Drug resistance is one of the most medically relevant forms of pathogen evolution. To date, vaccines have not failed with the same depressing regularity as drugs. Does that then make vaccines evolution-proof? In the face of vaccination, pathogens are thought to evolve in two ways: by evolving epitope changes at the antigenic target of vaccination (epitope evolution); or by evolving changes at other antigenic loci, some of which may involve virulence (virulence evolution). The fundamental difference between these two forms of evolution is that virulence evolution could lead to disease outcomes in unvaccinated people that are more severe than would have been seen prior to evolution. One of the theoretical assumptions of virulence evolution is that more virulent parasites will have a selective advantage over less virulent parasites in an immunized host, and are thus more likely to be transmitted. The assumption is that more virulent parasites may be competitively more superior in mixed infections, or may be better able to evade/modulate the host immune response. Thus, the aim of this thesis was to experimentally test whether more virulent parasites have a within-host selective advantage in an immunized host or whether vaccine efficacy is more likely to depend on genetic differences at the targeted sites of vaccination. I used clones (genotypes) of the rodent malaria Plasmodium chabaudi originally derived from wild-caught Thicket (Thamnomys rutilans) rats to infect laboratory mice and a rodent analogue of the candidate blood-stage malaria vaccine apical membrane antigen 1 (AMA-1). I found that within-host selection did not depend on parasite virulence, and that protective efficacy depended on genotype-specific differences at the vaccine target. Vaccine-induced protection was not enhanced by including a number of allelic variants. However, such genotype-specific responses were only observed when the vaccine was tested against genetically distinct P. chabaudi parasites. When one P. chabaudi genotype was serially passaged through naïve mice the derived line was more virulent and was subsequently less well controlled by vaccine-induced immunity. In other experiments I found within host competition not to be immune-mediated. Thus my results suggest that vaccination has the potential to select for more virulent parasites but that the selective advantage is likely to be independent of competition. The selective advantage may be attributable to the enhanced immune evasion of more virulent parasites. However, without genetic markers of virulence, the mechanisms that mediate this selection remain unknown. My thesis contributes towards a growing body of evidence that vaccines have the potential to differently alter the within-host parasite dynamics of particular pathogen genotypes and that the selection imposed is likely to be system specific, depending on the fine specificity of the vaccine-induced responses and the identity of infecting parasites. Although vaccine potency may not be enhanced by including more than one allelic variant of an antigen, multi-valent vaccines may be one of the best ways to avoid the inadvertent selection for more virulent malaria parasites.

The study of social evolution and virulence in parasites is concerned with fitness consequences of trade-offs between parasite life history traits and interactions between parasite species and/or genotypes with their hosts. I develop our understanding of social evolution and virulence in parasites in several ways. (1) I review empirical evidence for the fundamental predictions of virulence-transmission trade-off theory and demonstrate that the fit between theory and data is primarily qualitative rather than quantitative; that parasites differ in their degree of host generalism, and this is likely to impact virulence in four ways. (2) I take a comparative approach to examine the underlying causes of an observed statistical variation in the size of parasite infectious doses across taxa, revealing that mechanisms used by parasites to infect hosts are able to explain variation in both infectious dose and virulence. (3) I formally compare data on human pathogens to explain variation in virulence across taxa, revealing that immune subversion and not growth rate, explains variation in virulence. This allows me to predict that immune subverters and not fast growing parasites are likely to cause the most virulent clinical infections. (4) Using bacteria and their naturally infecting viruses (bacteriophage), I take an experimental approach to investigate the consequences of coinfection for parasite life history traits, and find that viruses cultured under a mix of single infections and coinfections evolved plasticity; they killed hosts more rapidly when coinfecting, and this resulted in high fitness under both single infection and coinfection conditions. (5) I experimentally investigate how selection within and between hosts and patches of hosts affects the fitness and virulence of populations of these viruses. I find that limited host availability favours virulent, faster killing parasites with reduced transmission; suggesting high, rather than low, virulence may be common in spatially structured host-parasite communities.

<p>Dengue is an important vector-borne virus that infects on the order of 400 million individuals per year. Infection with one of the virus's four serotypes (denoted DENV-1 to 4) may be silent, result in symptomatic dengue 'breakbone' fever, or develop into the more severe dengue hemorrhagic fever/dengue shock syndrome (DHF/DSS). Extensive research has therefore focused on identifying factors that influence dengue infection outcomes. It has been well-documented through epidemiological studies that DHF is most likely to result from a secondary heterologous infection, and that individuals experiencing a DENV-2 or DENV-3 infection typically are more likely to present with more severe dengue disease than those individuals experiencing a DENV-1 or DENV-4 infection. However, a mechanistic understanding of how these risk factors affect disease outcomes, and further, how the virus's ability to evolve these mechanisms will affect disease severity patterns over time, is lacking. In the second chapter of my dissertation, I formulate mechanistic mathematical models of primary and secondary dengue infections that describe how the dengue virus interacts with the immune response and the results of this interaction on the risk of developing severe dengue disease. I show that only the innate immune response is needed to reproduce characteristic features of a primary infection whereas the adaptive immune response is needed to reproduce characteristic features of a secondary dengue infection. I then add to these models a quantitative measure of disease severity that assumes immunopathology, and analyze the effectiveness of virological indicators of disease severity. In the third chapter of my dissertation, I then statistically fit these mathematical models to viral load data of dengue patients to understand the mechanisms that drive variation in viral load. I specifically consider the roles that immune status, clinical disease manifestation, and serotype may play in explaining viral load variation observed across the patients. With this analysis, I show that there is statistical support for the theory of antibody dependent enhancement in the development of severe disease in secondary dengue infections and that there is statistical support for serotype-specific differences in viral infectivity rates, with infectivity rates of DENV-2 and DENV-3 exceeding those of DENV-1. In the fourth chapter of my dissertation, I integrate these within-host models with a vector-borne epidemiological model to understand the potential for virulence evolution in dengue. Critically, I show that dengue is expected to evolve towards intermediate virulence, and that the optimal virulence of the virus depends strongly on the number of serotypes that co-circulate. Together, these dissertation chapters show that dengue viral load dynamics provide insight into the within-host mechanisms driving differences in dengue disease patterns and that these mechanisms have important implications for dengue virulence evolution.</p> / Dissertation

Marek’s disease (MD) is an oncogenic disease affecting chickens and is estimated to cost the worldwide poultry industry $1-2 billion annually. The causative agent of MD, Marek’s disease virus (MDV), provides a welldocumented example of virus virulence evolution occurring over a period of sixty years. The reason behind this evolution is unknown, although certain untested hypotheses have been suggested. These include vaccination (with increasingly potent vaccines) and other aspects of industrialisation, such as the decreased cohort duration of successive generations and an increased stocking density of the broiler flocks. In this thesis, four sections of work are undertaken. First, estimation of epidemiological parameters is tackled: virulence of MDV is quantified by looking at host mortality and virus shedding rates in vaccinated and unvaccinated birds. This is achieved via maximum likelihood estimation and Bayesian McMC techniques. Second, viral fitness is quantified by defining multiple lifetime fitness functions using the parameters previously estimated to understand the direction and force of virulence selection for different farm environments. Third, the impact of an outbreak of MDV on a broiler flock is examined by simulating a whole flock of birds. This provides an epidemiological understanding of the virus at the flock level and can help elucidate methods for disease control and surveillance and can also give a fitness measure to understand on-farm evolution of the virus. Fourth, a between-farm model is analysed to evaluate which MDV strains are able to persist in a network structure of farms and how this might be affected by biosecurity measures, different farm networks, farm size, bird lifespan and vaccination. This provides insights into how quickly a different strain can invade a farm network and the plausibility of it becoming endemic. Parameter estimation results show that the time to death for an infected bird decreases and its virus shedding rate increases with previous definitions of virulence in the literature. Model results suggest that the choice of fitness measure alters the conclusions reached. Increasing the amount of demographic structure introduced into the fitness measure shows that neither vaccine introduction nor decreasing a bird’s lifespan changes the ability of more virulent virus strains to outcompete less virulent strains. In any environment, more virulent strains are always selected for. Epidemiological results suggest that vaccination allows a low prevalence of virus on a farm although there are no deaths from the disease itself. Analogous results for between-farm spread suggest that if on-farm cleaning efficiency is low enough, a high prevalence of disease throughout a network of vaccinated flocks can exist but the farms themselves show no signs of increased mortality from the disease. The hypotheses for explaining the increase in virulence of MDV may not be consistent with the results of this work. Despite previous arguments that vaccines are driving the evolution of virulence ofMDV, this first quantitative work on the subject demonstrates that this might not be the case. This work also formulates new hypotheses to explain why MDV virulence has increased over the past sixty years which will pave the way for ongoing research in the area of virulence evolution in farm environments.

The health of pollinators, especially bees, is of the utmost importance to success of many agricultural ecosystems. Microorganisms can cause diseases in bees; such microbes are pathogenic. The ability of a pathogen to cause harm to its host (such as a bee) is termed its virulence. Studying the evolution of different levels of virulence can lead researchers to a better understanding of pathogens, and potentially predict how much harm a pathogen can cause in the future. We studied the evolution of virulence levels for a fungal disease of bees. This group of fungi is composed of 28 species, and some cause a disease in bees called chalkbrood while others do not. Using what we know about virulence evolution we wanted to see if the pathogens could infect all bees, if the pathogens varied in virulence when infecting at the same time as another pathogen, and if solitary bees had any behavioral adaptations that might increase or decrease chalkbrood infection. By using DNA sequences, the relationship between the genetic structures of each of the fungal species was studied, and we found that pathogens of solitary bees grouped together while pathogens of social bees (honey bees) were not part of this group. We then found that a solitary bee pathogen did not infect honey bees very well, and vice versa. The nuances of the relationship between two solitary bee pathogens were examined more closely to determine how the two pathogens interact in this bee. In this case, under varying conditions of infection, one pathogen always maintained a similar level of virulence and spore production, while the other pathogen varied in these measures. In addition, when doses of these fungi were fed to bee larvae at different times, more bees survived than when the doses were given at the same time, suggesting that bee immune responses are very important. Finally, we found no evidence of any specific behaviors of solitary bees exposed to infective spores that would suggest these bees have behaviors that are evolved to alter chalkbrood levels in populations.

Immunological heterogeneity is the norm in most free-living vertebrate populations, creating a diverse and challenging landscape for pathogens to replicate and transmit. This dissertation work sought to determine sources of immunological heterogeneity, as well as the consequences of this heterogeneity on pathogen fitness and evolution. A major source of heterogeneity in free-living host populations is the degree of exposure to a pathogen, as well as a host's history of exposure to a pathogen, which can create variation in standing immunity. We sought to determine the role of exposure heterogeneity on host susceptibility and immunity to secondary infection, and the influence of this heterogeneity on pathogen fitness and virulence evolution in a wild bird-bacterium system. We first determined that exposure level has a significant effect on host susceptibility to infection, severity of disease and infection, as well as immunity produced to secondary infection. Subsequently, we tested whether exposure history, and the immunity formed from this previous exposure, altered the within-host fitness advantage to virulent pathogens. We determined that previous low-level repeat exposure, which wild hosts likely encounter while foraging, produces a within-host environment which greatly favors more virulent pathogens. While within-host processes are vital for understanding and interpreting the evolutionary pressures on a pathogen, the ultimate metric of pathogen fitness is transmission. We therefore tested whether exposure history altered the transmission potential of a host and whether prior host exposure selected for more virulent pathogens. The transmission potential of a host significantly decreased with previous exposure, and high levels of previous exposure selected for more virulent pathogens. While we anticipated selection to be strongest at low-levels of exposure based on our previous results, we found here that high doses of prior exposure resulted in the strongest transmission advantage to virulence. This study also provided insight into the nuanced nature of transmission, which our results indicate is determined both by the degree of within-host pathogen replication as well as host disease severity. Together, our findings underscore the importance of exposure level and exposure history in natural populations in determining susceptibility, immunity and pathogen virulence evolution. / Doctor of Philosophy / Infectious diseases disrupt and threaten all life on this planet. To better anticipate and understand why some diseases are more harmful than others, it is vital that we consider the natural variability that exists in animal populations. A major source of variation in populations that experience disease is exposure level to a pathogen, as well as the history of exposure to a pathogen, which can alter an individual’s protection against future exposures. We sought to determine the role of exposure level on the likelihood of an individual contracting an infection, their protection from future infections, and the influence this has on pathogen evolution in a wild bird-bacterium system. We determined that exposure level has a significant effect on the likelihood an individual has of becoming infected, how severe the infection became, as well as how protected they were from future infections. Subsequently, we tested whether exposure history, and the immunity formed from previous exposure, altered the ability of pathogen strains that cause different levels of harm to replicate. We determined that previous low-level exposure, which hosts likely encounter in the wild, creates a level of immunity that favors more harmful strains of the pathogen. While understanding what happens within a host is important, the ultimate metric for predicting whether more or less harmful types of pathogens will persist is the ability of each pathogen type to spread from one host to another. We therefore tested whether exposure history altered the spread potential of a host and whether previous exposure preferentially favored the spread of more harmful pathogens. The spread potential of a host was much lower if that host had previously been exposed to the pathogen, and high levels of previous exposure in hosts only allowed the more harmful pathogen types to spread. We also found that a host’s spread potential was the result of both how much pathogen they had in their body, as well as how inflamed their affected tissues were. Together, our results indicate that natural variation in prior exposure to pathogens, which is common in all animal populations, including humans, can favor more harmful pathogen types.

Le système d'incompatibilité végétative a été décrit chez tous des champignons (Eumycètes) comme intervenant dans la limitation des fusions somatiques entre conspécifiques. Chez les champignons la fusion somatique est uniquement possible entre individus de même GCV (Groupe de Compatibilité Végétative). Comme tous les systèmes de reconnaissance du soi, le fonctionnement du système d'incompatibilité végétative des champignons est basé sur une grande diversité allélique. Cette thèse propose d'étudier la relation qui semble exister entre cette diversité des gènes impliqués dans l’incompatibilité végétative des champignons et la pression parasitaire exercée par des éléments cytoplasmiques délétères (ou DCE) transmis lors des fusions somatiques. Trois problématiques ont été abordées, avec trois approches différentes : (1) une approche conceptuelle générale portant sur l’évolution des systèmes de reconnaissance du soi, (2) une approche de modélisation sur le maintien de la diversité en GCV de la population de champignon par un DCE et (3) une approche expérimentale, pour étudier d’une part la perméabilité de la barrière d‘incompatibilité végétative et d’autre part l’interaction C. parasitica/CHV et les liens existant entre transmission et virulence du CHV. Ces études ont permis de montrer l'importance de la perméabilité de la barrière d'incompatibilité végétative à la fois au niveau du maintien de la diversité génétique de la population d'hôte et au niveau de la prévalence des DCE. Il semblerait donc que les DCE évoluent vers des niveaux de virulence faible du fait de la limitation de leur transmission par le système d'incompatibilité végétative de leur hôte. Nos résultats expérimentaux suggèrent que lorsque la diversité en GCV de la population d'hôte est faible, la virulence des DCE pourrait évoluer suivant le modèle du trade-off impliquant une évolution vers un niveau de virulence intermédiaire optimal. Ces travaux permettent donc de mieux comprendre les mécanismes agissant sur l'écologie et l'évolution des interactions champignon/DCE qui, au vu de cette étude, apparaissent comme de bon modèles pour l’étude des systèmes hôtes/parasites. / Vegetative incompatibility systems have been described in Fungi as controlling somatic fusion between conspecifics. For fungi, only fungi of the same vc type can fuse together. As other self recognition systems, this system involved high allelic diversity at specific genes. The issue of this work is to study the cause and effect relationship between the evolution of vegetative incompatibility systems and the selective pressure drove by cytoplasmic deleterious elements, transmitted during somatic fusion. Three problematics with three different approach were done : (1) a conceptual general framework on the evolution of self recognition systems (2) a theoretical work on the maintenance of vc type diversity by DCE and (3) an experimental work on the study of relationship between transmission and virulence in the C. parasitica/CHV host-parasite system. Ours results showed the key role of the permeability of the vegetative incompatibility barrier both for vc type diversity maintenance and on DCE prevalence. DCE would evolve toward avirulence in response to the transmission limitation by host incompatibility systems. Experimental work suggested a positive link between virulence and transmission in some population of CHV when host present a low vc type diversity, which could allow the evolution of the DCE toward an intermediate optimal virulence. This study would shed some light on mechanisms acting on the ecology and the evolution of fungi/DCE interaction which, according to our results, would be good study models for works on host-parasite systems.