarded as soil-living bacteria that can infect humans. (Neonatal tetanus is a major global problem, causing about 200 000 deaths per year in low-income countries.) It is thus likely that their toxins have evolved in the context of their normal, soil- (or sausage!) living existence and that their pathogenicity in humans is coincidental. All other things being equal, natural selection favours those pathogen strains that reproduce most rapidly because they leave more progeny. There are circumstances, however, in which slower pathogen reproduction is favoured by natural selection because the host lives for longer. There is thus a trade-off between rapid reproduction and host survival, that is, between high and low pathogen virulence. 2.2 Measures of virulence The earlier definitions of virulence of a pathogen are useful as qualitative descriptions but need to be quantified for formulating mathematical models of the epidemiology or evolution of disease. In studies using animals, the virulence of a pathogen is expressed as its lethal dose (LD50): the number of microbes required to kill 50% of infected hosts. In human studies, it is expressed as the case fatality rate: the number of infected individuals who die divided by the number of individuals infected, usually expressed as a percentage. These definitions measure virulence in terms of host mortality, and thus provide convenient, explicit and quantifiable measures that can be used in mathematical models. They ignore, however, many aspects of the harm that pathogens do to their hosts that do not involve death, but which can be considered to be aspects of virulence, such as how incapacitated a host becomes. 2.3 Virulence as a dynamic property of the host-pathogen system When thinking about the evolutionary biology of infectious diseases, we need to define virulence in terms of host fitness; the virulence of a pathogen is the reduction in the fitness of a host that results from infection by that pathogen. This definition is also problematic for two reasons. First, fitness is very difficult to measure and, second, as emphasised in Section 1, individual hosts vary considerably in their response to infection by a particular pathogen, which creates a problem for all definitions of virulence. Virulence is not a fixed property of a given pathogen but is a dynamic property of the interaction between host and pathogen, changing over time. For example, syphilis in Europe may have been a much more severe disease in the sixteenth century, killing people within months rather than years. Scarlet fever, caused by the bacterium Streptococcus pyogenes, was a disease with a very high case fatality rate in the late 1800s, killing very large numbers of children, but by the 1930s and 1940s it had become a relatively mild childhood disease. This may in part have been due to the development of effective treatments. Influenza is the best-known example of a disease in which periodic outbreaks are caused by the emergence of especially virulent viral strains, the most recent being the 1918 pandemic (see the Influenza Case Study). 2.4 Evolutionary trends The idea that some diseases may have been more severe in the past is often used as evidence that there is a general evolutionary trend by which virulent pathogens become more benign. Further evidence for such a process comes from the history of myxomatosis in Australia, a rare example in which evolutionary change in a disease was directly observed over several years. The flea-borne Myxoma virus is indigenous to South America, where it causes a non-lethal disease in rabbits. In 1950 the virus was introduced in Australia in an attempt to control massive populations of rabbits, themselves introduced from Europe. This action caused an epidemic that killed 99.8% of rabbits (Figure 5.3). A second epidemic killed 90% of the rabbit generation that resulted from the survivors of the first, but a third epidemic killed only 50%. This rapid change was due in part to a decrease in the virulence of the virus, and in part to rapidly evolving immune defences among the rabbits. Indeed, so prevalent is the idea that pathogens generally evolve from being very virulent towards being benign that it has been referred to as the 'conventional wisdom'. The emphasis of much of the recent literature in this area is to challenge this view, and to identify the circumstances in which it is and is not true. Arguments against the decreased virulence hypothesis Venereal syphilis is an example of a disease that may have evolved from one of two milder diseases - pinta and yaws. There is fairly general acceptance that pinta and yaws existed among humans long before syphilis evolved. One theory for the origin of syphilis is that it followed the early transition of human societies from a purely rural existence to urban life. Some suggest the pathogen evolved into a venereal pathogen when the wearing of clothes reduced the frequency of the skin-to-skin contact by which yaws and pinta are transmitted. As a result, a mild disease of children evolved into a severe disease of adults. Another example is provided by detailed genetic analysis of the plague pathogen Yersinia pestis, which suggests it evolved from the much-less virulent human enteric pathogen Y. pseudotuberculosis. However, you need to be wary about distinguishing between evolution of new strains or species with increased or decreased virulence and evolution of more or less virulence within a strain or species. Virulence factors In Block 1, it was noted that the virulence of bacteria is determined by a number of specific aspects of its biology, called virulence factors, often determined by genes derived by horizontal transfer. These virulence factors include the effectiveness of the bacterial pathogen in penetrating host cells, how successfully they evade their host's immune system, and their ability to obtain essential nutrients, particularly iron, from its host. (Such arguments about virulence factors can also be applied to pathogens other than bacteria.) At least one virulence factor - colonisation of the host - can change very rapidly. Cultures of the cholera pathogen Vibrio cholerae collected from pond water are less infectious than cultures collected from human faeces; passage through the gut of a human host induces a change in the bacterium that makes it more infectious. This change involves the activation of V. cholerae genes that are not expressed in pond water. In experiments using mice as hosts, cholera bacteria collected from human faeces were 700 times more infectious than bacteria collected from laboratory cultures. This hyperinfectivity lasts for at least 5 hours in pond water, but disappears in bacteria kept for 18 hours in the laboratory. The mechanism underlying hyperinfectivity may help explain why cholera epidemics spread very rapidly. Virulence in the context of alternative hosts Finally, it is important to consider that the virulence of a pathogen with respect to a particular host depends on whether or not that pathogen has alternative hosts. For such pathogens, it may be entirely inappropriate to consider virulence with respect to humans as being subject to natural selection in humans at all. Some of the deadliest human infectious diseases are caused by pathogens whose natural hosts are other mammals or birds, e.g. Ebola and hantavirus. For these diseases, virulence in humans appears to be much greater than it is in their non-human hosts and is best regarded, not as an adaptation by a pathogen to human hosts, but as a by-product of its coevolution with its non-human hosts.

Disease modelling

The modelling approach to investigating infectious diseases can be traced back to the Swiss mathematician Daniel Bernoulli (1700-1782). In the 1760s he became involved in an acrimonious controversy about the merits of smallpox inoculation with the French mathematician Jean le Rond d'Alembert (1717-1783), who questioned its benefits. Bernoulli developed a model, which he applied to available data to demonstrate that inoculation resulted in reduced smallpox mortality. The basic ideas behind Bernoulli's model are much the same as those used in today's 'compartmental' models, which you will learn about in this unit and in Unit 3.

Infectious dose

The number of pathogens required to make 50% of infected hosts ill. This is sometimes expressed as the 'ID50'.

Force of infection

The rate (per unit time) at which susceptibles acquire infection. Denoted by the symbol λ (lambda ).

Incidence rate

The rate at which new instances of disease occur in a population.

Transmission routes

The route of transmission is important from an epidemiological standpoint because contact patterns in a given population typically depend on socio-economic, cultural and other features of the population, which may vary between different populations, and between different social groups within populations. For example, overcrowded housing and large family sizes may result in increased contact rates for many infections, such as tuberculosis. Low personal and food hygiene due to lack of availability of clean running water may result in increased polio transmission. Differences in the incidence of disease between different social groups can illuminate causes and suggest interventions, as for example with human papillomavirus and cervical cancer (Box 2.4). Patterns over time can also reflect shifts in contact patterns. One example is chickenpox in England and Wales (Figure 2.12), the infection caused by the varicella zoster virus. • Compare the rates shown in Figure 2.12 for 0-4 and 5-14-year-olds. What do you observe? Varicella zoster is transmitted by close personal contact and airborne droplets. What explanation(s) for the pattern of the data might this suggest? • The consultation rates follow broadly similar patterns in the two age groups up to the early 1980s, from which point they diverge with the consultation rates in 0-4-year-olds increasing markedly thereafter. The divergent trends suggest that contact rates with 0-4-year-olds might have increased since the 1980s. This has been attributed to the increase in the proportion of children attending preschool in the UK over this period, resulting in increased contact rates within the 0-4 year age group.

Diversity of ecological communities

The term 'diversity' is often used rather loosely to refer to the variety of species in an ecological community. In fact, ecologists recognise two components of diversity: richness and evenness. Richness is a measure of the number of different species in an assemblage. For example, Trypanosoma has a large number of host species, as you will see shortly in Figure 3.15. Evenness, as the name implies, is a measure of the relative abundance of these different species. For example, there may be four hosts (species richness is four), one of which is ten times more abundant than the other three combined. This would be a highly uneven community. Abundance can be measured as number of individuals (preferably as population density) or amount of biomass. You may come across ecological measures of diversity, i.e. indices that represent the degree of diversity in a community. These indices combine measures of richness and evenness. High richness and even abundance contribute to a high diversity value.

Infectious diseases in 21st Century

This final block of SK320 focuses on the strategies available in the twenty-first century to tackle the full range of infectious diseases that still pose major threats to public health. The aim is to give you a clear insight into the following key areas of scientific investigation. • The recording and mapping of infection in human populations all over the world and the analysis and modelling of infectious disease outbreaks. • The biological, socio-economic and other circumstances influencing the transmission of pathogens and their impact on their human hosts. • The reasons underlying some successful community-based and international public health campaigns to prevent, treat or control infections, and the challenges that must be overcome in order to make further progress. This first unit of the block introduces public health approaches to reducing the massive global burden of infectious disease at the population level. Units 2 to 5 cover the ways in which infections spread, how these can be modelled and quantified, and the role of interactions between pathogens, people and other organisms. Units 2 and 4 focus on the dynamics of the infections, providing important insights that inform treatment and control strategies. Units 3 and 5 take a broader view of infectious diseases, exploring some of the underlying biology and considering the ways in which hosts and their pathogens have evolved. The module ends in Unit 6 with a closer look at the many challenges that face the world's population in our attempts to reduce the global burden of infectious diseases, which still cause so much suffering, disability and death.

Transmission risks and rates

Transmission risks and rates The key feature of infectious diseases, which sets them apart from other diseases, is that they can be acquired by contact with infectious agents. As previously, attention in the following sections is restricted to infectious agents directly transmitted between humans, i.e. by direct person-to-person contact. 6.1 Effective contact The transmission of infection requires three conditions: • presence of an infectious individual • presence of a susceptible individual • effective contact between them. An effective contact is defined as contact between two individuals, A and B, such that if A is infectious and B is susceptible then A infects B. What constitutes an effective contact depends on the infectious agent and its route of transmission. 6.2 Routes of transmission Some of the most important routes of transmission for directly transmitted infectious agents include the following. • Respiratory: this is the typical mode of transmission of many infectious agents such as those causing measles, mumps, rubella, whooping cough, influenza and tuberculosis, all of which can be transmitted by airborne droplets. • Faecal-oral: this is the typical transmission mode for the infectious agents of polio, hepatitis A, rotavirus and cholera, for example; either by direct contact or through contaminated foodstuffs or water. • Sexual: the typical mode of transmission of the infectious agents of AIDS, syphilis, gonorrhoea and hepatitis B. • Vertical: from mother to child, often in utero, as may occur with HIV and hepatitis B viruses. Many infections are transmitted by more than one route: for example, close physical contact plays a role in passing on the measles virus as well as transmission by airborne droplets; polioviruses can be transmitted by the respiratory route as well as by the faecal-oral route; hepatitis B virus can be transmitted by injecting drug use as well as by sexual contact. Other modes of transmission include iatrogenic transmission, that is, transmission caused by medical procedures such as injection or transplantation of infected material. An example of the latter is the transmission of Creutzfeldt-Jakob disease through injection of contaminated growth hormone or corneal transplants.

Changes in numbers of viral pathogen

Viruses such as the influenza virus can be considered here as a second example of pathogen increase in the host. Recall that antibody production against the virus begins 3-4 days after infection. Eventually, this may lead to elimination of the virus. But what numbers of the virus are present by the time that antibody production takes effect? Viruses replicate by infecting a host cell and then producing multiple copies of themselves (Figure 3.13), which are released (usually) by lysing the host cell. Therefore, after each replication cycle, the number of viruses is multiplied by a certain number. To estimate the number of virus particles after 3 days, you need to know three variables: • the number of virions (V) at the start of infection (denoted by V0, i.e. the number at time zero) • the average number of virions produced per host cell (r) • the average time taken for replication (t). Before dealing with the specific values for influenza, we can generate a simple formula that could be used for any virus. • In the light of the variables listed above, what is a simple way of expressing the average number of viral replication cycles in 3 days? • The answer is 3/t, but you need to be careful with the units. If the viral replication time is expressed in hours, the total time under consideration also needs to be in hours. Thus, if viral replication time was 12 hours, the number of viral replications in 3 days (i.e. in 72 hours) would be 72/12 = 6. In order to generalise, you can replace the number of hours over which replication is taking place (72 in the example above) by T. Thus there are T/t replication cycles. If each viral replication produces r virions, the equation for the number of virions after T hours (denoted by VT) is: number of virions after T hours (VT) = number at start (V0) × number produced per replication cycle (r) to the power of T/t or (Eqn 3.2) VT = V0 × rT/t This uses the same logic as the previous bacterial example. In that case, the number of replication cycles was the number of doubling periods (12 = 6/0.5). The number of bacteria produced per replication cycle was 2, i.e. r was 2. So Equation 3.2 is sufficiently general to also be able to predict the number of bacteria after a fixed period of time (BT), knowing the initial number (B0) and the doubling time t. Equations of this type are used widely in ecology to describe the increase in the number of organisms over time, with the assumption that there is no limit to the rates of increase.

Transmission rate

β, is directly related to the basic reproduction number, R0, such that: R0 = β × D where D is the average duration of the infectious period. It was also pointed out that in most circumstances it is difficult to estimate β because it is not usually possible to document all contacts with an infection. In this unit you will consider an important setting in which it is possible to estimate R0 directly for some endemic infections.

Redia

(Plural, rediae.) A larval form with an oral sucker, it will produce either more rediae, or cercariae.

Metacercariae

(Singular, metacercaria.) Cercaria larvae encysted and resting, prior to the parasites being transferred to a definitive host.

Oxidase test

A diagnostic test that is employed to distinguish between different species of bacteria by testing for cytochrome oxidase, using the ability of this enzyme to oxidize the substrate TMPD to produce a purple product.

Slide agglutination test

A diagnostic test used for cholera in which test antiserum against particular O antigens is added (on a microscope slide) to a sample of culture obtained from a patient. If the serum recognises the O antigens in the sample, it binds to them, clumping and immobilising the bacteria.

cryptosporidiosis

A diarrhoeal disease caused by members of the genus Cryptosporidium.

Anthroponosis

A disease that is spread from humans to other animals.

Indirect life cycle

A parasite life cycle in which an intermediate host or vector is involved.

Prions

A transmissible, pathogenic particle believed to consist only of protein. The term prion is derived from their description as proteinaceous infectious particles, and unlike any other infectious agents, prions have no nucleic acid. The exact nature of prions is still poorly understood, but they appear to consist entirely of protein and pass from one animal or species to another via dietary and other routes. Prions give rise to a group of fatal neurodegenerative diseases in humans and animals called transmissible spongiform encephalopathies (TSEs). These include, scrapie in sheep and goats, chronic wasting disease (CWD) in wild deer and elk, bovine spongiform encephalopathy (BSE) in cattle, and Creutzfeldt-Jakob disease (CJD) and variant Creutzfeldt-Jakob disease (vCJD) in humans. Common to all of these diseases, prion proteins form insoluble fibrous aggregates (plaques) within cells of the brain, leading to cell death. This action produces characteristic lesions in the brain, forming vacuoles in the grey matter and giving the tissue a sponge-like appearance upon microscopical examination. TSEs are characterised by lengthy incubation periods (10-30 years for human TSEs), and because of this they are also known as 'slow diseases'. Do not confuse these with the viral slow infections. A small proportion of TSEs are not infectious, but are instead spontaneous or hereditary, caused by mutations in a gene called PrPc. The best characterised TSE is scrapie, a disease of sheep and goats, which has been known in Europe for at least two centuries. The scrapie agent has been studied extensively, and has provided much of the information known about prions. Animals suffering from scrapie have difficulty walking and suffer itching to such an extent that they may rub themselves raw, hence the name scrapie. Although the disorder was long suspected to be infectious, it wasn't until 1936 that the first experimental sheep-to-sheep transmission was achieved. This was done by injecting spinal cord homogenate from a diseased sheep into a healthy animal, satisfying the last of Koch's postulates. The same year, brain homogenates from scrapie-infected sheep were shown to retain their infectivity after passage through filters too small to allow anything except viruses through, leading to the assumption that the disease was viral in origin. Examination of partially purified prion material under the electron microscope reveals rod-shaped fibrils (small fibres) of various shapes and sizes. These fibrils are not destroyed by treatment with nuclease enzymes, and in fact are aggregates of prion protein. They provide a morphological marker for scrapie infection or disease. The first TSEs to affect humans were identified in the 1920s by Creutzfeldt and Jakob. The most common of these diseases is now known as Creutzfeldt-Jakob disease or CJD, which mostly affects middle-aged people. Patients with this disease suffer rapid dementia and other neurological disorders, usually dying within 4-8 months from the onset of symptoms. Other related diseases have subsequently been described: Gerstmann-Straussler-Scheinker syndrome in the 1930s, kuru in the 1950s, and variant CJD (vCJD) and fatal familial insomnia in the 1990s. A blood-based assay for vCJD prion infection has recently been developed which is hoped could be used to diagnose infection before symptoms develop.

Filarial worms

Filariasis is caused by roundworms that inhabit the lymphatic vessels and subcutaneous tissues. Four species are responsible for much of the morbidity due to the disease: • Wuchereria bancrofti (most tropical areas) and Brugia malayi (Asia), which cause lymphatic filariasis, commonly known as elephantiasis • Onchocerca volvulus (mostly in sub Saharan Africa, Yemen, some parts of Central and South America), which causes onchocerciasis, commonly known as river blindness • Loa loa (in equatorial Africa), the eye worm causing loiasis. The worms are long and slender with females around 60-100 mm long and males shorter (up to 40 mm). A more detailed summary of the characteristics of these four species appears in Section 5.2.3. • What are the two main differences between this cycle and that of hookworms? 1. A vector is involved to transfer the larval worm from host to host. 2. There is no free-living larval form.

Brugia timori

Vector is Anopheles barbirostris. Host is humans. Found in regional lymph vessels. Causes acute filarial fever and lypodema of lower legs. Found in Eastern end of Indonesian archipelago.

Chlamydia

We are now going to look at the genera in more detail, highlighting how the Chlamydia life cycle relates to the disease it causes, and the epidemiology of Chlamydia trachomatis and other sexually transmitted infections (STIs). STIs include diseases caused by viruses, bacteria, fungi and parasites. As their name suggests, STIs are produced by infectious pathogens transmitted from person to person and primarily by sexual contact - although several important STIs can also be passed: • from mother to child during labour • from mother to child during birth • via breast milk • via blood products • during transplant surgery • from contaminated medical devices or illegal drug-injecting equipment. Young people are a particularly difficult group to reach with 'safer sex' messages, especially in low- and middle-income countries (LMICs) where services to prevent and treat STIs are inadequate to meet population needs. Widespread cultural taboos that exist almost everywhere against sex outside marriage mean that young, single people who are affected are even less likely to come forward for treatment than married people. Study note The disease chlamydia and the organism that causes it both have the same name. The organism's name should always be written in italics and with an initial capital letter (Chlamydia). When writing about either the disease or the organism, you should be careful to write it accurately.

Rice water stools

A symptom of cholera. People with the disease often have watery stools, with a vaguely fishy odour, containing flecks of mucus. This substance is said to resemble water from boiled rice, hence the name given to it.

STIs

Chlamydia trachomatis chlamydia Discharge from the penis; asymptomatic in up to 70% of women, but can develop into pelvic inflammatory disease (PID) with abdominal pain, fever, damage to the fallopian tubes and possible infertility. Mother-to-newborn transmission can cause blinding eye infections. Neisseria gonorrhoeae gonorrhoea Discharge from the penis and pain when urinating in men; asymptomatic in up to 70% of women, or vaginal discharge and pain when urinating, which can progress to PID, and sepsis if bacteria get into the bloodstream. Mother-to-newborn transmission can cause blinding eye infections. Treponema pallidum syphilis Initially produces sores typically on the genitals, anus, mouth or hands; later, pain in the bones, lymph nodes, headache, fever, tiredness; ultimately, widespread tissue destruction. Viruses Human immunodeficiency virus (HIV) acquired immune deficiency syndrome (AIDS) Initially asymptomatic or transient mild fever; later, repeated opportunistic infections, particularly in the lungs and gut, weight loss, fevers, certain cancers, brain deficits; relapsing disease is ultimately fatal if untreated. Mother-to-newborn transmission leads to paediatric AIDS. Herpes simplex virus type 2 (HSV2) genital herpes Recurrent attacks of painful genital blisters, which burst and become painful ulcers that heal in 2-3 weeks. Human papilloma virus (HPV) genital warts Initially asymptomatic, then painless warts develop in the genital area and around the anus; certain HPV subtypes are associated with cervical cancer in women. Hepatitis B virus (HBV) hepatitis B Initially asymptomatic, then sudden onset of headache, fever and jaundice; chronic cases may lead to fatal liver failure or liver cancer. Protist Trichomonas vaginalis trichomoniasis Can remain asymptomatic for years (particularly in men); women may develop inflammation and itching in the genital area and a foul-smelling vaginal discharge. Fungus Candida albicans candidiasis (or thrush) Inflammation of the head of the penis and foreskin; in women, itching/burning and a thick white discharge from the vagina; in both sexes, oral candidiasis produces small white sores in the mouth that may spread to the gut. Bacteria of the family Chlamydiaceae are obligate intracellular pathogens with a dimorphic life cycle. They have a lipopolysaccharide-rich membrane similar to that of Gram-negative bacteria, but no peptidoglycan in their cell walls. The extracellular form of the organism, the elementary body (EB), possesses characteristics that enable its survival outside the host cell. It is very small, from about 0.2 to 0.6 μm in size, and has a tough outer membrane. This is the infectious form of the organism, and it enters the host cell by binding to receptors on the cell surface. The intracellular form, the reticulate body (RB), allows intracellular multiplication of the organism. RBs are much larger, being up to 1.5 μm in diameter and are clearly visible within cells (Figure 1.1). The RBs use nutrients from the host cell to divide numerous times by binary fission, where identical daughter cells are made. Note that a feature shared by most STIs is that the causative pathogens often develop in their human host during a long asymptomatic period. • What problems for the control of STIs arise from so many having a long asymptomatic period? • Infected people can pass on the pathogen to subsequent sexual partners (or from mother to newborn) before they realise that they are infected. This also results in a delay before seeking medical help, so the condition can become much harder to treat. Finally, it is difficult to estimate the prevalence of infection in a population and provide services to meet diagnostic and treatment needs when so many cases are 'hidden'. With this word of caution in mind about the reliability of case estimates, particularly in LMICs, this case study focuses on the epidemiology of one of the most prevalent and most readily curable STIs in the world. In this case study, you will learn about the causative pathogen Chlamydia in more detail, the clinical progression of symptoms and disease stages, the diagnostic tests and the methods of treatment. We also consider the number of people infected, looking both at the worldwide picture and also at data from the UK and Europe. At least two days after the host cell has been infected, the numerous newly generated RBs become EBs again. They stop metabolising (i.e. using their own enzymes to break down nutrients for energy) but they are capable of infecting other cells. The host cells rupture in the process of cell lysis, and release the EBs and RBs, which can then go on to infect other cells and continue the cycle. The life cycle of Chlamydia is superficially similar to the propagation cycle of viruses. However, Chlamydia are bacteria, meaning that unlike viruses they metabolise. The release of so many infectious EBs very rapidly through cell lysis indicates why Chlamydia is so infectious. The fact that it is an intracellular parasitic bacterium impacts on the immune response to, and the methods for diagnosis of, this infection, as you will see later. The family Chlamydiaceae is split into two genera: • Chlamydia, in which the main species of human pathogen is Chlamydia trachomatis • Chlamydophila, where the two respiratory pathogens Chlamydophila psittaci and Chlamydophila pneumoniae have now been placed. The species are differentiated in several ways: growth characteristics, staining properties of the inclusions, DNA/ RNA sequences, and serology (Greenwood et al., 2007). It is the Chlamydia genus that is the focus of this case study. C. trachomatis can be divided into two biovars based on invasiveness: the oculogenital biovar and the lymphogranuloma venereum (LGV) biovar. The oculogenital biovar is less invasive, and causes most of the eye infections and many genitourinary infections, while the LGV biovar causes a more invasive genital infection, lymphogranuloma venereum, which gives it its name. These two biovars can be further divided into serovars (serotypes) on the basis of the major outer membrane protein (MOMP), Biovar Serotype Site of infection Disease Method of spread Main area of occurrence Oculo-genital A, B, Ba, C eye trachoma, conjunctivitis hand to eye, fomites, flies Asia, Africa D-K eye conjunctivitis hand to eye, fomites, flies worldwide D-K eye ophthalmia neonatorum perinatal worldwide D-K genital tract urethritis, cervicitis, salpingitis, pelvic inflammatory disease sexual worldwide LGV L1, L2, L3 genital tract lymphogranuloma venereum (LGV) sexual worldwide but more common in Africa, Asia and South America 3 Pathogenesis of C. trachomatis Little is known about how Chlamydia organisms produce disease. It is most commonly transmitted from person to person through sexual contact, and is generally most prevalent in teenaged women and in men in their early twenties. The incidence of infection declines with advancing age, which may be due to the development of immunity to the infection. There are relatively few cases of chlamydia in women over 35. The bacterium can also infect newborn babies through maternal transmission as they pass down an infected birth canal. Infants tend to have chlamydial eye infections within a week of birth to a mother infected with Chlamydia; pneumonia may also develop from this. Most of the damage caused by Chlamydia to host tissue is due to the cell destruction and scarring following the host immune response (see Section 4). If chlamydia is left untreated it can lead to sterility in women. In men, where symptoms are present, it causes non-specific urethritis, which is inflammation of the tube connecting the bladder to the penis for urination (where 'non-specific' means that this is a symptom that has a number of possible causes, not just chlamydia), but rarely leads to any long-term health effects. In about 50% of cases in men and about 75% of cases in women, there are no symptoms at all. As well as urethral infection in men and women, Chlamydia can also infect the pharynx (throat), the rectum and the female cervix. Chlamydia is also a major cause of the eye disease conjunctivitis. Trachoma Another infectious eye disease caused by C. trachomatis is trachoma. This is probably one of the oldest infectious diseases known to humankind, although the causative agent was unknown until recently. The WHO estimates that trachoma affects approximately 84 million people, of whom 8 million are visually impaired, and it accounts for more than 3% of the world's blindness (WHO, 2011). It is common in rural areas of Africa, Asia and South America. The causative organism is spread by direct contact with discharge from an infected person, e.g. by fingers, handkerchiefs or flies. The scarring caused by repeated eye infections can cause the eyelid to turn inward, resulting in blindness from abrasions on the cornea. The resulting inflammation also affects the flow of tears, which can lead to further secondary bacterial infections. Because Chlamydiae are intracellular, it is very difficult for antibodies to reach them. Cell-mediated immunity therefore plays a major role in the body's response to infection. The organisms themselves also have characteristics that enable them to persist in the body and prevent it developing immunity. These characteristics are summarised in Box 1.1. Box 1.1 Immune-evasion mechanisms of C. trachomatis (from Brunham and Rey-Ladino, 2005) Enhanced survival outside host cells • Presence of antigenically diverse surface proteins, such as MOMPs and polymorphic membrane proteins, avoids detection by antibodies. Enhanced survival inside host cells • Replication within a membrane-bound inclusion body limits exposure to antibodies and to host-cell antigen-processing and antigen-presentation machinery. • Inhibition of mitochondrial release of cytochrome c, which is required for caspase-9-mediated apoptosis, inhibits apoptosis of infected host cells. • Presence of a particular tyrosyl radical site in the bacterial ribonucleotide reductase is probably responsible for increased resistance to nitric oxide. C trachomatis shares this feature with other intracellular pathogens, including Mycobacterium tuberculosis and M. bovis. Reduced inflammatory responses • Presence of a lipopolysaccharide (LPS) of reduced potency decreases LPS-mediated activation of host cells. C. trachomatis LPS is at least 100-fold less potent at activating host cells than other types of bacterial LPS. Reduced adaptive immune responses • Cytoplasmic secretion of a C. trachomatis protease that degrades transcription factors required for the transcription of MHC genes downregulates IFNγ-induced expression of MHC class I and class II molecules. Ability to persist as alternative intracellular forms • Development of persistent, non-replicating forms after exposure of C. trachomatis to antibiotics, nutrient deprivation or cytokines (such as IFNγ) or after infection of monocytes. • Upregulation of genes involved in intracellular survival, so C. trachomatis persistence in response to IFNγ is controlled at the transcriptional level. • Maintenance of viability of persistent forms, which can rapidly regain the normal developmental cycle on removal of IFNγ. • Expression of genes encoding tryptophan synthase and a tryptophan repressor by genital strains of C. trachomatis suppresses the growth inhibitory effect of IFNγ if indole is available. As C. trachomatis infects the epithelial cells, the host generates an inflammatory response by producing cytokines such as IL-1α, which stimulates other epithelial cells to produce pro-inflammatory cytokines such as TNFα and IFNγ. Other cytokines such as IL-8 and IL-6 are also important. The combination of cytokines and chemokines produced causes a protective Th-1 type response as well as antibody production. Chlamydia is the most common bacterial sexually transmitted disease not only in the UK but also in the world, with approximately 89 million cases each year. Remember, however, that such figures are difficult to determine because so many cases are asymptomatic. Up to 80% of women who have chlamydial infections do not show any symptoms, and in those who do, the symptoms may be non-specific, such as pelvic pain. This means that many people develop complications from the infection that can have far-reaching consequences, such as infertility due to scarring of the fallopian tubes. Up to 40% of women with sexually transmitted chlamydia develop PID, which makes them six to ten times more at risk of an ectopic pregnancy (when the embryo implants in the fallopian tube or elsewhere outside the uterus). Damage to the fallopian tubes is responsible for 30-40% of cases of female infertility. Mother-to-newborn transmission also results in significant disability: up to 33% of babies born to women with chlamydia develop a potentially blinding eye infection. Even in countries with good data collection and healthcare systems, estimates have a wide range: for example, 4-5 million new cases may be occurring annually in the USA. Chlamydia is the most commonly diagnosed STI in the UK, accounting in the 16-24 age group for around 55% of all STI cases in men and around 75% in women (HPA, 2008). Diagnostic surveys of selected population groups in various low-income countries suggest that the prevalence of chlamydia is reaching epidemic proportions among young people. The morbidity due to chlamydia and its disproportionate impact in Africa and South-East Asia is evident from Figure 1.3. (Recall that DALY is a measure of the impact of a disease on people's lives.) For many diseases or conditions, diagnosis follows a visit to the doctor because a person has unexpected symptoms. In the case of chlamydia, this does not happen. • How can chlamydia be detected if there are often no symptoms? • Because of the asymptomatic nature of chlamydia infection, it is often not detected. In view of its impact on female fertility, it is a prime example of an infection which would benefit from a screening programme. For this reason, there are now national screening programmes for chlamydia in many countries. Indeed, the WHO has introduced a grading scheme to help with clinical diagnosis. The programmes aim to screen young people annually or each time they change sexual partners. It is offered at routine medical check-ups, with vaccinations (e.g. MMR inoculations), during other screening procedures, and in sexual health clinics, youth clinics and at other convenient places. Many people are embarrassed about going in person to a clinic and getting a test; for this reason many over-the-counter (OTC) kits are available for chlamydia testing. Most of these OTC tests require urine samples to be taken, which are sent off to a laboratory by post for subsequent analysis. However, such kits are less likely to be available to people in low- and middle-income countries, where the effects of the disease are felt most keenly. Unusually for a screening programme, the tests used are also diagnostic tests. They are very accurate and correctly identify the vast majority of positive cases. The methods used to diagnose chlamydia in a sample are described in Section 6. A national screening programme for chlamydia was set up in the UK in 2003 in an attempt to control the spread of the disease and prevent the complications that follow non-treatment. The programme targeted those people who are most at risk of infection. The following activity provides more information about this particular scheme. Because the organism causing chlamydia is an intracellular parasite, some of the tests you have read about for diagnosing or identifying other pathogens are less appropriate here. • Why does the fact that Chlamydia is an intracellular parasite complicate diagnosis? • In order to be able to visualise the organism through microscopy, host cells would also need to be present, which means that the collection of samples will be more invasive. • What other diagnostic methods might be suitable? (Think of as many examples as you can from your study of this module.) • Culture may be possible but, as it is an intracellular parasite, the pathogen will need to be grown in particular cells in the laboratory, which are expensive to produce and maintain. • Microscopy may be difficult as the EBs are small; however, EBs and RBs can be imaged with appropriate staining. • Immunological methods - detecting either the bacterial antigen or antibodies produced by the infected individual - may be used, although the latter depends on extracellular antigens being present for long enough for an immune response to have developed. • Nucleic acid amplification techniques may be the most suitable. Table 1.3 shows the different types of tests available for diagnosis. The most widely used of these methods are discussed below. Table 1.3 Tests used for diagnosis of C. trachomatis Test method Specimen Comments, and amount of time required for test Cell culture urethral swab; cervical swab specialised culture medium and culture conditions; skilled staff: 72 hours Direct immunofluorescence urethral swab; cervical swab simple to process; subjective results; skilled staff: 30 minutes Nucleic acid amplification test (NAAT) vulval, cervical or urethral swab; first-void urine needs specialised equipment and qualified staff: 4-24 hours ELISA urethral swab; cervical swab easier to perform than culture; easy to read: 3 hours Gene probe vulval, cervical or urethral swab needs specialised equipment and qualified staff: 4-24 hours (Source: Watson et al., 2002) Some of these test methods are discussed further in the rest of this section. Study note A short note on the terminology of diagnostic tests: recall that sensitivity is a measure of how much (or how little) of a sample needs to be analysed before a clear-cut result is obtained in a test. It measures the proportion of true positives that are correctly identified. Specificity relates to the accuracy of the recognition between the test reagent and the patient's sample, and measures the proportion of true negatives that are correctly identified. You learned about this in Block 1 Unit 9 (Section 1.5). 6.1 Cell culture Because C. trachomatis will only grow inside host cells, culture of the organism is difficult, and needs to be carried out in a specialist laboratory. The organisms are grown in McCoy or other cultured cell monolayers for 48-72 hours, and the inclusions are stained with iodine (Figure 1.5), or by using fluorescent antibodies that bind to the MOMP. • What else would these antibodies allow you to detect? • Since they are specific for MOMPs, they allow identification of the serovar. The test has a specificity of almost 100%, and is used as the gold standard for comparison with other tests. Also, because the organism is in culture, antibiotic susceptibility tests can be carried out. The drawbacks to using culture are its lack of sensitivity (said to be as low as 50%; Watson et al., 2002), the length of time it takes to achieve a result, and the difficulty of obtaining correct samples for the test: cervical swabs are needed, and this procedure is more invasive than, say, obtaining urine specimens. Also, the sample needs to be stored and transported at a low temperature to maintain bacterial viability. For these reasons, other methods have been developed that overcome some of these problems, two of which are discussed in the remainder of Section 6. A direct fluorescent antibody (DFA) method can be used for detecting Chlamydia. It is very specific (about 99%) and sensitive (96%). • What do the figures for specificity and sensitivity mean for the percentages of false negatives and false positives? • With 99% specificity, 1% of those who do not have a chlamydial infection will test positive (false positives); 96% sensitivity means that of those who do have the infection, 4% nevertheless test negative (false negatives). The DFA test uses a monoclonal antibody to the MOMP. In this technique, labelled antibodies conjugated to fluorescein are mixed with, and allowed to bind to, any MOMP present in the swab sample smeared on a microscope slide. The slide is then viewed with a fluorescence microscope. The presence of visible fluorescence indicates that the MOMP antigen is present and thus the sample is positive for Chlamydia. The advantage of this test is that it does not require samples to contain viable cells (living cells which can be cultured), which means that sample handling is less important than with culture methods. It does require a skilled operator to carry out the microscopy, however, and it is not very sensitive for samples taken from the male urethra. 6.3 Nucleic acid amplification Instead of the cell culture method, many laboratories are now moving towards nucleic acid amplification tests (NAATs). The National Chlamydia Screening Programme (NCSP) in the UK uses NAATs, and the OTC chlamydia tests also use this technique. There are several reasons for adopting this method. • First, the sample used in the test is urine, which is easily obtained, non-invasive and acceptable to patients. • Second, the test is rapid, selective (i.e. a high degree of preference for the target analyte) and sensitive. • Third, most of the test can be automated. Although NAATs are not inexpensive, they are preferred for many screening programmes because of these factors. The sample must be prepared for analysis before the NAAT. The procedure requires many more steps for urine samples than for swab samples. Commercial systems are available that automatically process the samples and analyse them using the NAAT method. Amplification tests include PCR (see Block 1 Unit 9, Section 2.6.1), strand displacement amplification (SDA), which uses restriction enzymes to target the area of amplification on the genome, and transcription mediated amplification (TMA), which targets ribosomal RNA. All of these tests are available as commercial kits. 7 Treatment of chlamydial infection Uncomplicated chlamydial infection is easy to treat and cure once it is diagnosed, although all the people involved in sexual contact with the infected person (current and recent) should be tested and treated to prevent the spread of infection. The antibiotics generally used to treat the condition are Doxycycline (taken twice daily for 7-14 days), and Azithromycin (single dose). • Why do you think the treatment regimes are so different? • Chlamydia has a complex life cycle, so treatment has to be for a prolonged period to cover the entire cycle, hence 7-14 days. Azithromycin can stay in the tissues for a long time, so can be effective after only one dose. As with many infectious diseases, prevention is better than cure, so the 'safe sex' message must continue to be broadcast, particularly to the most vulnerable group: under-24-year-olds. An effective vaccine against chlamydia would be ideal, but the intracellular nature of the causative organism means that the task is complex. is exactly what the model predicts. There we can see the trend in the epidemic. And here this is the mean partner change rate per year over time. And we can see that over time, as the epidemic takes its course, the mean rate of partner-change falls because higher activity individuals are dying more quickly than lower activity individuals. This slide, graph bottom rights shows that if nothing else changes, such as the impact of intervention efforts, HIV prevalence will level off when the incidence of infection and the incidence of death come into equilibrium, which is what we're seeing towards the end of this simulation here. Note that in this very simple model we've not modelled any volitional behaviour change, i.e. we haven't modelled any population response to the epidemic. As such this model predicts that you'd expect to see falls in HIV prevalence even if you didn't have any volitional behaviour change by the population. Slide 20 (22:02) This slide here shows the trends in HIV prevalence in three countries with the most severe HIV epidemics - South Africa, Zimbabwe and Uganda and our model predictions in the background. It is too early to tell from the data here, what's going to happen in South Africa with it recent and explosive HIV epidemic, but we can see in the more mature epidemics in Zimbabwe and Uganda the epidemic appears to be following the rough trajectory predicted by our very simple model. However, you might think that the decline predicted in Zimbabwe and Uganda is larger than we predicted, which might suggest that our models are too simple, or parameter values are wrong, or indeed there have been volitional reductions in risk behaviour in these populations, and indeed we do have empirical data that show that to be true. Slide 21 (22:47) In this very brief introduction to the modelling of sexually transmitted infections we've looked at the characteristics of STIs and how the models are adapted to account for these characteristics. We've seen how to alter the force of infection equations so we don't assume frequency dependence and how modellers usually split the transmission parameter into a behavioural and a biological component. We've seen how simple models of curable STIs can help us understand how the importance of risk heterogeneity in allowing STIs to spread and become endemic, and we've seen how simple HIV models can be used to predict HIV trends. Models have been enormously useful tools to understand the epidemiology and control of STIs. But despite recent advances in the HIV controls such as male circumcision and anti-retroviral therapy, the global burden of disease remains high and millions of people continue to be newly infected with HIV each year. As such, STI modelling will continue to be an important tool for research. Slide 22 (23:44) To download any of the models used in the session and many more, point your browser at these links. These are the references, This slide also shows the references we drew on today. To read about STI modelling in more detail, including relaxing the random mixing assumption, and for models of HIV/STI co-infection and many more aspects of infectious disease modelling, an introductory modelling book is now available. Happy modelling!

Cysts

Highly resistant structures formed by many parasitic protists, which enhance survival and transfer to new hosts.

Device-associated infections

Infections entering via an indwelling respiration tube, urinary catheter, intravenous blood line or other invasive device; a category of healthcare-associated infections (HCAIs).

Zoonosis

Infectious disease caused by pathogens or parasites transmitted to humans from their principal reservoir in another vertebrate species (e.g. rodents, antelope, monkeys, cattle, and poultry). Transmission can occur through direct contact with host animals, or in contaminated food or water, or via an intermediate vector (e.g. fleas, lice, flies). Once established in a human population, the causative agents of some zoonosis (e.g. pneumonic plague, pulmonary tuberculosis) can also be transmitted directly from person to person.

Cystericus

Larval stage of a tapeworm, consisting of a single invaginated scolex enclosed in a fluid-filled cyst.

Malaria

Malaria is caused by various species of Plasmodium. Severe, often fatal, disease is predominantly caused by Plasmodium falciparum. Milder forms of malaria are caused by P. malariae, P. ovale, and P. vivax. Another species, P. knowlesi, causes malaria in macaque monkeys, but is also an emerging zoonosis in humans. Around 225 million cases of malaria occur annually, resulting in 0.8 million deaths Symptoms include anaemia with recurrent episodes of fever and headaches, progressing in severe cases to coma and death. Around 225 million cases and 0.8 million deaths from malaria are reported globally each year. Ninety percent of these deaths occur in Africa, where malaria accounts for one in five of all childhood deaths (source: UNICEF, 2011). Infection during pregnancy results in severe maternal anaemia, contributing to low birthweight and increased risk of infant mortality. Paradoxically, child malaria infection within the first few months after birth in endemic areas often results in relatively mild symptoms. This is because many women have developed antibodies against Plasmodium due to repeated exposure to the pathogen. Maternal antibodies are actively transported across the placenta to protect the child in utero and remain in the baby's circulation for the first few months after birth. This temporary immune protection is known as passive immunity. Reports of the characteristic periodic fevers associated with malaria are first recorded in ancient Chinese writings of 2700 BC (Cox, 2002) and references to the disease are found throughout recorded history in many other cultures. However, the association of Plasmodium with humans stretches back more than 50 000 years (Joy et al., 2003) and at least one species may have originally been a zoonosis from gorillas (Liu et al., 2010). Much progress was made in the eradication of malaria in the late nineteenth and early twentieth centuries by targeting the mosquito vectors. For instance, programmes of wetland drainage, spreading a layer of oil on standing water, and the introduction of the pesticide spray DDT (also known as the National Malaria Eradication Program) managed to eliminate malaria from the United States within 4 years (1947-1951). Similar programmes made significant impact on malaria in tropical areas and Africa. However, indiscriminate use of DDT pesticides during the 1960s led to the development of resistance in Anopheles populations. The use of DDT was withdrawn and funding for eradication programmes was not sustained. Years of neglect followed, and the number of malaria cases grew dramatically. It is only comparatively recently that the international community has returned its attention to malaria, resulting in increased funding for research and the introduction of programmes of vector control, including insecticide spraying of houses and the distribution of insecticide-impregnated bed nets. Currently, the two simplest methods for diagnosing malaria are microscopy and antibody techniques (also known as rapid diagnostic tests (RDTs)). Both types of approach can be carried out in the field and are described in the sections that follow. Molecular techniques for diagnosing the disease (which are discussed briefly in Section 2.3) are more sensitive than the microscopy and antibody approaches mentioned above, but require additional equipment and resources. The most straightforward diagnostic test for malaria comprises the detection of asexual parasites in a blood sample taken from a peripheral area (such as an ear lobe or finger) and examined with a microscope. The sample is taken before any antimalarial drugs are given to the patient, usually by pricking their finger. Drops of blood are placed directly onto a microscope slide to give both a thick film (comprising several large drops) and a thin film (composed of one small drop that is smeared across the slide). The thick film of blood is allowed to dry in the air and is then stained with Giemsa (a standard stain of blood samples) or Field's stain. Staining unfixed red blood cells in this manner causes them to undergo haemolysis (lysis of red blood cells) so, on viewing the sample with a microscope, just their empty cell membranes or 'ghosts' can be seen. This allows the stained Plasmodium parasites to be seen easily, despite the thickness of the film. Using a thick film sample allows a large volume of blood to be examined for the malaria parasites, and provides quantitative evidence concerning the number of parasites in the patient's blood. These thick films allow as few as five parasites per microlitre to be detected, and at least 200 different fields of view at ×1000 magnification are examined before a negative diagnosis can be made. The thin film of the blood sample is first fixed to the microscope slide by heating and then, like the thick film sample, stained with Giemsa. The process of fixing the sample means that the red blood cells are not haemolysed when stained. Instead, the staining agent acts mainly on white blood cells, turning their nuclei blue and the cytoplasm pinkish. Thin films of blood take around ten times longer to examine than thick films, so are not used for routine diagnosis. They do provide a much clearer view, however, and allow the species of Plasmodium parasite to be identified. This factor is crucial, since it allows the risk to the patient to be assessed and the most appropriate treatment to be chosen. Even the trophozoites, schizonts and gametocytes of the four malaria-causing species can be distinguished microscopically using thin films, and the stages seen provide information on disease prognosis. The appearance of the infected red blood cells, which may change size, shape, colour or develop particular markings (called Schüffner's stippling) also differs depending on the species of parasite involved. Interpreting blood films requires time, expertise, appropriate reagents and a relatively expensive microscope. Antibody-based rapid diagnostic tests (RDTs), however, can provide a cheaper and easier alternative. For these tests, a drop of the patient's blood is applied to a dipstick or plastic cassette containing antibodies specific for malarial antigens. If a series of bands is visible within 10-15 minutes of the blood being applied, and these bands match those given by a positive control sample, the patient has malaria. An example of such a test is illustrated in Figure 1.3, which shows three dipsticks, all with a broad (positive control) band showing that the tests have worked. The negative control shows no other bands, but samples containing Plasmodium give rise to a banding profile characteristic of the particular species present. Although this test allows different Plasmodium species to be distinguished, it does not provide quantitative information about the parasite load, which must instead be determined by microscopy. Molecular methods, such as an enzyme-linked immunosorbent assay (ELISA) and polymerase chain reactions (PCRs), have been developed for Plasmodium spp. to determine current infection and/or past exposure to malaria. However, they are more often employed for epidemiological research and patient screening in specialist laboratories, and are not suitable for routine diagnosis. A new molecular technique that may acquire a role in routine diagnosis is loop-mediated isothermal amplification (LAMP). This is similar to PCR in essence, but has the advantage that it does not rely on cycles of temperature change to cause amplification of target DNA sequences, and is faster than traditional PCR. LAMP permits earlier diagnoses when parasite levels are low and, therefore, more controllable with drug treatments. Primer sets can be used that recognise target DNA sequences unique to each species of Plasmodium, and to a variety of emerging drug-resistant strains. 3 Host responses to malaria The simplest strategies are often the most effective, and this is also true when it comes to tackling infectious diseases like malaria. For example: • Programmes of education about malaria and the malaria cycle teach people how to reduce their risk of getting the disease, and how to care for those that are affected. • Drainage or removal of the wet, stagnant breeding places of mosquito larvae is very important. • Spraying with insecticides for the control of mosquitoes can be effective, and often involves pyrethroids or DDT. Indoor residual spray (IRS) and insecticide-treated bed nets (ITN) are also effective techniques for prevention and control. However, these practices must be supported by surveillance data to establish the species of Plasmodium being transmitted, and the species and numbers of mosquitoes that are spreading the infection. It is also important to record and analyse the transmission rates among human hosts. Together, this data allows for more efficient targeting of resources in the on-going fight against the disease. Currently, there are three classes of antimalarial drugs in use (see Table 1.1), and all of them target the blood stages of the parasite's life cycle. Table 1.1 The three current classes of antimalarial drug and their mechanisms of action Drug category Specific examples Mechanism of action Quinolines Quinine, chloroquine, mefloquine, amodiaquine Inside the erythrocyte, Plasmodium gains its nutrition from digesting haemoglobin within its food vacuole. Toxic heme is produced as a by-product of this proteolysis, which the parasite converts to non-toxic hemozoin. Quinoline drugs are transported to the food vacuole and inhibit this conversion, thereby poisoning the parasite with its own waste products. Antifolates Pyrimethamine, sulphadoxine, proguanil, dapsone Products of the folate metabolic pathway are essential for producing the purines and pyrimidines needed for DNA synthesis, and for the metabolism of certain amino acids (Met, Gly, Ser, Glu, His). In common with many microorganisms, Plasmodium synthesises folates from simple precursors whereas higher organisms, such as humans cannot do this and rely on dietary intake of folates. Although Plasmodium can scavenge some of these folates from the host blood, these are in short supply and the parasite relies heavily on its own enzymes for folate biosynthesis. Antifolate drugs inhibit the enzymes dihydrofolate reductase (DHFR) and dihydropteroate synthase (DHPS); thereby starving the parasite of this essential nutrient. Artemisinin-based combination therapies (ACTs) Derivatives of the Chinese herb Artemisa annua Artesunate, artemeter, dihydroartemisinin, artelinic acid, artenimol, artemotil. Artemisinins are sesquiterpene lactones that act in a similar way to quinolines by blocking the conversion of heme to hemozoin. Although highly effective, artemisinins are rapidly broken down in the body. Thus, they are often used in combination with longer-lasting quinoline drugs or antifolates. • Some strains of Plasmodium have developed resistance to quinoline drugs. What biological mechanism might lead to such resistance? • Many resistant strains have acquired mutations in genes encoding membrane transporters that are required to transport the quinoline drug into the food vacuole. Thus, the drug never reaches its target in sufficient quantities to kill the parasite. • Other strains of Plasmodium have developed resistance to antifolate drugs. What biological mechanism might lead to such resistance? • These strains of Plasmodium have mutations in the genes encoding the DHFR and DHPS enzymes that hinder the binding of antifolate drugs, thus minimising their impact on the parasite's ability to make folate. The malaria parasite has a complex life cycle, progressing through several stages whilst in the host and in the mosquito vector. This presents a 'moving target' to the host immune system, which has to adapt its methods of attack as the parasite progresses through its lifecycle. However, the variety of stages that the parasite moves through also increases the number of potential targets for vaccine development. The three main lifecycle stages that have currently been targeted for vaccine development are: • Pre-erythrocytic stage: covering the stages from when the mosquito injects sporozoites into the blood and they infect hepatocytes, to when the merozoites burst out of the hepatocytes into the bloodstream. • Asexual erythrocytic stage: occurring within red blood cells in a cycle of invasion, replication, rupture and release of further infective merozoites. • Gametocyte stage: covering the stages when a small number of asexual merozoites convert to sexual gametocytes in the host, and the entire developmental stages within the mosquito vector. The details of drugs that have been developed to target these stages are not discussed here, since the field of malaria vaccine research is so active at present that any description of vaccines could be out of date within a few months. Instead, there are websites where you can find the latest information (see Box 1.1). Activity 1.3 also provides the opportunity for you to learn more about antimalarial vaccines by reading a selected short review article on the subject. Malaria parasites are caused by Plasmodium, which is a member of the Apicomplexa. The Apicomplexa are protozoa characterised by having an invasive stage or more invasive stages during the life cycle, which can actually get into cells. And they have an arrangement or burrowing into cells, they have, the actual invasive stage is able to move along surfaces and actually get into a cell and live inside it. Slide 3 (00:42): This is a list of the malaria parasites of man, and I am going through the different stages in the life cycle. Transmission of malaria parasites is indirect viaa definitive host, which is a vector mosquito. The reservoir for transmission is man, the intermediate host, and so this is an anthroponosis. This is exceptfor P. knowlesi, which has been shown recently to be fairly widespread in Borneo for instance, which is malaria acquired from a monkey reservoir, obviously through a mosquito. But this is actually a zoonosis, not very widespread at the moment. The life cycle of the parasite has an asexual and a sexual stage, and is mainly haploid. The asexual stages of schizogony in liver of the human are seen after the bite of mosquito. Merozoites from actively growing and dividing liver schizonts invade red cells to give dividing schizonts in a continuous erythrocytic cycle. Dormant liver hypnozoites are present in P. vivax and P. ovale. They are responsible for relapses. The immature sexual stages in the blood are ingested in the blood-meal by the female mosquito, and they mature into male and female gametes in the mosquito stomach, and then the sporogonic cycle, or sporogony starts by the syngamy, that's the fusion of the gametes in the mosquito, meiosis and multiple division of the zygote to give infective sporozoites, which infect the liver of the next human. Slide 4 (02:19): The species of Plasmodium which infect humans throughout the tropics are P. falciparum, which is a malignant tertian malaria: we see drug-resistance in this species, it causes severe anaemia in children and deaths occur due to organ failure. In the case of the "benign" tertian malarias, P. vivax and P. ovale, death is not usually the end of the infection, and it is more of a benign infection, but the actual fevers that occur during these infections are not particularly benign. In P. malariae, the "Quartan malaria", which also occurs throughout the tropics, wherever the other malarias are transmitted, basically, this particular infection if it is repeated in childhood can lead to diuretic kidney and nephrotic syndrome. Slide 5 (03:17): In endemic areas, the presence of parasites in the blood does not exclude another pathology, in other words, you can mistake malaria symptoms for something else, and you can also mistake the symptoms of something else for malaria. Malaria can be confused with acute respiratory infection, tonsillitis, pharyngitis and meningitis. Slide 6 (03:41): Here we have our female anopheles mosquito which is taking blood from a human's skin, and at the same time injecting infective sporozoites in the saliva. Slide 7 (03:54): The sporozoites are about 11 microns long, and are carried in the blood to the liver, where they invade the heptocyte cells of the liver. Slide 8 (04:07): Each sporozoite rounds up, and then starts to devide inside a liver cell, and it takes about from 5.5 to 15 days, depending on the species, for the liver schizonts to be mature, that is, fully divided and basically composed of invasive merozoites, which will then escape from the liver into the blood stream. Slide 9 (04:33): This picture shows a cultured pre-erythrocytic stage of P. berghei, the rodent malaria parasite, and this has been particularly altered so that the nuclei, nuclear stain, the nuclei showing up blue and the apicoplasts, which we will see later on more details about, the apicoplasts of which there is one in each cell are stained green with green florescence protein, which they are actually producing. Slide 10 (05:03): This shows an early blood stage of P. vivax inside a red cell. Notice the red cell appears to be slightly stippled, as the parasite is in a small ring stage, with a strong nucleus and delicate ring, and the cell may possibly be slightly larger than the other cells, on average. Slide 11 (05:27): The blood stages of the parasite are responsible for all the pathology of malaria, and here we see the red blood cell containing a malaria parasite. And let's go through the points in the list on the right. We go to the nucleus, which is the green one there, that's the nucleus. The cytoplasm which is blue, the mitochondrion which is pink, the apicoplast which is gray. Notice that the apicoplast and the mitochondrion both have a little bit of green inside, the apicoplast has a circular bit of green and that is actually the circular genome, 35 KB, it's the genome of the apicoplast. And in the mitochondrion there's a linear genome, which is usually for the mitochondrion, which is about 6.5 KB. And we also have a large nucleus which contains most of the genome of the malaria parasite. And also, to see the cytostome, cst, where the parasite actually engulfs lumps of the red blood cell content, in other words, largely hemoglobin which it's going to digest. Those lumps are actually transported to the digestive vacuole, DV, which is brown, where they are going to be digested and the hematin from the hemoglobin is going to be combined together into a crystal, the hemozoin crystal. Slide 12 (06:56): Here we see a late dividing stage of P. vivax, and you can see here that the stippling on the red cell is basically visible outside the parasite. The parasite is the more darker blue area inside the cell. Inside the blue area we see some brown spots, and those are spots of malaria pigment, hemozoin. The stippling on the red cell is called Schuffner's dots, just to make things confusing. Anyway, this is the pre-schizont stage of P. vivax, and the red cell is quite appreciably enlarged in this stage. This is characteristic of this particular species. Slide 13 (07:38): Here is the slightly later stage, where the red cell is still enlarged and slightly pinkly-stained, and the parasite has divided up into at least 16 different merozoites, which shall go out and burst into and infect adjacent red cells. Slide 14 (07:56): When the red cells burst and release the merozoites, they usually do it in relative synchrony during the day or night, and they tend to produce a rise in fever. So, if we look at these three different fever charts in this slide, we see one that has a peak every day, which is called the sub-tertian or quotidian fever, that's an everyday fever. And then we have one that's called the tertian fever, which is every third day, and one that is called the quartan fever, which is every fourth day. And these are related to the species of malaria, but not completely tied in. In the case of the sub-tertian or quotidian fever, you do get this in P. falciparum, although the actual cycle is 48 hours, because you may get several broods of the parasite in the same infection. In the case of the tertian fever this can represent falciparum, vivax or ovale because they all have 48 hour cycles of development in the red cell. In the case of the quotidian fever this is usually P. malariae, because it has a 72-hour cycle in the red cell. Slide 15 (09:13): After several blood cycles the merozoites from the schizonts in the blood, some of them actually become sexual stages. They get into a red cell and they don't divide and they just enlarge. In the male ones, the nucleus tends to enlarge, and in the females the cytoplasm tends to enlarge a lot and the nucleus remains compact. Both of them produce malaria pigment in the cytoplasm, and it is generally rather scattered. The scattering of the pigment will help you to tell the sex of the gametocytes. And these are waiting in the blood until a suitable mosquito should bite, so that they can infect it. Slide 16 (10:01): In this case we have some P. falciparum gametocytes, which are unusual in being sausage-shaped or banana-shaped. We can see again that there are two different types; on the left we have ones with blue cytoplasm and pigment rather clustering around the nucleus in the centre, and on the right we have a lighter colour more pink cytoplasm and the pigment actually more scattered around a larger nucleus in the centre. All those features are features which distinguish male from female gametocytes, if you think it's important, that is. Slide 17 (10:46): And then, when these get into the mosquito on a feed, then the maturation of the gametes, the fusion of the gametes and sporogony, that is the development of these sporozoites, will take place in the vector. Slide 18 (11:00): First of all, the male gametocytes will round up, and then divide into eight. Eight motile spermatozoa really, basically male gametes, and the females tend to round up and become rather smaller. And the male gametes search for the females in the stomach contents and fuse with them and produce a zygote. After the zygote is produced there's a series of divisions which follows a meiotic division, and so the rest of the cycle is again haploid. Slide 19 (11:38): The product of the zygote is the ookinete, or motile egg, which actually swims in the content of the stomach and actually burrows its way out of the stomach into the hemocoel of the mosquito, and it then cysts on the other side of the stomach as a little globule which actually divides and develops about a 1000 sporozoites infective stages in the globule oocyst. Slide 20 (12:08): This shows a mosquito stomach which has been dissected out, showing the individual globules of the oocyst of the stomach wall, and this is an experimental model so it's probably a rodent malaria. And there's a lot of globules there, a lot of oocysts on the stomach. In fact, when you're dissecting mosquitoes in the wild, you only find two or three or half a dozen, and you feel very lucky if you find half a dozen. And then, when these are mature, they actually burst and release the sporozoites into the hemocoel, and they actually circulate through the hemocoel and end up in the salivary glands, waiting to be injected to a human being. Slide 21 (12:55): And this shows a section of the salivary glands, showing the duct to the bottom, where the saliva goes and the sporozoites, these sort of lance-like things in the cells of the salivary glands, they are waiting to go into the saliva and be injected into a human being. Slide 22 (13:14): So, when the female Anopheles mosquito bites between dusk and dawn it injects the infected sporozoites in saliva when it's taking blood. Slide 23 (13:25): This picture shows an outline of the whole cycle, from 1-11, and I don't think I'll go into it in detail, but just to notice that when the sporozoite goes into the liver, that's between 1 and 1a and 2, we find that some of the sporozoa (it's from vivex and ovale) will actually just sit in the liver, round up in the liver cell, and sit there and wait, and they're called the dormant stages or hypnozoites, the sleeping stages which can actually wait for several months or even a year or several years before they actually start developing. Normally, the pre-erythrocytic stages develop by dividing as a liver schizont, that's number 2, and in only a few days they are ready to burst and the merozoites are released into the blood stream, that's between 2 and 3. And the merozoites individually go into red cells, and it takes, say, 48 hours for vivax or falciparum, to develop to a mature dividing stage in the blood cell, which will then burst and infect more uninfected red blood cells. After a few days of that we get production of gametocytes, that I've mentioned before, and they are then waiting be picked up by the mosquito. And the whole of the cycle is on that picture there, although not terribly clear, but I think it is all there. Slide 24 (14:56): Now malaria is studied so closely because it is a severe problem in the world. Every year there are at least 200 million clinical cases of malaria, and also about 2 million deaths from malignant tertian malaria, that's falciparum malaria, particularly in Africa. And the people who die are pregnant women and children. In the endemic areas, these are the at risk groups that are seen in West Africa they are the main at risk group for malaria and a lot of other things. In areas where there is less transmission, that is the epidemic areas, there are fewer infections but all age groups are at risk from the disease. Slide 25 (15:42): What makes malaria occur in a particular place? Well, basically, the things that control malaria are climatic and seasonal. Whether you are in a humid area and a warm area, which is ideal for mosquitoes to grow, and also the temperature needs to be ideal for the malaria parasites to grow in the mosquito. Mosquitoes won't bite if the humidity is below 45% relative humidity. So, humidity is absolutely vital to the transmission of malaria. So, people living up mountains, at the slopes of mountains, tend to have less malaria than people living on the plain. Slide 26 (16:27): The temperature of growth of the stage in the mosquito is quite critical, and it distinguishes between the vivax and the falciparum. The vivax is a temperate zone malaria on the whole, and P. falciparum is a tropical zone malaria. And the requirements of temperature for the growth of the parasite in the mosquito are quite critical; the falciparum needs a higher temperature, vivax doesn't need such high temperature. So, vivax could actually infect much of the UK in the past, falciparum, at the moment, could never do that, because of the temperatures. Anywhere where there's a good summer you can get malaria, generally. Anyway, the length of time the mosquito survives at the particular temperature is also critical, because if the female mosquito dies before it's able to pass on the infection to another person, then that particular malaria parasite is lost. And so, the time and temperature are very important and the ability of the mosquito to survive for more than 25 days, and from the point of view of the infection for more than 30 days if possible, then that is very important. So, a lot of factors affect the transmission of malaria from the mosquito point of view. Slide 27 (17:54): If we look at malaria in different parts of the world, we can contrast very crudely, stable malaria and unstable malaria. If we look, let's say, at West Africa, the stable malaria and unstable malaria for the borders of an island possibly, or even up a mountain, at the slopes of Mt. Kenya, for example, for the unstable malaria. The transmission of stable malaria is high every year, there's a lot of transmission every year. If we look down at the bottom of this chart, we see infective bites per year, that's the entomological infection rate, and for stable malaria this goes from about 10-300. Imagine how in 300 infective bites per year - that's amazing. It can even be more, to a 1000. If we have unstable malaria the transmission varies from year to year, the likely parasite will be, possibly in a temperate zone, is vivax, but you could have falciparum as well, but the most likely parasite will be seen to be vivax. The adults' immunity in the stable malaria area is high, and in the unstable malaria is low. Epidemics don't occur usually in stable malaria areas, because only parts of the population get seriously ill from malaria. And in the case of unstable malaria all the population may get malaria when there is an epidemic. Notice the infective bites per year in unstable malaria can be zero some years, and it can go up to 10. Usually, it's somewhere in between. The control of the infection, of actual transmission, is relatively easy, just a few precautions, in an unstable malaria area, but if we're talking about a stable malaria area then the control is extremely difficult and you have to attack the infection on several fronts. Slide 28 (20:02): Here's an example of the growth of the spleen, enlargement of the spleen, in children in a stable malaria area. Notice that from the left the spleens of these children have been outlined in white ink. That little arrow in the middle is the joining of the ribs and there is where the spleen is actually protruding below the ribs. Normally, this spleen can't even be felt or hardly be felt under the ribs. 50% of children aged 2-9 show a reflection of recent infective rates, in this particular case are actually showing splenic enlargement, and that is an area where malaria is probably transmitted year-round. You can also do a blood film and look at parasite positivity rate in children of the same age, and it gives you very much the same idea as the splenic enlargement of continuous exposure to infection. Slide 29 (21:09): Controlling malaria is really both controlling the transmission and controlling the morbidity, and starting off with the transmission, we need to know what mosquito is important for transmission. Slide 30 (21:25): Here's an example of a human biting catch, not very much done in our days, but is very valuable from the point of view of deciding which mosquito in a particular area are actually biting human beings. You sit out there with the lights, mosquitoes come along, and they may bite you. And the ones that are going to bite, you catch them and put them in a bottle and take them back to the lab, identify the species and also you can dissect them to actually see if you can find any malaria parasites in them. It gives you a very good idea of which mosquito is mainly responsible for transmitting malaria in the area. Of course this has been done in many areas around the world. Slide 31 (22:10): Here we have an example of mosquitoes which are breeding, they do have an aquatic larva stage, they are breeding in a rice paddy in southern China, and obviously the adult mosquitoes will come from that rice paddy in the local area. It is possible to attempt some sort of control by draining the water out of the rice paddy for weeks and letting it partially dry without damaging the rice. It can be done, it's a bit difficult and farmers will complain, but it can be done. I will go on to the next picture. Slide 32 (22:48): An example of urban malaria which is usual in India, A. stephensi will colonise and grow in water tanks on the top of high-rise buildings, and examining these is a very important job for people to do, because if there are mosquitoes in there they can actually expend polystyrene beads on the surface of the water to prevent egg laying and also to prevent escape of the mosquitoes that are already in the tanks. Slide 33 (23:21): In African situations, Anopheles gambiae is so prolific a mosquito that it will grow in a cow's footprint that's just filled with water in the rain storm during the afternoon, and that is practically impossible to prevent infection from that. There must be other ways of stopping the mosquitoes in those cases. Slide 34 (23:46): For instance, the adult mosquitoes, the adult females, when they come to feed, can you catch them, or can you kill them with insecticide? Yes you can, in the right cases. In the mosquitoes that will come inside houses to feed, in many cases this is the case, if you spray the inside of the house with a residual insecticide, and the mosquitoes that rest in the house during the day are actually likely to die or to live for a shorter time, and that will affect the transmission of the infection. It is also possible to prevent man-mosquito contact by using repellents and insecticide-impregnated bed nets, curtains and hammocks. Insecticides are generally repellent as well as insecticidal, and so the bed nets and the hammocks will repel the mosquitoes as well as killing any mosquitoes that are so keen as to come and sit on the bed net. Slide 35 (24:47): Here's an example of pyrethroid-impregnated bed-net preventing man-mosquito contact after dusk. This person is going to get into the bed-net before the sun goes down, and be protected from being bitten. Health education in local areas is extremely important aspect of malaria control. Everybody should know about the malaria cycle and how it's organised and ways of stopping it. Slide 36 (25:19): To control morbidity and mortality we need to have accurate diagnosis of malaria and suitable drug treatment. And diagnosis and drug treatment are absolutely crucial to getting malaria under control. Slide 37 (25:37): For diagnosing malaria blood films have been used for many many years, although they are being replaced by rapid diagnostic tests. They are much cheaper, but they do demand an extremely skilled technician to read them and to make. And the thick blood film is used routinely in the field because it's more sensitive, at least ten times more sensitive than the thin blood file, although the thin blood film is better for actually working out which parasite it is of the four species or so you've got available, because of the morphology of the parasite is in a better condition on the thin film. Slide 38 (26:15): How does the infection persist? The only reservoir is man, because mosquito populations lose their infections within weeks in the absence of infective humans. The persistence in man depends on the persistence in the blood stage for P. falciparum up to 3 years and in P. malariae more than 50 years, or in the case of P. vivax and P. ovale, it can be persistence in the liver stage as the dormant hypnozoites. Dormant parasites may remain in the liver for up to 5 years before developing and infecting the blood. Slide 39 (26:53): Some of the sporozoites that are injected by the mosquito, they are destined to become dormant, and they sit in the liver for 3, 6 or more months up to 5 years, then they form a schizont and infect the blood as usual. The timing of the relapse in P. vivax and P. ovaleis strain specific, and varies with different climatic patterns. Slide 40 (27:18): Here are examples of sections of liver showing, on the right, a pre-erythrocytic schizont, that is the ordinary dividing stage after the sporozoite has infected the liver, and on the left of the right-hand side of the slide, there is a single hypnozoite, very small, showing up under the same florescent antibody system as the actual developing stage. This is a dormant stage. And, on the left we have a dormant stage which has been stained with florescent antibody and then the top stained with an ordinary gleam staining, showing the size, it's about 3 or 4 microns in diameter. Slide 41 (27:59): Malignant tertian malaria is characterised by the blood stream infection, apparently from the peripheral point of view, from the finger or the arm, in mainly young stages of the parasite, and later on in patients' gametocytes. But the later stages of the parasite don't generally circulate through the whole of the body. Slide 42 (28:22): Here we have a late stage which in a particular patient has actually come out from inside and actually circulating, and it shows the presence of malaria pigment, black/brown pigment hemozoin, formed in the digestive vacuole of the parasite by dimerisation of toxic hematin released during hemoglobin digestion. Slide 43 (28:47): Look at this slide here, shown on the left are parasitized cells called the hemozoin, and electron micrograph of it on the right, inside the vacuole, showing the hemozoin crystals inside the digestive vacuole. Slide 44 (29:05): Now, the anatomy of hematin is capable of generating free radicals and damaging membranes, which can damage the parasite when it's released in the digestive vacuole. Mammalian cells which digest hemoglobin actually also split up the hematin ring and let out the iron and produce what is known as a bile pigment, whereas the malaria parasite can do it to detoxify the hematin. So what it does, in the next slide, we see two of those molecules of hematin are actually fused together. Slide 45 (29:40): This shows how the iron atom from one of the molecules is detoxified by oxygen from the other molecule, and vice versa. So, two iron atoms are actually detoxified by this dimerisation process. Slide 46 (29:58): Another instance of seeing the hematin in infected red cells is looking at sections of brain post-mortem. We see here a large vessel and then some capillaries leading off it. And what we see, in the vessel there's a good proportion of normal red cells, but in the capillaries the infected red cells have actually been bound to the endothelium of the lining of the capillaries so that those vessels appear to be blocked up with infected cells. This is the latest stage of the P. Falciparum parasite, when it is developing merozoites and these are going to burst and release the merozoites to infect new cells. Slide 47 (30:45): In falciparum malaria, binding of CD36 and ICAM1 type adhesion proteins on the vascular endothelial surface to molecules from parasite such as PfEMP1 expressed on the infected red cell causes cytoadherence with pathological effects related to localised slowing of the circulation. Production of TNF and lactate and increased expression of binding factors themselves, that is, CD36 and ICAM1. Cytoadherence functions to reduce the exposure of red cells containing later division stages to the splenic circulation, where they would be recognised and removed, as well as specific organ-damaging effects of cytoadherence on kidney, brain, lungs and intestine due to alterations in microcirculation, the higher parasitaemias developed are damaging in themselves. Cytoadherence leads to higher gametocytaemias, because of the higher parasitaemias, probably enhancing transmission efficiency for the parasite. Slide 48 (31:52): This picture shows a diagrammatic form of the infected red cell, with knobs on the surface containing PfEMP1, which are actually binding to CD36 on the vascular endothelial cells in, for instance, the brain or the kidney. Slide 49 (32:10): In pregnancy malaria can cause abortion, an exacerbate anaemia, and is associated with low birth weight. Though semi-immune before hand, and able to pass on this immunity to the neonate, the pregnant woman loses much of her immunity to malaria. Clones of parasites which express chondroitin sulfate binding proteins on the red blood surface accumulate in the maternal circulation of pregnant women, and they accumulate particularly in the maternal circulation of the placenta where they cytoadhere. In subsequent pregnancies there isn't such a problem with malaria in pregnancy because the women are getting immune to the parasites expressing chondroitin sulfate. So, second pregnancies or third are not so badly effective by malaria as firsts. Slide 50 (33:03): Human beings have been exposed to malaria for many years, and 5% of them carry abnormal hemoglobins with changes in the α- or β- globin genes, which have been selected through generations of malaria exposure. Sickle cell haemoglobin, for instance, is a balanced polymorphism where HbS, the sickle cell haemoglobin, has a sequence change from Glutamate to Valine which causes crystallisation at low oxygen tension. People with homozygous sickle cell haemoglobin suffer from sickle cell anaemia, and they may even die before puberty in endemic areas, but people that are heterozygotes are actually protected from the adverse effects of falciparum infection, probably due to enhanced recognition of infected cells by the spleen. Slide 51 (33:58): Red cell enzyme deficiencies also occur, which are linked to protection from malaria. Glucose 6-Phosphate Dehydrogenase deficiency, an X-linked character, causes a defect in reduction of methemoglobin which particularly affects infected red cells. G6PD heterozygotes and hemizygotes, they all be male, are relatively protected from severe effects of malaria. There are red cell membrane protein polymorphisms related to the proteins, the variant glycophorins En(a-) and S-s-U protect because they are used by P. falciparum for invasion. The duffy binding factors Fya and Fyb, which are rare in West Africa, are required for P. vivax invasion, and P. ovale replaces P. vivax in Western Africa because of this factor. Changes in band 5 in the red cell membrane lead to ovalocytosis which protects against cerebral malaria, and this is seen in New Guinea. Slide 52 (35:02): Sterile immunity to malaria is almost never found. Immunity to symptomatic infection is species specific, and develops during childhood in endemic areas. This clinical immunity has to be maintained by constant exposure to infection, and develops more rapidly in more highly endemic areas. The adult population, even when asymptomatic or even aparasitaemic, is an important reservoir for transmission. There is antigenic diversity in the multiclonal population, and antigenic variation, that is of PfEMP1 for instance, within clones. 2% of parasites in vitro will change their RBC surface phenotype every 48 hours. Back Transcript Malaria is produced by infection with single celled parasites belonging to the genus Plasmodium. This disease, which affects up to 100 million people, killing 2 million each year is a major pathogen in tropical regions, where it is spread by biting insects. There have always been sporadic cases in temperate regions in travellers returning from the tropics, and sometimes near airports when infected mosquitoes have been transported on aeroplanes. However, with gradual changes occurring in the global climate, the range of the mosquito vector is now extending, so that areas of Southern Europe and the United States, which have been free of the disease for many years, are once again vulnerable. Malaria is an example of a disease which is transferred between individuals by an insect vector; in this case mosquitoes of the genus Anopheles. The life-cycle of malaria is complex, with different stages occurring in the mosquito, and in the blood and the liver cells of the infected individual. The complex life cycle presents a moving target to the immune system, which uses different types of immune defence against different stages of the parasite's life cycle. To consider the immune responses against malaria, we must first look at the different stages of infection. If an individual is bitten by a female mosquito carrying the parasite, sporozoites will be injected into the blood stream, and will find their way within a few minutes to the liver where they infect cells of the liver called hepatocytes. If the person already has antibodies present which react against sporozoites, then this initial transfer through the blood may be intercepted. Notice that at this stage the sporozoites are extra-cellular and therefore susceptible to antibodies. Once the sporozoites enter the hepatocytes they undergo a period of rapid division, so that each infected cell can produce up to 30 000 progeny over a period of 5-10 days. During this time the parasites are inaccessible to antibody, but cell-mediated immune mechanisms may be able to destroy infected cells. After division in the hepatocyte the progeny are released once more into the blood stream as the cell ruptures. However, at this stage the parasites have changed their form and are now called merozoites. Merozoites attach to, and infect red blood cells, in which they undergo another programme of cell division, so that each infected red cell can produce 16-32 merozoite progeny. These are released into the blood plasma by rupture of the red cell, where they can once more infect fresh red cells. The waves of infection in the blood are associated with the fever and debilitating symptoms of the disease. Repeated rounds of infection are typical of malaria. While the parasites are within the red cells they are not directly accessible by antibodies. However antigens produced by the parasites can be inserted into the plasma membrane of the red cells and these are accessible by antibodies in the plasma. The surface antigens are potential targets for antibody-mediated cell mediated cytotoxicity effected by large granular lymphocytes also called killer cells. The surface antigens may also act as targets for antibody and complement-mediated lysis of the red cells. One species of Plasmodium, called Plasmodium falciparum is particularly dangerous since infected red cells can attach to the endothelial cells which line the blood vessels in the brain. This causes blockage of the blood supply within the brain and local inflammation. The attachment is mediated by antigens of the parasite inserted into the infected red blood cell. This type of malaria is called cerebral malaria and it is associated with high mortality. During the infection of the red cells, some of the parasites develop into the sexual stage of their life cycle called gametocytes. These are different again from previous forms, and may be taken up from the blood when another mosquito feeds on the infected person. Antibodies to the gametocytes can also interfere with the parasite's life-cycle, although they do not protect the host from the effects of the infection. It is interesting to note that antibodies from the plasma, taken up by the insect as it feeds, can still inhibit parasite development even within the gut of the vector. The complexity of the malarial life-style means that the immune system must deploy defences against both intracellular and extracellular stages of the disease. Moreover parasite antigens vary from one stage to the next - some are present on several stages, while other antigens are specific for one stage, and sometimes only for a particular phase of that form. The micrographs here show three different stages of the malarial life cycle within red cells, which are called ring-forms, trophozoites and schizonts. The gel shows proteins from these three different stages. Each protein appears as a stained band; some proteins, such as the one labelled '6' persist throughout all three stages, but others such as number 7 at 46 kiloDaltons are only present on the schizont stage, just before the red cell ruptures. This means that antibodies which are effective against one stage may be ineffective against others. Many of the antigens of malaria are proteins which have segments containing numerous repeats of 4 or 5 amino acid residues. An example of this is the circumsporozoite antigen or CS antigen, which is present on the surface of the sporozoite. This antigen is involved in the initial attachment to hepatocytes, and antibodies against it can limit the initial infection. When sporozoites are treated with serum from an immune individual, the CS antigen is shed from the cell surface. It appears that the parasite is releasing its surface antigens, in order to act as a decoy - antibodies in serum will bind to shed antigen and then be unable to engage the parasite itself. At first, it is difficult to understand what particular value repetitive antigens would have to the parasite. Possibly they present the immune system with a large antigenic load, which it finds difficult to cope with. Certainly the antigens released during a malaria infection, contribute to secondary damage within the host. For example if immune complexes formed of antigens and host antibody are formed in large amounts, they tend to become deposited in the kidney. You can see this in the immunofluorescence picture, which identifies IgG in the glomerulus of a patient with late stage malaria. Phagocytes attempt to clear these deposited complexes from the kidney. Although they clearly do not succeed fully in the attempt, they do produce a deal of colateral damage caused by their armoury of cytotoxic molecules, which would normally be used to kill the parasites. This is an example of an immunopathological reaction where the immune response itself produces the damage. Let us return to the cell mediated immune response. It has been thought that CD8+ T cells could recognise parasite antigens presented by MHC class I molecules, and therefore kill parasite-infected hepatocytes. This would be important in protection against the earliest stage of infection when the parasite develops inside the cell. It is clear that CD8+ T cells are important since animals which lack class I molecules due to a genetic deficiency are not protected by immunisation with sporozoite vaccines. However it appears that CD8+ T cells act primarily by the production of cytokines and not by T cell mediated cytotoxicity. For example, the most effective vaccines against malarial sporozoites induce high levels of cells secreting interferon-γ, including CD4+ T cells as well as CD8+ T cells. Unfortunately however the immune response is strain-specific and numerous strains of the parasite occur. There is also evidence from work on mouse malaria that the interferon-γ activates macrophages to produce inducible nitric oxide synthase thereby generating nitric oxide and peroxynitrites which are toxic for the parasites. Further evidence that T cells are important comes from genetic studies. In West Africa, for example the MHC class I allele HLA-B53 and an independent class II allele are both associated with protection against severe disease, implying that CD8 and CD4+ T cells respectively both contribute to protection. The graph compares the frequencies of the HLA-B53 allele in normal subjects and those with mild or severe malaria in the Gambia. HLA-B53 is under-represented in individuals with severe malaria indicating that it could be protective. 43 other alleles showed little or no such protection. Interestingly, this allele is common in areas of the world where malaria is endemic, and uncommon in other areas, as indicated on the map. This implies that malaria has acted as a selection factor to increase the frequency of the protective gene in endemic areas. Different studies using pairs of twins also imply that disease severity and immune responsiveness to malaria has a major genetic component, even though host genetic variation has little effect on the initial susceptibility to infection. So we can see that protection against malaria involves both cell-mediated and antibody mediated immune defences, which act at different stages of the parasite life-cycle. The level of the immune response depends on the genetic make-up of the individual and also depends on the species and strain of the parasite.

Protozoa

Single-celled protists that have 'animal-like' cells and are unable to photosynthesise. They have one or more nuclei and are usually classified according to their cell structures.

Sporadic

Cases of a disease that are isolated and unpredictable in their occurrence.

Trypanosomiasis

Disease caused by members of the genus Trypanosoma, including sleeping sickness in Africa, and Chagas' disease in South America. The human infected trypanosomes in Africa are highlighted in yellow here belong to the genus Trypanosoma; subgenus, Trypanozoon. This subgenus, otherwise known as the 'brucei' group after the type species Trypanosoma brucei brucei, also include species that only infect animals. The actual taxonomic status of Trypanosoma brucei brucei T. b. rhodesiense and T. b. gambiense relative to one another has proved difficult to determine but data from various analytical techniques since the 1980s such as isoenzyme electrophoresis and various genetic techniques outlined by Gibson in this 2007 Paper now suggest a resolution. Isolates of T. b. brucei from animals in East Africa and T. b. rhodesiense from humans although genetically heterogeneous are clearly more related to each other than to T. b. gambiense. Historically East African T. b. brucei isolate that is human infective or has shown human serum resistance on testing has been designated T. b. rhodesiense whereas non-human infectivity or serum sensitivity has meant a T. b. brucei designation. Trypanosoma brucei brucei isolates fail to survive incubation in human serum whereas T. b. rhodesiense and T. b. gambiense isolates do. The identification of the SRA serum resistance gene in T. b. rhodesiense isolates by de Greiff and colleagues in 2007 and the fact that [a] this gene is not expressed constitutively and [b] it could be transferred to another genotype as a result of genetic exchange almost certainly makes the separate designations T. b. brucei and T. b. rhodesiense unsustainable. T. b. gambiense is genetically more distinct and falls into two groups. Group One a group genetically distinct from T. b. brucei and T. b. rhodesiense causing classic Gambian disease. And Type Two covering a small group of isolates from Ivory Coast running a more acute disease course. Disease caused by members of the genus Trypanosoma, including sleeping sickness in Africa, and Chagas' disease in South America. T. b. rhodesiense has been found principally in East and previously also Southern Africa and is associated with an acute disease course, untreated cases dying in weeks or months. Whereas T. b. gambiense is found in West and Central Africa and is associated mostly with a disease running a more chronic course. Untreated cases surviving a number of years. Infection results in fever and tissue wasting initially before the trypanosomes cross into the central nervous system causing an alteration in sleep pattern from where the disease 'Sleeping Sickness' gets its name. And later psychological changes and eventually coma and death. The tsetse transmitting human infections in East Africa are Savannah flies such as Glossina morsitans and here the disease operates as with game animals particularly large antelope species such as the waterbuck shown here acting as reservoirs of infection. T. b. gambiense in West and Central Africa is transmitted by forest flies such as Glossina palpalis gambiensis. And although the occasional animal infection has been identified the disease appears primarily to be transmitted by the flies between humans. No vaccines are available and drug treatment is problematic as the drugs are few, frequently toxic when and if they are available which itself is a problem. Work is now underway in various laboratories to find new drugs but no new compounds are as yet anywhere near clinical trials. As with many other diseases then diagnosis is better early as the drugs for use before CNS involvement are less toxic. A diagnostic serological test the CATT test for T. b. gambiense was developed in the late 1970s and 1980s based on the detection of antibody to a commonly occurring surface antigen from trypanosomes from clinical cases. Fixed stain trypanosomes are used as antigen and when mixed with the patients serum containing antibody a glutination occurs as can be seen on this next slide. Transcript Slide 1 (00:00): African trypanosomiasis, listed by WHO as one of the neglected tropical diseases, currently appears even in endemic areas to have a low incidence compared with other diseases on the WHO list with current estimates of perhaps ten to fifty thousand cases annually. But its incidence over the last century has varied considerably with a few large epidemic outbreaks, the last being in the 1990s affecting Angola, the Democratic Republic of Congo, Uganda and Sudan where anywhere between forty thousand and half a million people may have been infected. Slide 2 (00:48): The African trypanosomes are like the South American trypanosomes and the kinetoplastid protists - that's to say single celled flagella organisms. They are internal parasites of a range of vertebrates hosts including humans and are transmitted by biting flies. Slide 3 (01:17): Slide 6 (05:14): Slide 7 (05:38): Slide 8 (05:49): Slide 9 (07:19): Slide 10 (07:52): Slide 11 (07:57): This test is easily carried out in temporary field laboratories as here in a village in South West Cameroon. And large numbers of individuals can be screened during the course of a day as in this school again in South West Cameroon. Other forms of more labour intensive diagnostic test are in use also. With no vaccines and limited drugs one very practical and sustainable control strategy is to interrupt transmission by reducing vector numbers. This can be done very effectively by the use of fly traps which are relatively inexpensive and can be made locally. The flagellate trypanosomes are elongate, varying between 15 and 30 microns in length. The flagellum comes onto the surface of the trypanosomes via an in tucking in the surface membrane - the flagella pocket. And then runs along the surface of the cell connected to it by an undulating membrane. The extent of the flagellum varies with species and life cycle stage. In the infected host the trypanosomes swim freely in the blood and lymph and penetrate various tissues but remain extra cellular. The upper panel here illustrates the various kinetoplastid lifecycle stages although only the trypomastigote and the epimastigote are represented in the genus trypanosomer. In brucei group trypanosomes the blood stream trypomastigotes exhibit pleomorphism varying from long slender to short stumpy forms, shown diagrammatically in B-D of the lower panel of this slide. The slender blood stream trypomastigotes proliferate by a process of asymmetric binary fission. You may have noted that with one of the trypanosomes in that earlier blood fill. Some eventually transform via intermediate forms to short stumpies but no longer divide and have pre-adapted metabolically to life in the tsetse should a blood meal be taken. In the long slender forms, seen at three o'clock on this life cycle diagram, a single tubular mitochondrian essentially devoid of crystii shows no respiratory oxidative phosphorylation activity whereas the mitochondrian in the short stumpy form at five o'clock has developed extensive crystii and electron transport system enzymes are expressed, presumably to increase the efficiency of use of the glucose respiratory substrate in the ... Should the trypanosomes be taken up by tsetse in a blood meal the bloodstream trypomastigotes transform to procyclic trypomastigotes in the fly mid gut and again start dividing prolifically by asymmetric binary fission. After a varying period of time according to the trypanosome species procyclics migrate to the salivary glands where they transform via an epimastigote state into metacyclic trypomastigotes, the infective form, injected into a new host by the tsetse at a further blood meal. There are further important lifecycle associated changes to refer to and I will come back to these as we discuss the course of infection in the vertebra host. As you can see represented on this slide soon after infection as a result of tsetse bite trypanosomes appear in the blood and as the trypanosomes divide the numbers increase. But after a few more days the numbers in the blood crash and fall below the threshold of detection by conventional wet film microscopy and then a few days later again the trypanosomes can once again be detected in the blood and numbers increase. The paucity of detectable number of trypanosomes in the blood makes direct diagnosis of infection problematic especially with gambensi where parasitemias are much lower than with rhodesiense. This fluctuation in numbers continues for the duration of the infection. So what is the explanation? It's the organisms profound ability to evade the host immune response. The parasite surface is covered by a layer of around ten to the seven identical antigenic protein molecules an antigenic coat the parasite can change. We call these proteins variant specific glyco proteins VSGs so depicted here very diagrammatically, the coloured peaks representing the fluctuating parasitemia. Parasites with the type A antigenic VSG coat grow up in the first parasitemic peak provoking an anti A antibody response. The antibodies bind to the parasite with observations suggesting clearance of the antibody coated trypanosomes from the circulation by fixed tissue macrophages primarily in the liver. A few days later in this example trypanosomes reappear in the circulation. Anti A antibodies are still present but ineffective as the parasites now present are expressing VSGC. A new antibody response is initiated eventually clearing Type C for these to be replaced here by Type F and so on. It seems that before each antibody clearance some of the trypanosomes start to express a different VSG and as the antibody clears the major population so the new minor population grows to replace it. In this transmission electron micrograph of the brucei group trypanosome surface you can see the electron dense VSG surface coat overlying the phosphor lipid cell membrane. The VSG molecules ... for insertion into the surface membrane via a glycosylphosphatidylinositol GPI anchor on each VSG molecule with Myristic acid chains inserting into the phosphor lipid of the membrane. Brucei group VSG molecules have a molecular size around 55 Kilodaltons comprising over 400 amino acids. The end terminal sequence of amino acids some 350-400 amino acids varies between antigenically distinct VSGs while the C terminus around 50-100 amino acids shows more homology with the sequences falling into 3 VSG groups. Despite low amino acid sequence homology 2 VSG monomers subjected to X-ray crystallographic analysis showed remarkable 3 dimensional structural similarity. In these ribbon diagrams the red coloured areas highlight the principle structural differences between the two presumably reflecting antigenic differences. Returning to the life cycle only the metacyclic and blood stream trypomastigotes express a VSG coat. Once taken up by the tsetse during a blood meal the trypanosomes lose their VSG coat replacing it within hours by a new stage specific coat comprising two different classes of the protein procyclin the EP and GP EET procyclins characterised by internal dipeptide and pentapeptide repeats respectively. Recent experiments by Isabel ... and colleagues using deletion variants that did not express procyclins determined that while procyclins are not essential for fly transmission prevalence and intensity of saliva gland infection was much reduced in the absence of expression. Other experiments over recent years have established that trypanosomes are capable of genetic exchange at some point during the tsetse fly stage. VSG surface expression is reacquired at the metacyclic trypomastigote stage in the tsetse salivary glands unlike the perhaps nonessential nature of the procyclin code referred to previously. Trypanosomes not expressing VSGs do not survive in the blood. Early experiments suggesting that such trypanosomes would be lysed as a result of complement activation by the alternative pathway and clearly as detailed before they have a major role in immunivation involving antibodies. So what is the mechanism bringing about antigenic variation - VSG variation - in African trypanosomes? It appears that T. b. brucei has somewhere of the order of 1000 to 1500 VSG gene alternatives in its genome although many of these are incomplete that is to say pseudo genes. Slide 22 (19:09): To understand the mechanism of VSG change we need to look at the trypanosome genome. Trypanosome chromosomes cannot be directly visualised in for instance a chromosome spread. The chromosomes do not condense at division. But it is possible to visualise chromosome sized DNA by pulsed field gel electrophoresis. The trypanosome is deployed with respect to a set of eleven paired megabase sized chromosomes perhaps .5 to 6 megabases in size on which resides structural and housekeeping genes as well as VSG genes. T. b. brucei trypanosomes also have several hundred unpaired mini chromosomes in the size range 50 to 150 kilobases on which reside VSG genes plus a few less than 10 additional unpaired chromosomes of a size intermediate between the large megabased chromosomes and the mini chromosomes. VSG genes are also found on these intermediate sized chromosomes. T. b. gambiense however has many fewer chromosomes and therefore its VSG repertoire seems less extensive than other bruceis. Slide 24 (20:47): VSG expression itself is controlled by expression sites found near to the ends of the megabase and intermediate sized chromosomes. A VSG gene must occupy an expression site for an expression to occur whereas the majority of VSG genes occur in arrays at internal sites on the chromosomes. Estimates of the number of subtelomeric expression sites present in the genome are given as between 10 and 20 with only one being active in VSG expression at any one time. The lower diagram B in the top panel here shows the organisation of a so-called blood stream expression site facilitating the expression of VSGs in blood stream trypomastigotes from the promoter towards the left hand end to the telomere at the right hand end. The orange coloured boxes represent the VSG gene in residence in the expression site under the influence of the promoter represented by the black arrows some 40 to 60 kilobases upstream. Immediately upstream of the VSG gene are a series of non-coding 70 base pair repeats and between these repeats and the upstream promoter are a series of other genes whose expression is under the influence of the same promoter so-called polycistronic transcription. In the smaller lower panel on this slide you can see part of an array of VSG genes in a chromosome internal array with a few 70 based pair repeats 5-prime to each VSG gene. Only about 5 to 10% of the VSG genes in the internal arrays appear to be complete VSG genes. The rest are pseudo genes. The VSG genes cannot be expressed directly from their position in the arrays. The predominant mechanism of VSG expression and VSG variation appears to involve recombination within an expression site. Slide 25 (23:20): Let's consider a VSG gene. If we look at the yellow box in the example on the left this VSG gene chosen for expression here from an internal array in a manner completely unknown is first copied including some or all of the upstream 70 based pair repeats. The copy is then moved to a subtelomeric expression site and replaces the resident VSG gene. The displaced VSG gene is lost although the ability to re-express that VSG gene is retained as the gene occupying the expression site is only a copy with the original VSG gene retained elsewhere on one of the chromosomes. Subsequently the green gene replaces the yellow and then the mauve replaces the green. Recombination appears to occur within the 70 based pair repeats 5-prime to the VSG gene and probably within the conserved 3-prime end of the gene. This cassette recombination process is also known as gene conversion and is also seen in yeast. Slide 26 (24:41): Some expression sites appear to have few if any 5-prime 70 based pair repeats. And the VSG genes resident there A appear fixed and B appear to be those expressed by the metacyclic trypomastigotes in the tsetse salivary glands. As there are a number of these so called metacyclic expression sites reactivation of VSG expression in the metacyclic trypomastigotes results in a mixture of VSG variant antigen types in the salivary glands. With regard to bloodstream forms early in infection it seems to be that complete VSG genes are copied and transferred to expression sites but later more complex gene recombination occurs with all or parts of pseudo genes transferred to generate complete mosaic genes - complete VSG genes capable of expression. So a process of segmental gene conversion. So-called transcriptional switching may also occur whereby the active promoter on one chromosome is turned off and one on another chromosome is turned on. At least one suggestion in relation to this mechanism is that it's more about switching to a different set of expression site associated genes --ESAGs - rather than a different VSG. But that a VSG switch occurs as a consequence of it. The VSG molecules coat the entire surface of the trypanosome including lining the flagella pocket. But in the membrane of the flagella pocket are a series of other surface proteins that are coded by the ESAGs located between the promoter and the VSG expression site. The function of all of these ESAG products is not yet confirmed but a number appear to be flagella pocket proteins that act as receptors in receptor-mediated endocytosis for essential nutrients such as transferrin. ESAG 6 and 7 together code for a transferrin receptor and different versions of these genes occur on different chromosomes with some transferrin receptor protein versions better able to bind the transferrin of one host species compared with another. So transcriptional switching referred to previously will allow for instance the expression of a different transferrin receptor. Slide 27 (27:45): Gene conversion is the major mechanism underlying VSG expression and variation in bloodstream forms and appears to be hierarchical with telomeric VSGs expressed earliest in infection followed by the expression of intact VSG genes from internal arrays and finally recombined pseudo genes. However other processes resulting in a variation of VSG expression seem also to be able to occur. A further recombination mechanism telomere exchange may occur bringing an inactive expression site on one chromosome into the influence of an active promoter on another. And as I mentioned previously it appears that promoter switching can also occur so an expression site with a resident VSG gene can become inactive and another expression site with its resident VSG can become activated so-called in situ switching. Slide 28 (28:54): Returning to VSG expression site control only one expression site is active at any one time. And the mechanism or mechanisms of the control afis are still the subject of much research. Navarro and Gull have described within the trypanosome nucleus an extra nucleolar body ESB - arrowed in the photographs. An ESB containing the pole 1 RNA ... enzyme involved in the transcription of VSGs. In the nucleii of other cells Pole 1 appears only to be associated with Ribosomal DNA expression in the nucleolus. Navarro and Gull propose a model whereby VSG expression requires a chromosomal expression site to become associated with the ESB and that only one expression site may associate at a time thus providing the required control. Slide 29 (30:01): A further gene located in the region of the expression site and the ESAGs in T.b rhodesiense is the SRA serum resistance gene or genes as there are two versions. I referred to the SRA gene earlier. The SRA gene is a truncated VSG molecule and is expressed in the lysosome or membrane of human infective T. brucei isolates. By definition then T. b. rhodesiense. The SRA gene is not in all expression site locations on all chromosomes and therefore may not be expressed at all times depending on which expression site is in use. The SRA protein in T. b. rhodesiense binds ApoL1 - the human serum molecule that brings about serum sensitivity in T. b. brucei thereby preventing its activity. T. b. gambiense is also serum resistant but based on an unknown mechanism not involving SRA genes. Slide 30 (31:18): Overall the trypanosome's complex genome that provides the organism with an incredibly powerful facility to vary its surface antigen provides a molecule that allows survival in human serum and allows genetic exchange between trypanosomes with different genotypes makes this organism a formidable disease causing adversary. The disease causing organisms, the trypanosomes are transmitted by the bites of the tsetse fly - genus Glossina. Trypanosomes are unicellular animal-like cells with flagella, and cause trypanosomiasis. Unlike Giardia, these parasites have a complex life cycle involving more than one type of host and an insect vector. Trypanosomes cause a chronic disease that is commonly known as sleeping sickness or human African trypanosomiasis (HAT) in Africa, and Chagas' disease in tropical and subtropical Central and South America. These two different diseases are considered in more detail in Sections 2.3.1 and 2.3.2. Trypanosomes exist in different morphological forms in the definitive host and the vector. As can be seen in Figure 6.2 the position of the kinetoplast relative to the nucleus and the origin of the flagellum differ in each of the four morphological types. Typically, the trypomastigote form occurs extracellularly in both the definitive host and the vector (e.g. in humans infected with Trypanosoma spp.). The amastigote is an intracellular form occurring in the definitive host (e.g. in humans infected with Trypanosoma cruzi or Leishmania (see Section 2.4); while the promastigote and epimastigote are forms.

Hydatid cyst

Fluid-filled sac within a host tissue, formed by the oncosphere of a tapeworm of the genus Echinococcus. The sac enlarges gradually, producing protoscolices (juvenile tapeworm larvae) that are eventually released from the cyst.

Miracidium

Free-living motile form, covered with cilia, which penetrates or is ingested by the intermediate host (mollusc) to become a sporocyst.

Cholera

Koch showed that cholera was caused by a comma-shaped bacterium, subsequently referred to as a 'vibrio' and named Vibrio cholerae. Koch isolated the bacterium again in India in 1884, and showed that the bacillus lived in the human gut and was spread by dirty water. Cholera is typically an acute disease, with an incubation period of two hours to five days, but asymptomatic infections are common. Over 90% of infections are, in fact, very mild or moderate. Fewer than 10% of infected individuals go on to develop the typical disease, with its sudden onset of profuse watery diarrhoea, effortless vomiting and in some cases fever. Cholera toxin binds specific gangliosides in the membranes of gut epithelial cells (which makes the membrane more permeable to potassium (K+) ions), and activates membrane-bound adenylate cyclase. This increases cyclic AMP production, which causes efflux of chlorine (Cl−) and bicarbonate (HCO3-) ions and across the membrane and into the gut lumen, and prevents sodium (Na+) ions being reabsorbed from the gut lumen. The presence of these ions in high concentrations in the lumen 'pulls' water out of the surrounding cells by osmosis, and the result is copious amounts of water in the gut. Symptoms are profuse, watery diarrhoea known as rice-water stools,which are pale grey, flecked with mucus, and have a fishy smell. Can lead to severe dehydration and the loss of important electrolytes (Na+ and Cl− ions. Potassium (K+) and bicarbonate (HCO3−) ions are also lost in large quantities. Potassium plays a key role in skeletal and smooth muscle contraction. Thus, the loss of K+ ions results in cramps in the abdominal muscles, whereas a reduction in HCO3− ions can upset the pH balance of the body. The resulting severe dehydration causes the production of urine to cease, the skin to become wrinkled, and sometimes the eyes to appear sunken. The loss of fluid volume causes a drop in blood pressure and circulatory shock. Patient has reduced level of consciousness and become progressively weaker, sometimes to the point of death, within 12-24 hours. Cholera is caused by the Gram-negative, rod-shaped bacterium Vibrio cholerae, which on first isolation may appear curved. The bacteria have a single, polar (at one end) flagellum, which renders them motile. Based on the properties of their O antigens, more than 130 groups have been identified, but only two of them, O1 and O139 ('O' for O antigen), have been known to cause epidemics of diarrhoeal disease. Until recently, only O1 Vibrio strains were known to cause disease. These strains fall into two biotypes (or biovars), distinguished by their metabolic activities - in this case, their different haemolytic activity, relative resistance to the antibiotic polymyxin B, and their different susceptibilities to bacteriophage. The two biotypes are called 'classical' and 'El Tor'. The classical biotype is further divided into two serovars (also known as serotypes), based on the antisera that recognise them, and named after the place where they were first isolated: Inaba and Ogawa. Thus, any pathogenic strain of Vibrio cholerae has a name that reflects both the biotype and the serovar; for example, strain 569B has a classical biotype and the Inaba serovar. Until recently, humans were thought to be the only host for Vibrio cholerae, especially as there are particular populations in which cholera is endemic. However, in general the vibrios are aquatic bacteria, and studies have shown that V. cholerae is a normal member of natural aquatic environments, such as oceans, lakes, rivers, estuaries and brackish water. Here they attach to small (1-2 mm long) copepod crustacea (a major component of zooplankton), to other crustacea such as shrimps, to algae, and to the intestines of filter-feeding shellfish. Some of these species are natural reservoirs of V. cholerae that can be eaten by humans. In addition, fish that ingest these species can contain sufficient vibrios to cause disease when eaten by humans, so that properly cooking all seafood is essential in areas where cholera is endemic. Infection with the cholera bacterium occurs by the faecal-oral route, as a result of consuming food or water contaminated with V. cholerae. The infectious dose is high, requiring a minimum of 108 bacteria for classical V. cholerae in a healthy host, but figure falls when acid production by the stomach is impaired. The vibrios that survive passage through the stomach are able to multiply in the alkaline environment of the small intestine. In fact, the tolerance of cholera vibrios to alkaline conditions is one of their distinguishing features. Interestingly, people of blood group O appear to be more susceptible to V. cholerae infection, but the mechanism underlying this is not understood. Once in the small intestine, the vibrios must reach the epithelial cells that line it. They are probably aided in this task by their flagella, which help in propulsion, and by their ability to produce mucinase (an enzyme capable of degrading mucus). Another enzyme produced by vibrios is chitinase, which degrades chitin. Chitin is found throughout all kingdoms and is the main component of the cell wall of fungi and the exoskeletons of crustaceans. Whilst of no direct use in infecting humans, chitinase is used by V. cholerae to attach to the surface of aquatic crustacea and the radula ('tongue') of shellfish, and subsequently to digest chitin as a source of food for the vibrios. Actual attachment to the epithelial cells is mediated by a number of factors on the bacterial surface, the most important of which is a molecule called toxin-coregulated pilus (TCP), which is discussed later in this section. A number of haemagglutinins, fimbriae and the O antigens of lipopolysaccharide have also been implicated in the attachment process. The cholera vibrios do not penetrate the gut epithelium, but they release a potent enterotoxin called cholera toxin, CT, which is largely responsible for the symptoms of cholera. CT has an A-B structure comprising five B subunits and one A subunit, which is an enzyme. The B subunits mediate attachment to the villi of the epithelial cells by binding to the ganglioside receptor GM1, and the CT is taken up by receptor-mediated endocytosis, The A subunit is then released and goes on to catalyse the transfer of an ADP-ribose from NAD to the α subunit of a G protein. The modified G protein is no longer able to switch off the enzyme adenylate cyclase. Adenylate cyclase produces cyclic adenosine monophosphate (cAMP) from ATP as shown in the equation below: ATP → cAMP + PPi + H+ cAMP is a key regulatory molecule that carries information from extracellular signalling molecules such as hormones, to the cell's interior. For this reason it is called a second messenger. • What would be called the first messenger? • The hormone itself. The level of cAMP within a cell therefore influences its activity, but its precise role depends on the cell type. In a gut epithelial cell, cAMP level influences ion transport, whereas in a pancreatic cell, it regulates insulin secretion. As mentioned above, adenylate cyclase is regulated by a protein, known as G protein, which is a complex of three different subunits, α, β and γ. The A subunit of the cholera toxin catalyses the transfer of ADP-ribose group from NAD to the α subunit of the G protein. This prevents the G protein from switching off the synthesis of cAMP, with the result that cAMP is produced constitutively (i.e. at a constant rate). The high levels of cAMP cause Cl- and HCO3- ion channels in the apical cell membrane of the small intestine to open, resulting in uncontrolled secretion of these ions into the gut lumen (Figure 1.5d and e). The high cAMP levels also inhibit the uptake of Na+ ions into the epithelial cells, with the net result that Na+ and Cl− ions accumulate in the lumen (Bharati and Ganguly, 2011). In parallel, the clustering of GM1 receptors by the cholera toxin B subunit causes nearby K+ channels in the membrane to open, allowing efflux of K+ ions (Mulhern et al., 1989). High concentration of these ions causes water to move out of the epithelial cells into the lumen by osmosis. Thus, both water and essential electrolytes are lost from the body as the copious diarrhoea of cholera. The virulence factors that are responsible for the pathogenesis of Vibrio cholerae are encoded by two chromosomally integrated lysogenic phages: CTXφ (pronounced CTX fie) and VPIφ. (Remember that 'phage' is a shortened version of term 'bacteriophage', which you met in Block 1 Unit 5, Sections 5 and 8.2.) Phage CTXφ encodes the cholera toxin (CT) subunits A and B, and also two other toxins, accessory cholera toxin (Ace) and zona occludens toxin (Zot). Zot increases the permeability of the small intestine by affecting the structure of the intercellular tight junctions. Phage VPIφ encodes the toxin-coregulated pilus (TCP, see above) which, curiously, is the receptor for the CTXφ phage when it infects Vibrio cholerae. The procedures used to diagnose cholera differ depending on whether the disease is endemic, epidemic or usually absent from an area. Where cholera is endemic or epidemic, laboratory diagnosis is not always appropriate, because immediate treatment of the patient is required and laboratory tests may take several days to complete. Since cholera is transmitted by the faecal-oral route, it tends to occur in areas without proper sanitation, and many of these areas are found in low- and middle-income countries (LMICs). In addition, the resources for the laboratory diagnosis of cholera may simply be unavailable in some of these areas, and because patients can be successfully treated without a laboratory diagnosis being made, this procedure is largely unnecessary (apart from supporting the conclusions of the clinical assessment). Where cholera is usually absent from an area, laboratory diagnosis is needed to confirm that a patient has the disease, partly since medical staff may be unfamiliar with the symptoms and partly because a positive identification must be made before any necessary control methods are initiated. Cholera is usually absent from HICs, and these countries normally have a readily available laboratory service. The main symptom of cholera is profuse watery diarrhoea. The assessment of patients with diarrhoea has three stages: 1. First a history of the patient's disease is taken, including the details of any vomiting or fever, and their stools are examined by eye. The duration and consistency of the diarrhoea, and whether it contains blood, pus or mucus (which might suggest a diagnosis other than cholera) can point to the probable cause, as is the case for the characteristic rice-water stools of cholera patients. 2. Next the patient is examined to assess the degree of dehydration they have suffered, so that the appropriate rehydration therapy can be started. 3. Diarrhoea patients, particularly children, are also examined for any underlying nutritional deficiency, so that a suitable feeding programme can be administered. The malnutrition may have been caused by a cholera infection (if there is one), or by another disease such as pneumonia or measles. After this initial assessment, laboratory investigations may be carried out, depending on the information already obtained and the epidemiology of diarrhoea in the locality. In acute cholera the watery nature of the stools means that the commensal biota of the bowel is much diluted, and the main bacteria present are cholera vibrios. These can be seen on examination of a wet mount of the stool under the microscope. They appear as curved rods, in contrast to the majority of faecal bacteria, which are straight rods. The identity of the bacteria can be confirmed as Vibrio cholerae by adding antiserum against the H antigen to the wet preparation of the stool. • Where are the H antigens of a bacterium located? • They are on its flagella (see Unit 5 Section 6.2) H antiserum may be produced by injecting material containing a bacterium's flagella into a laboratory animal, waiting for an immune response to the antigen to occur, and then taking some of the animal's blood. The serum portion of the blood will contain antibodies that specifically bind that particular bacterium's H antigen. When Vibrio cholerae H antiserum is added to a wet mount of this bacterium, it will bind the organism's single polar flagellum. • Can you predict what effect addition of antiserum will have on the cholera vibrios? • It will render them immotile, so bacteria that were swimming around in the wet preparation will quickly become stationary. The serovar can also be determined at this stage by observing the immobilisation of the vibrios with the addition of either Inaba or Ogawa antisera (see Section 4.2.2). If a patient has a gastrointestinal disease in a country or area where instances of cholera are rare then a faecal specimen is usually requested, and screened for the bacteria and parasites routinely isolated from such samples. In the UK and many other countries Vibrio cholerae is not included among these bacteria and is only sought when the patient has recently travelled to an area where cholera is either endemic or epidemic. Faecal specimens should be processed within three to four hours, otherwise they will require refrigeration at 4 °C. As you saw in Section 4.1, in the acute cholera patient the faeces are so watery that the commensal biota is much reduced and Vibrio cholerae is the main bacterium present. The next stages of diagnosis involve growing and testing sample cultures. From stool samples suspected of harbouring Vibrio cholerae a tiny portion (a bacteriologist's loopful) is taken and inoculated into a broth of alkaline peptone water (APW). Peptone water is a general-purpose medium that can be adapted for a variety of uses. If it is prepared using a phosphate buffer with a (weakly) alkaline pH of 8.4 (as APW), the cholera vibrios are able to grow, while the other faecal bacteria cannot, since this pH inhibits many bacteria. This selective action of APW significantly increases the number of Vibrio cholerae bacteria in the culture and thus aids identification. In the investigation of a suspected cholera case, the material from the APW enrichment broth is plated out on a type of solid selective medium called thiosulphate-citrate-bile salts-sucrose (TCBS) agar following an incubation period of 6-8 hours. The selective agents in TCBS agar are the sodium thiosulphate, sodium citrate and the bile salts. These agents inhibit the growth of the majority of other Gram-negative bacteria that would otherwise grow from a faecal sample. Vibrio cholerae is also inhibited by this medium, but to a lesser extent than the other bacteria. The selectivity of TCBS in favour of Vibrio cholerae is further increased by its high salt (NaCl) concentration and its pH of 8.6. Bacterial colonies have a characteristic appearance on a particular agar, but sometimes different bacteria look very similar. Despite all the tailoring of TCBS to promote the growth of Vibrio cholerae some other organisms do still manage to grow, and they can be distinguished from the vibrios by incorporation of indicators into the medium. The first indicator comprises sucrose and a pH indicator, bromothymol blue, which turns yellow at low pH. Vibrio cholerae uses sucrose anaerobically for an energy source, i.e. the bacterium ferments sucrose. When sugars are fermented in this way, acid is produced so the pH around the bacterial colony falls. This change in pH can be clearly seen, since it causes the pH indicator to change colour from blue to yellow, and this in turn changes the colour of the bacterial colony. The colonies of V. cholerae thus appear yellow, while the colonies of bacteria that are unable to ferment sucrose appear blue, since they are coloured by the unchanged bromothymol blue. The second indicator detects the production of the gas hydrogen sulphide (H2S), by turning the centre of the colonies black. Vibrios do not produce H2S, but other organisms that grow on TCBS agar do, which makes them easily identifiable. You can see black and yellow colonies on a TCBS agar in Figure 1.6. After overnight incubation, the yellow colonies on the TCBS agar without black centres are most likely to be Vibrio cholerae. They are then Gram-stained. • What will be the appearance of the bacteria after Gram staining if the colonies are of Vibrio cholerae? • The Gram stain will show Gram-negative bacilli, which may appear comma-shaped. (Look again at Figure 1.4 to see the shapes of Vibrio cholerae bacteria.) If the bacteria have a comma-like appearance after being Gram stained then a pure culture is plated out onto a rich general-purpose agar (such as fresh blood agar: see Table 9.3 in Unit 9) and incubated for 6-8 hours, in order to carry out further tests. The bacterial growth taken directly from the TCBS plate is unsuitable, since substances in the agar can affect the test results. The first test carried out on the pure culture is a standard test used to distinguish between different species of bacteria, known as an oxidase test. Bacteria that can oxidise glucose to release energy have an electron transport chain similar to that used in respiration in human mitochondria. This chain comprises various cytochromes and other enzymes, ending with the enzyme cytochrome oxidase. This enzyme can also oxidise tetramethyl-para-phenylenediamine hydrochloride (TMPD), more simply known as oxidase reagent, to produce a purple-coloured product in less than 10 seconds. Vibrio cholerae is oxidase-positive, so it quickly turns TMPD purple when a colony from the plate is brought into contact with it. The final test for the colonies suspected to be cholera vibrios is the slide agglutination test (described next) to determine the group of their O antigens. • What are the groups of Vibrio cholerae that have been known to cause epidemics of diarrhoeal disease? • Group O1 is the main group of epidemic V. cholerae, but group O139 has also been responsible for epidemics in recent years. During a slide agglutination test, culture from a non-selective solid medium is suspended in sterile saline on a microscope slide and a small bacteriologist's loopful of test antiserum, which recognises the O1 antigen, is added and mixed in by rocking the slide. If the solution becomes 'clumpy' (see the top example in Figure 1.7) within 1 minute of mixing then agglutination has occurred, which means that the antibodies in the antiserum bind to the O1 antigen of the bacterium, cross-linking them into a clump. Conversely, if clumping does not occur then there is no agglutination and the O1 antigen is not recognised by those particular antibodies. The slide agglutination test is illustrated in Figure 1.7. If the organisms isolated from the TCBS plate that are oxidase-positive Gram-negative bacilli are group O1, they are presumptively identified as epidemic Vibrio cholerae O1. If this happens, then the authorities are usually informed since cholera is a notifiable disease in many countries, and member states of the WHO are required to report cases of the disease as part of the International Health Regulations. Any organism identified as presumptive Vibrio cholerae O1, is sent to a reference laboratory for confirmation, biotyping (as classical or El Tor) and serotyping. Remember that the biotypes are distinguished on the basis of their haemolytic activity, resistance to the antibiotic polymyxin B and their different susceptibilities to bacteriophage (see Section 3). The classical biotype can be further differentiated into the serovars Inaba and Ogawa. These typing tests are important because V. cholerae O1 and non-O1 cannot be distinguished biochemically. Molecular methods have been developed for diagnosing cholera, using a variety of techniques (Hoshino et al., 1998; Faruque et al., 1997; Yamazaki, 2008), but the main target of these methods is the gene encoding the A subunit of the cholera toxin (CT). Another test currently in development senses the presence of cholera toxin in water using iron oxide nanoparticles (Kaittanis, 2011). The immune response to enteric pathogens is initiated in the gut-associated lymphoid tissue (GALT), known as Peyer's patches. The gastrointestinal tract is an important route of entry for pathogens, so this lymphoid tissue constantly 'samples' the gut contents for antigenic material. After appropriate processing and presentation of antigen (see Block 2), B cells are stimulated to differentiate into plasma cells that produce antibodies. The most important antibody classes in combatting a cholera infection are secretory IgA and, to a lesser extent, secretory IgM, which are released from the gut mucosa (you will learn about the different classes of antibody and how they protect against pathogens in Block 2). • What would you expect to happen to antibodies secreted into the gut? • Antibodies are protein molecules and so should be digested along with proteins in food. In fact, the gut antibodies are thought to be protected from this degradation. For instance, secretory IgA has an additional polypeptide called a secretory component that protects against digestion by gut enzymes. The IgA and IgM antibodies bind to the B subunit of cholera toxin, thereby inactivating it, and they also bind to the lipopolysaccharide molecules of the bacteria themselves. IgG antibodies are produced, too, in response to a cholera infection, but they are largely confined to the blood. However, some of them do end up in the gut after leaking from the bloodstream through the damaged epithelium, or following the migration of a B lymphocyte, and these antibodies are thought to be significant elements of the immune response to cholera. Recovery from a cholera infection brings about long-lasting natural immunity, but infection with cholera of the classical biotype may provide better protection than a similar infection with El Tor cholera biotype. So far, the vaccines developed to combat cholera have been unable to produce the same results. • What would be the best route for administration of a cholera vaccine? • Since Vibrio cholerae is an enteric pathogen, an oral vaccine would have the most chance of provoking an immune response that resembled the response to a natural infection, and so result in a similar immunity to the organism. Several vaccines against cholera have been licensed and can be grouped into different types: • killed whole-cell vaccines • killed whole-cell plus subunit vaccines • live attenuated vaccines (a weakened, less virulent form of the pathogen). A killed whole-cell (WC) vaccine, comprising a mixture of El Tor and classical biotypes and Inaba and Ogawa serovars, was developed in 1894 by Waldemar Haffkine and was tested in India. This vaccine was administered by injection and did not perform well, giving 48% protection that only lasted for around three months. The WHO (2011b) described it as conveying 'incomplete, unreliable protection of short duration' and they do not recommend its use. • Suggest a reason why the vaccine performed poorly. • Administration of a WC vaccine by injection would not bring the cholera vibrios into contact with the Peyer's patches and the most important part of the immune response against cholera would not be provoked. At the time of writing (2011) there are two oral vaccines on the market. One (Dukoral) is a killed WC vaccine with an added toxin B subunit (i.e. a killed WC plus subunit vaccine). It provides 85-95% protection against O1 strains and lasts for more than two years, in all age groups. The other (Shanchol) is a killed WC vaccine and provides good protection against O1 and O139 in children under 5 years of age. Both vaccines are administered in two doses given between seven days and six weeks apart. Live attenuated cholera vaccines have also been developed; the strains used are altered so that they cannot cause cholera but still retain the ability to colonise the small intestine when given orally. Thus, the disease-causing classical Inaba strain (569B) has been genetically engineered to remove the cholera toxin A subunit gene (ToxA) and to add a mercury resistance gene (HgR) to distinguish the modified strain from wild-type strains (which are otherwise killed if grown in media containing mercury). The resulting non-disease-causing strain, CVD103-HgR, was used as the basis of the vaccine Orochol, and was tested in Indonesia between 1993 and 1997. The vaccine induced an IgG as well as a secretory IgA antibody response. It performed well in a small field trial (80% protection for 3 months), but in a larger trial of 67 000 volunteers in Indonesia it did not give sufficient protection, and was discontinued by the manufacturer in 2004 (WHO, 2009). Since the current seventh pandemic is caused by El Tor cholera (see Section 7), a vaccine containing El Tor V. cholerae might provide more protection than one based entirely on a classical strain. A cholera vaccine based on CVD103-HgR, but with the addition of an attenuated derivative of an El Tor strain lacking the genes encoding CT, Ace and Zot (strain CVD 111), has since been developed and was very promising when tested in the USA and Peru (Taylor et al., 1997). As yet, routine cholera vaccination is not cost-effective, because more than one dose is usually required, immunity does not last long and the disease is relatively rare. Nevertheless, pre-emptive vaccination of refugee populations seems to be effective in preventing large-scale epidemics. At present, no country demands a certificate of cholera vaccination as a condition of entry. Although cholera vaccines are available, other measures such as treatment and prevention are of paramount importance. Prevention of cholera depends on interrupting the faecal-oral route, so that food and drink remain uncontaminated with faecal material. This can be achieved with proper sanitation, clean water and hygienic methods of food preparation, which are fairly simple practices, but sadly, well beyond the means of many people. This is particularly true of populations of refugees, and those whose countries' infrastructures have been damaged by war or natural disasters. If a person contracts cholera, it is imperative that they are treated as quickly as possible since untreated cholera can have a case fatality rate as high as 50%. Fortunately, effective treatment can reduce this rate to less than 1%. More importantly, for many countries, effective treatment is cheap, comprising oral rehydration to replace lost fluid and electrolytes. Table 1.2 shows the composition of oral rehydration solution recommended by the WHO. The listed substances are dissolved in one litre of clean drinking water. Effective oral rehydration therapy relies on the following five factors: 1. a solution that is isotonic to blood plasma (to re-establish osmotic balance and maximise sodium and water re-absorption) 2. a sodium concentration close to that of blood plasma (optimises Na+ uptake) 3. a potassium concentration slightly higher than that of blood plasma to promote K+ absorption 4. a citrate (or bicarbonate) concentration to overcome blood acidosis due to loss of alkaline ions (bicarbonate) in diarrhoea and dehydration. 5. The presence of glucose, which is co-transported across the gut epithelium with Na+ ions, facilitating the uptake of Na+ from the gut. Without glucose, rehydration therapy will not work. The case fatality rates for cholera have dropped dramatically since the 1950s. What are the probable reasons for this fall? • Case fatality rates are likely to have dropped partly because treatment has become more easily available, and, as explained in Section 7, partly because El Tor cholera is less virulent than classical cholera. Oral rehydration therapy was first introduced in the early 1970s, and became widely available during the 1980s. In 80-90% of cases, oral rehydration alone is sufficient treatment, but when dehydration is severe, intravenous fluids are required. In these circumstances, antibiotics may be used to lessen the volume and duration of diarrhoea and reduce the carriage of vibrios in the faeces. Tetracycline is the drug of choice, but some V. cholerae are resistant to it and alternatives such as azithromycin, furazolidone, ciprofloxacin or others have to be used instead. The resistances exhibited by cholera vibrios can change rapidly; in their study of diarrhoeal stool samples, Kaur and Lal (1998) reported that the strains they isolated were sensitive to chloramphenicol until 1993, but resistant to this antibiotic from then onwards. Conversely, these strains were resistant to cotrimoxazole until 1996, but 100% sensitive to it in 1997. Although global case fatality rates have stabilised at around 2-3% for the past decade they differ around the world, averaging around 1% for Asia, Europe and the Americas but rising to around 5% in Africa. This marked difference reflects disparities in access to treatment, rather than any variation in the virulence of V. cholerae. The seventh and current cholera pandemic began in 1961, immediately after the 'end' of the sixth. It arose, surprisingly, not in India where the disease has a particularly long history, but in Indonesia in Sulawesi (Celebes Islands). This time, the causative agent was not the usual, classical cholera, but a new biotype, El Tor. This biotype was first isolated among pilgrims in a quarantine station called El Tor, and belonged to the same group (O1) as classical Vibrio cholerae, but was a different serovar from either Inaba or Ogawa. The disease it caused was different too: the El Tor strain was less virulent, produced a less effective immune response and was better able to persist in the environment. The new El Tor cholera began a relentless advance across the world that continues today. In 1963 it reached Bangladesh, in 1964 it got to India, and over the following years it went on to invade the former Soviet Union, Iran and Iraq. Africa was reached by 1970, but it took another 20 years for El Tor cholera to reach South America. In January 1991, a Chinese ship released bilge water infected with the cholera bacillus into the waters of Lima's port city, Callao. It was summer time and some locals would probably have been eating the popular dish ceviche, which contains raw shellfish. The shellfish quickly became contaminated with cholera bacteria, which soon infected the human population. The spread of the epidemic was made easier because Lima's water supply was unchlorinated. In that year, cholera spread rapidly, causing 400 000 reported cases and 4000 deaths in 16 South American countries. Nowhere in the world had seen an epidemic of these proportions since 1969, the year that cholera was made reportable under International Health Regulations. The question of exactly how Vibrio cholerae ended up in the bilge water of a ship is an interesting one. El Tor vibrios are known to persist longer in the environment than classical cholera bacteria. Research has shown that the El Tor bacillus is capable of parasitising algae and even goes into a kind of reversible suspended animation when suddenly placed in cold saltwater. Algae could therefore provide a reservoir of infection for epidemic V. cholerae. Some studies have lent weight to this theory. An investigation into cholera in Bangladesh and the El Niño weather pattern of southern Asia found that the two were linked. This suggested that cholera patterns were related to temperature changes in the region (Pascual et al., 2000). • How might temperature affect algal populations? • Warmer water temperatures might encourage algal blooms to flourish. Rita Colwell, a scientist who contributed to this investigation, has claimed for years that tracking the oceanic algal blooms that originate from Bangladesh and India, would allow the prediction of likely cholera outbreaks. A group from Lima has also found evidence that the occurrence of cholera may be affected by environmental factors. Twelve environmental sites were sampled for cholera vibrios, each month, from November 1993 to March 1995. In the winter, no vibrios were found in the samples, but in summer, the bacteria were detectable before cases of cholera occurred in the local community. The researchers suggested that an increase in environmental vibrios is followed by the appearance of cases of cholera in the human population and that increasing temperatures might cause the increase in vibrio numbers (Franco et al., 1997). The idea that the cholera bacterium alone is not sufficient to cause epidemic cholera, but the correct environmental conditions may also be required, suggests that von Pettenkofer's soil theory on the cause of cholera, although untrue, may have been based on some contributing evidence. While South America was reeling under the impact of the seventh cholera pandemic, scientists in Bangladesh and India were wondering if an eighth pandemic was just beginning. In 1992, in the southern coastal region of Bangladesh, a previously unknown strain of Vibrio cholerae was causing a cholera epidemic. This emergent group was the 139th to be distinguished by its O antigen, and so was labelled O139 and given the name 'Bengal'. In 1992 and 1993, Bengal cholera caused large epidemics in India and Bangladesh, which killed 5000 people. Then, in 1994, cholera O139 suddenly disappeared. In Bangladesh it was displaced by a resurgent El Tor strain. However, there was a resurgence of Bengal cholera during 1995 and 1996. The strain was identified in Dhaka and surrounding districts in Bangladesh (Faruque et al., 1997), and also in Kolkata (Calcutta) in India (Mukhopadhyay et al., 1998). These events illustrate the rapid changes in cholera epidemiology that can occur. Vibrio cholerae O139 had been reported in 11 South-East-Asian countries by 2000, but for the time being, remains confined to Asia. The origin of epidemic cholera strains is an area of fast-moving research. Bengal cholera lacks some of the genes that code for the O antigen of serogroup O1 cholera, and has instead a different DNA sequence that is unique to this strain. This finding has prompted the suggestion that Bengal cholera may have emerged from El Tor cholera by serovar-specific genetic changes (Faruque et al., 1997). A Dutch group (Mooi and Bik, 1997) has also suggested that horizontal transfer of genes encoding enzymes involved in cell wall polysaccharide synthesis may have played a key role in the emergence of cholera O139. In their summary document on cholera for 2009 the WHO concluded that: There is a need to shift from response to prevention in order to avert outbreaks by expanding access to improved sources of drinking-water and sanitation and by working with communities to encourage behavioural change to diminish the risks of infection.... The dynamic of cholera occurrences since 2005 combined with the emergence of new strains that lead to a more severe clinical presentation, increased antimicrobial resistance and climate change, suggest that cholera may well return to the forefront of the global public health agenda. (WHO, 2010, p. 297) The seventh cholera pandemic has had an enormous impact on the Americas and Asia, as you saw in Figure 1.8 in the previous section. However, this trend has now shifted to a high (and ever increasing) incidence in Africa, where the WHO reports that the failure of effective epidemic control has led to an increasing number of areas becoming endemic for cholera. New major outbreaks of cholera continue to occur, and many of these are associated with climate changes, such as El Niño, or the displacement of people into refugee camps and populations crippled by natural disasters. Table 1.3 lists the largest cholera outbreaks (according to the number of cases) for each year between 2000 and 2010. Moreover, the ease of intercontinental travel these days makes it easier to unwittingly transfer strains of Vibrio cholera from one region to another, and this is of growing concern. For instance, following an earthquake in January 2010, cholera broke out in Haiti, infecting 60 240 people and killing 1415 by November of that year. Cholera had not previously been seen in Haiti for decades and the strain (O1, Ogawa, El Tor) closely resembled one from India (Ceccarelli et al., 2011). The suggestion is that this V. cholerae strain was introduced to Haiti in a single event, possibly via contaminated food, water, or an infected person. It was even suggested that one of the rescue workers helping after the earthquake brought the disease into the country. However it arrived, this imported strain is unfortunately now likely to become endemic in Haiti. Thus, it would seem that wherever there is upheaval, whether social or environmental, there is the risk of cholera.

Miasmas

Name from ancient Greek for 'bad air' or 'pollution'. The miasma theory held that diseases such as cholera, plague and Chlamydia were air borne, and were caused by bad smells in the air.

Toxoplasmosis

Name given to any infection with the parasite Toxoplasma gondii. This organism is found throughout the world in mammals and birds, but cats are the definitive hosts, since sexual reproduction of T. gondii is confined to them.

Recombinant subunit vaccine

Preparation for injection or ingestion containing fragments (subunits) of an infectious agent obtained from the expression of certain pathogen genes, which have been cloned (copied many times) and inserted into the genome of a harmless strain of bacteria or yeast (the expression vector). This vector then manufactures the vaccine subunits in commercially useful quantities.

Antimicrobial resistance

Resistance of microbial pathogens (viruses, bacteria, fungi and protists) to antimicrobial drugs using one of several molecular mechanisms, e.g. target molecule modification or replacement, antibiotic-inactivating enzymes, reduced permeability to the antibiotic, and low-specificity membrane pumps that expel the antibiotics from the cell.

Definitive host

The host inside which a parasite reaches reproductive maturity.

Resistance to chemical attack

The range of anti-infective chemicals available to treat pathogens in human hosts, destroy them on surfaces or equipment (e.g. in healthcare settings), or repel or kill their vectors in the environment, may seem vast. However, the speed at which resistance to overused chemical agents develops in target organisms presents a major barrier to infection control.

Antigenic disguise

A mechanism evolved to evade the host immune response by Schistosoma species, in which the worms 'disguise' their own surface antigens by coating themselves with host macromolecules. The best known example of antigenic disguise occurs when schistosome parasites enter a new human host. An immune response begins against their unique surface antigens, but within a few hours the worms are no longer susceptible to damage by the host's antibodies and complement cascade. Schistosomes have the ability to 'cloak' themselves with molecules taken from their host, including human blood group antigens and major histocompatibility (MHC) molecules. The host antigens mask the parasite antigens, so the host's antibodies can no longer bind to the worms and initiate complement-mediated damage.

Amoebae

Amoebae are animal-like, single-celled protists that move and capture prey by means of cytoplasmic extensions known as pseudopodia. A few are pathogens, and the disease they cause is called amoebiasis.

Sporocyst

An elongated sac that produces a form of parasitic larva (rediae) or, in some cases, more sporocysts.

Climate effects

An obvious way in which climate impacts on the distribution of infectious diseases is in shaping the environments, and so conditions of temperature and humidity, in which pathogens and/or their vectors can replicate. This is well illustrated by the interaction between malaria parasites and Anopheles mosquitoes, which is described in the following two sections of this unit. 3.2.1 Malaria transmission and climate constraints The optimum temperature for Plasmodium larvae to complete their development in the mosquito in the shortest possible time (as little as 8 days, depending on the species) is around 27 °C. Parasite maturation takes longer at lower temperatures and does not occur at all below 14 °C or above 40 °C. Even a small deviation from the optimum temperature may slow the parasite's development beyond the lifespan of an adult mosquito, which has to survive long enough after it becomes infected for the parasite larvae to complete their development. Rainfall is another important factor for mosquito breeding: too little and the shallow water collections that support mosquito larvae dry up - too much rain and they flood and wash the larvae away (Figure 6.9). The average atmospheric humidity must also exceed about 60% saturation during the mosquito breeding season: below this humidity level, the mosquitoes' lifespan is too short to support the maturation of the parasite larvae. Climate change is gradually increasing the geographical range of malaria-carrying mosquitoes by expanding the territory in which the temperature, humidity and rainfall support breeding populations. And they are not alone in this respect, as the next section explains.

Ascaris lumbricoides

It is estimated that about a quarter of the human population is infected with Ascaris lumbricoides, making it the most common parasitic intestinal worm in humans. Although it occurs worldwide, it is most prevalent in the tropics and subtropics and is frequently found with other intestinal worms such as the hookworms. Also like hookworm, the effects of infection with Ascaris lumbricoides include malnutrition and anaemia. A person becomes infected with the parasite when embryonated eggs are ingested in faeces-contaminated food, water or soil. The egg hatches in the small intestine releasing the first- or sometimes second-stage larva. The larva then undergoes a migration akin to that of the hookworms, which some believe to be a remnant of a life cycle that once took place in an intermediate host but is now all carried out in a single host. Specifically, the larva burrows through the intestinal wall and into the bloodstream after which it is carried first to the liver, where it moults to a third stage, and then to the heart and lungs. It then migrates up the trachea, is swallowed by the host, and returns to the intestine where it started about 21 days previously! In the small intestine the larvae moult, mature to adult stage, copulate and begin passing eggs roughly nine weeks after ingestion. Ascaris are very large worms with males averaging 150-300 mm and females even longer (200-350 mm) in length. Egg output is prodigious, with in excess of 200 000 passed in faeces every day for up to two years, The egg is extremely resistant to desiccation and low temperatures, and can remain viable for several months. These properties are two of the main factors that ensure successful transmission of the parasite. • What other factors favour the transmission of this parasite? • The organism's simple, direct life cycle; the high rate of egg production; social living conditions; and poor sanitation. A burden of a few worms causes no symptoms or only minor effects such as abdominal pains, but a heavy infection can cause blockage of the intestine or of the bile duct and malnutrition. The larval migration through the lungs may cause coughing and wheezing and can lead to pneumonia. • What evidence would be required to diagnose an infection with Ascaris? • The presence of eggs in a stool sample. If pneumonia due to Ascaris is suspected, a sputum sample may reveal the presence of ascarid larvae. Treatment is by oral albendazole, or mebendazole (a related drug with a similar mechanism of action).

Antiprotist drugs

One reason why protist-mediated diseases have such serious consequences is that there are few really effective drugs to treat them. Here are some of the drugs and their mechanism of action: i). Metronidazole. Metabolised to yield a molecule that interacts with DNA. Treats giardiasis, amoebiasis and trichomoniasis ii).Suramin. Inhibits enzymes involved in ATP synthesis. Treats stage one African trypanosomiasis, especially T. b. rhodesiense iii). Pentamidine. Binds to DNA. Treats stage one African trypanosomiasis, especially T. b. gambiense leishmaniasis iv).Melarsoprol. arsenic compound that inactivates thiol (-SH) groups of several enzymes especially pyruvate kinase inhibiting synthesis of ATP. Treats second stage African trypanosomiasis v).Nifurtimox-eflornithine. Combination interferes with cell replication. Treats second stage T. b. gambiense vi). Tinidazole. Inhibits glucose metabolism, interferes with mitochondrial function and inhibits DNA synthesis. Treats E. histolytica amoebiasis, trichomoniasis, giardiasis vii). Benzimidazole. Inhibits glucose metabolism and interferes with mitochondrial function. Treats Chagas' disease viii). Iodoquinol. Blocks DNA replication. Treats E. histolytica amoebiasis ix). Paromomycin. Blocks ribosomal activity. Treats E. histolytica amoebiasis; cryptosporidiosis, cutaneous leishmaniasis x). Pentavalent antimonials. Inhibits ATP synthesis. Treatsall forms of leishmaniasis xii). Amphotericin B. Antifungal drug. Treats visceral leishmaniasis xiii). Miltefosine. Oral drug which alters the integrity of the cell membrane leading to apoptosis. Treats visceral leishmaniasis xiv). Nitazoxanide. Blocks ATP production. Treats cryptosporidiosis and metronidazole-resistant Giardia, Entamoeba and Trichomonas xv). Propamidine. Binds to DNA. Treats acanthamoeba amoebiasis; T. b. gambiense xvi). Pyrimethamine. Folic acid antagonist. Treats toxoplasmosis during second and third trimester of pregnancy xvii). Sulfadiazine. Folic acid antagonist. Treats toxoplasmosis during the first and second trimesters of pregnancy xviii). Spiramycin. Blocks ribosome activity toxoplasmosis during whole pregnancy. xix). Clindamycin. Blocks ribosome activity. Treats ocular toxoplasmosis. Protists are eukaryotes, so have the same type of molecules and metabolism as their hosts. Thus it is difficult to achieve selective toxicity. For instance, drugs that bind to DNA and disrupt its structure affect both protist and eukaryotic host cells, and rely for their efficacy on subtle differences between host and pathogen, such as different rates of drug uptake. However, two classes of drug mentioned in Table 6.1 do exhibit selective toxicity towards protist parasites. The first are the folic acid antagonists: protists are unable to absorb folic acid from their host, so they are sensitive to inhibition of their own folic acid metabolism. Their hosts, however, are not sensitive to these drugs as they are able to take up folic acid from the diet. The second type of the drug showing selective toxicity towards protist parasites is melarsoprol. This is a drug based on arsenic, which inhibits an enzyme found only in protists.

DNA vaccine

Preparation of 'naked' (cell-free) DNA encoding pathogenspecific antigens, which is coated onto gold particles and injected directly into the muscles of vaccine recipients using a gas-pressured gene gun. In theory, the injected DNA is incorporated into muscle-cell DNA and enough of the pathogenspecific antigens are synthesised to induce a protective immune response in the vaccine recipient against subsequent infection.

Acanthamoeba spp. amoebiasis

Species of Acanthamoeba and a species called Naegleris fowleri are environmental amoebae, common in freshwater and moist soil, which can cause a brain infection called primary amoebic meningoencephalitis, as well as eye infections. Eye infections usually follow an injury, such as a foreign body in the eye or the damage caused by wearing contact lenses for long periods. These infections result in gradual ulceration of the cornea and keratitis (inflammation of the cornea), which can cause blindness. Eye infections are diagnosed by taking a scraping of the eye, staining the amoebae with a specific antibody and viewing them under a microscope. For nervous system infections, wet-mount microscopy of a cerebrospinal fluid sample sometimes reveals motile trophozoites. If not, they can be revealed with Giemsa staining. Specific targeted drug treatments for these eye infections have not yet been devised, since the amoebae are often resistant to the antibiotics in normal use. However, a combination of more common antibacterial antibiotics has been used with some success.

Progress in new vaccines

The biological challenges outlined above partly explain why there are still only just over 20 vaccine-preventable diseases out of the hundreds of infectious diseases caused by pathogens around the world. However, the most significant barriers to success are difficulties in isolating suitable target antigens from pathogen species and problems in devising effective delivery systems that can be used in vaccines. Block 2 Unit 10 reviewed the strategies currently being trialled for the design of new vaccines, so the issues will not be repeated here. However, it is sobering to note that, at the time of writing (2012), the RTS,S/AS01 anti-malaria vaccine is the only one available with any effect against a parasite and its efficacy in initial trials was no more than 50% (RTS,S Clinical Trials Partnership, 2011). Even BCG, one of the longest established antibacterial vaccines, only protects infants from the most severe forms of TB-associated meningitis and extra-pulmonary TB if it is given under the age of 12 months. Vaccination beyond this age offers highly unreliable protection. This brief summary illustrates the extent of the challenges facing medical science in devising new vaccines to keep pace with pathogen evolution. The adaptability of pathogen and vector species presents the pharmaceutical industry with similar difficulties in devising effective drug treatments, antimicrobials, insecticides and other chemical agents to treat or control them.

Anisakis simplex

The disease anisakiasis was first recognised in 1962 in The Netherlands when larval nematodes were recovered from an intestinal lesion of a patient suffering from severe abdominal pain. The causative agent is the larval stage of Anisakis species of which A. simplex is the most common. The parasite is sometimes called the herring worm and although it is found worldwide it is particularly common in countries where the eating of raw or lightly cooked fish is usual such as Japan, where 90% of cases occur. Cases are increasingly reported from Europe especially The Netherlands, Spain, Germany and France. The life cycle is complex and entirely marine-based, with humans acting as accidental hosts. The definitive hosts are marine mammals (porpoises. dolphins, whales). Adult nematodes in the host's intestine shed eggs in their faeces, which hatch in the ocean to release the free-living infective third larval stage. These organisms are ingested by crustaceans such as shrimps (the first intermediate host), which are then eaten by fish such as cod, herring, hake, salmon, sardine and squid (the second intermediate host). The larvae penetrate the muscle of the fish so marine mammals (and humans) become infected by eating infected fish. The larval stages in fish flesh are easy to see as they are creamy-white and 18-36 mm long. (Huss and Ben Embarek, 2003, Section 5.1.4). Problems arise in cases of human infection because the larvae penetrate the gut wall, causing tissue damage and strong allergic reactions. In fact the occurrence of allergic reactions is so high now that A. simplex is included in the standard sets of allergens used to investigate food allergy. What is even more concerning is that the cooking or freezing of infected fish, which kills the parasite, does not destroy the A. simplex allergens (Audicana et al., 2002), so a contaminated fish can still produce undesirable consequences if consumed by a human.

Trichinella spiralis

The disease trichinellosis mainly affects people in Asia and parts of Europe. In fact, it is now considered to be endemic in China and Japan and is a notifiable disease in many countries, including the USA. It is caused by the nematode Trichinella spiralis, which are unusual nematodes because they are able to infect a wide variety of mammalian hosts, and because they live a part of their lives as intracellular parasites. The domestic pig is the main reservoir host, and the cause of most human infections is consumption of raw or undercooked pork products. For the parasite to be passed on the infected host must die and be eaten by another mammal. Humans are a dead end host unless cannibalism is involved or a dead infected person is scavenged by another mammal. The nematode is present in infected meat in individual 'nurse cells'. This is one of the intracellular phases of the parasite's life cycle. Upon ingestion, larvae are released from the meat by digestive juices in the stomach. They then locate to the small intestine and penetrate the epithelial cells lining the intestinal villi. Here they develop very quickly moulting four times in a 30 hour period to reach the adult stage. The smaller female adults (which are around 3 mm long) begin producing larvae, not eggs, only five days after mating. These larvae are equipped with a sharp sword-like 'stylet' in their mouth, which they use to penetrate the epithelial basement membrane and gain access to either the lymphatic system or the bloodstream. The larvae migrate around the body entering a variety of cell types, but become resident in skeletal muscle cells. Most other cells types that larvae enter die as a result of the invasion, and it is this that causes most of the pathology of the infection. Once inside skeletal muscle cells the larvae initiate a remarkable series of changes to the host muscle cell known as nurse cell formation. Over a period of two weeks the fully differentiated muscle cell becomes totally transformed into a cell that is no longer contractile but is instead a support system for the developing larva. This nurse cell or parasite complex can remain viable for the life of the host, but most become calcified after a few months. There is no specific treatment for trichinellosis, although mebendazole administered early in the infection may limit the number of migrating larvae and so reduce clinical symptoms. Later on in an infection, anti-inflammatory corticosteroids are recommended. Infection is prevented by cooking meat thoroughly before eating it, or initially freezing it for at least three days at -20 oC.

Constraints on public health progression

The prospects for progress are constrained by many factors in addition to the most obvious impacts on health due to poverty, shortage of food, inadequate housing, and lack of education. These inhibiting circumstances include: • the biological adaptability of important pathogens infecting humans and their vector species, which enables the rapid evolution of evasion mechanisms that render vaccines ineffective and promote resistance to chemical treatments and controls • difficult geological and climatic conditions, the risks posed by major natural disasters - drought, flooding - and the impact of climate change and agriculture on the distribution of pathogens and vectors • the lack of finance available in heavily indebted countries for installing sanitation and piped water, or providing vaccination programmes, soap, bed nets and other preventive strategies • the shortage of essential skills, knowledge, trained personnel and equipment, not only in the health service, but also in engineering, management and many other sectors • cultural contexts and social norms that limit the effectiveness of health-education initiatives and make individuals and communities highly resistant to behavioural change that would promote health and reduce the incidence and severity of infectious disease.

Apicomplexans

These organisms belong to the phylum Apicomplexa (see the Interactive Taxonomy diagram), and do not have flagella, except for the male gametes. They are named after the arrangement of organelles at one end of the cell, called the apical complex. The apical complex is instrumental in achieving penetration into host cells, and all apicomplexans are in fact parasites of animal cells. Apicomplexans have very complex life cycles that alternate between sexual and asexual reproduction. • Asexual reproduction produces many infective organisms via mitosis in a process called schizogony. • Sexual reproduction involves the fusion of gametes to produce a zygote known as an oocyst, where infective spores are formed by meiosis in a process called sporogony. The life cycles may also involve more than one host. If there is more than one host then the asexual stages take place in the intermediate host with the sexual stages occurring in the definitive host. The diseases, caused by apicomplexans, include toxoplasmosis, cryptosporidiosis and malaria, which are the subjects of the following three sections.

Trichomoniasis

This common sexually transmitted disease is caused by the small (10-20 μm) flagellate protist, Trichomonas vaginalis. The WHO estimates that globally over 170 million cases are acquired each year, with about 8 million cases in each of the USA and Europe. Unlike Giardia, there is no cyst stage in the life cycle of Trichomonas vaginalis so transmission is direct via the trophozoite stage. But, like Giardia, these protists do not possess mitochondria. Instead they have an organelle called a hydrogenosome, which has evolved from a mitochondrion and again is possibly an adaptation to an anaerobic way of life. The parasites are motile, each possessing four flagella. When passed from one person to another during sexual intercourse they move into the vagina and attach to the vaginal mucosa via specific adhesins, causing inflammation (vaginitis). Symptoms of trichomoniasis appear 4-28 days after infection and include mild to severe irritation and a foul-smelling discharge. Erosion of the epithelial cells occurs exposing the basement membrane to which the parasites attach via fibronectin receptors. This accounts for the long term nature of the disease in women and may also be the reason why infected women are more susceptible to HIV. Although infection rates in men are the same as those in women, the disease in men is largely asymptomatic and self limiting. In some cases there may be irritation and/or urethritis. Diagnosis is by using fluorescent antibodies that detect the pathogen in urine samples or urethral swabs. More recently, polymerase chain reaction (PCR) of urine or urethral swab samples has proved to be a highly accurate diagnostic technique (Hobbs et al., 2006). You will learn more about this technique in Unit 9.

Onchocerca volvulus

Vector is blackflies, especially Simulium damnosum and S neavei. Found in sub-cutaneous nodules. Causes river blindness. Found in Sub- Saharan Africa and Yemen and some tropical areas of Central and South America. One of the species of nematode responsible for cutaneous filariasis. Members of this species of worm live as adults in nodules in subcutaneous tissue, but most of the symptoms of the disease they induce, onchocerciasis, are caused by the microfilariae. These juvenile forms of the organism can cause inflammation of the outer layers of the skin, resulting in intense itching. If the larvae invade the cornea, iris or optic nerves, they can cause blindness, and this happens in about 5-8% of all cases. As the vector is the blackfly (Simulium), which breeds in fast-running rivers in tropical Africa, this severe symptom is termed 'river blindness'. Adult worms can survive for up to 9 years in a host, during which time they produce high numbers of microfilariae. The microfilariae migrate out of the nodules into the skin and blood capillaries, from where they can be ingested by biting blackflies with the blood meal. The blackfly is a true intermediate host as well as a vector, because the larvae undergo a migration out of the fly gut into the haemocoel and enter thoracic muscle cells where they develop into third stage infective larvae. The process takes about 2 weeks, and it is this infective stage of the parasite that migrates to the fly's proboscis and is released into the next host during a blood meal. Diagnosis of onchoceriasis is by the antibody-based technique ELISA to detect parasite antigens in a blood sample, or a skin biopsy for microscopic detection of microfilariae.

Climate change on vector species

Vector-borne diseases kill approximately 1.1 million people and cause the loss of almost 50 million years of healthy life every year. In recent decades, evidence is emerging that climate change is favouring the expansion in numbers and territories affected by the vectors of several important pathogens. For example, some species of Anopheles mosquitoes are now reproducing at higher altitudes than a century ago because the climate has become warmer. The inhabitants of highland areas of East Africa, including cities like Nairobi, which were built above the 'malaria line' are increasingly vulnerable to infection because the climate has become more hospitable to the vector mosquitoes. The WHO (2009) report on Protecting Health from Climate Change predicts that, by the year 2030, climate effects will have increased the size of the population at risk of malaria in Africa by 170 million people. Dengue haemorrhagic fever is also increasing rapidly in urban and semi-urban areas, where uncovered water sources and rising temperatures promote the breeding of its vector, the Aedes mosquitoes. The WHO (2009) report estimates that two billion people could be exposed to dengue viruses by the 2080s as a result of climate change. In China, Zhou et al. (2008) have estimated that, by 2050, climate change may have increased the distribution of schistosomiasis across a further 8.1% of China's vast landmass, due to the expansion of habitats for the freshwater snails that are its intermediate host. • Suggest two or three other ways in which climate change could increase exposure to pathogens in human populations. Answer You may have thought of the displacement of population due to increased exposure to extreme weather events, such as flooding and droughts. Crowded refugee settlements inevitably expose their inhabitants to a range of infections; floods create new breeding grounds for pathogens and vectors (e.g. Anopheles mosquitoes have returned to marshland in southern England) and increase the transmission of diarrhoeal diseases; droughts expose populations to food shortage and the increased vulnerability to infection that always accompanies malnutrition. The WHO report concludes that: Effects on infectious disease will not be restricted to developing tropical regions. For example, climate change is also expected to change distributions of diseases such as Lyme disease and tick-borne encephalitis, and to increase rates of Salmonella and other foodborne infections in Europe and North America.