Genotypic variation in parasite avoidance behaviour and other mechanistic, nonlinear components of transmission

Parasite transmission in size-structured populations

Host heterogeneity can affect parasite transmission, but determining underlying traits and incorporating them into transmission models remains challenging. Body size is easily measured and affects numerous ecological interactions, including transmission. In the snail-schistosome system, larger snails have a higher exposure to parasites but lower susceptibility to infection per parasite. We quantified the effect of size-based heterogeneity on population-level transmission by conducting transmission trials in differently size-structured snail populations and competing size-dependent transmission models. Populations with greater proportions of large snails had lower prevalence, and small snails were shielded from infection by co-occurring large conspecifics. Furthermore, a fully dependent transmission model that incorporated body size in both exposure and susceptibility outperformed other candidate models considered. Incorporating traits such as body size, which are affected by and directly affect host ecology, into transmission models could yield insights into natural dynamics and disease mitigation in many systems.

Age Structure Eliminates the Impact of Coinfection on Epidemic Dynamics in a Freshwater Zooplankton System

AbstractParasites often coinfect host populations and, by interacting within hosts, might change the trajectory of multiparasite epidemics. However, host-parasite interactions often change with host age, raising the possibility that within-host interactions between parasites might also change, influencing the spread of disease. We measured how heterospecific parasites interacted within zooplankton hosts and how host age changed these interactions. We then parameterized an epidemiological model to explore how age effects altered the impact of coinfection on epidemic dynamics. In our model, we found that in populations where epidemiologically relevant parameters did not change with age, the presence of a second parasite altered epidemic dynamics. In contrast, when parameters varied with host age (based on our empirical measures), there was no longer a difference in epidemic dynamics between singly infected and coinfected populations, indicating that variable age structure within a population eliminates the impact of coinfection on epidemic dynamics. Moreover, infection prevalence of both parasites was lower in populations where epidemiologically relevant parameters changed with age. Given that host population age structure changes over time and space, these results indicate that age effects are important for understanding epidemiological processes in coinfected systems and that studies focused on a single age group could yield inaccurate insights.

Natural variation in host feeding behaviors impacts host disease and pathogen transmission potential

Animals ranging from mosquitoes to humans often vary their feeding behavior when infected or merely exposed to pathogens. These so-called "sickness behaviors" are part of the innate immune response with many consequences, including avoiding orally transmitted pathogens. Fully understanding the role of this ubiquitous behavior in host defense and pathogen evolution requires a quantitative account of its impact on host and pathogen fitness across environmentally relevant contexts. Here, we use a zooplankton host and fungal pathogen as a case study to ask if infection-mediated feeding behaviors vary across pathogen exposure levels and natural genetic variation in susceptibility to infection. Then, we connect these changes in behavior to pathogen transmission potential (spore yield) and fitness and growth costs to the host. Our results validate a protective effect of altered feeding behavior during pathogen exposure while also revealing significant variation in the magnitude of this response across host susceptibility and pathogen exposure levels. Across all four host genotypes, feeding rates were negatively correlated with susceptibility to infection and transmission potential. The most susceptible genotypes exhibited either strong anorexia, reducing food intake by 26%-42%, ("Standard") or pronounced hyperphagia, increasing food intake by 20%-54% ("A45"). Together, these results suggest that infection-mediated changes in host feeding behavior-which are traditionally interpreted as immunopathology- may in fact serve as crucial components of host defense strategies and warrant further investigation.

Partitioning variance in immune traits in a zooplankton host-Fungal parasite system

Host immune traits arise from both genetic and environmental sources of variation. When immune traits have a strong genetic basis, the presence and severity of disease in a population may influence the distribution of those traits. Our study addressed how two immune-related traits (gut penetrability and the hemocyte response) are shaped by genetic and environmental sources of variation, and how the presence of a virulent disease altered the relative frequency of these traits in natural populations. Daphnia dentifera hosts were sampled from five Indiana lakes between June and December 2017 before and during epidemics of their fungal pathogen, Metschnikowia bicuspidata. Collected Daphnia were experimentally exposed to Metschnikowia and assayed for their gut penetrability, hemocyte response, and multi-locus genotype. Mixed-effects models were constructed to partition variance in immune traits between genetic and environmental sources. We then isolated the genetic sources to produce genotype-specific estimates of immune traits for each multi-locus genotype. Finally, we assessed the relative frequency and dynamics of genotypes during epidemics and asked whether genotypes with more robust immune responses increased in frequency during epidemics. Although genotype was an important source of variation for both gut penetrability and the hemocyte response, environmental factors (e.g., resource availability, Metschnikowia prevalence, and co-infection) still explained a large portion of observed variation, suggesting a high degree of flexibility in Daphnia immune traits. Additionally, no significant associations were detected between a genotype's immune traits and its frequency in a population. Our study highlights the power of variance partitioning in understanding the factors driving variation in Daphnia traits and motivates further research on immunological flexibility and the ecological drivers of immune variation.

Evolution and Ecology of Parasite Avoidance

Parasite avoidance is a host defense that reduces the contact rate with parasites. We investigate avoidance as a primary driver of variation among individuals in the risk of parasitism and the evolution of host-parasite interactions. To bridge mechanistic and taxonomic divides, we define and categorize avoidance by its function and position in the sequence of host defenses. We also examine the role of avoidance in limiting epidemics and evaluate evidence for the processes that drive its evolution. Throughout, we highlight important directions to advance our conceptual and theoretical understanding of the role of avoidance in host-parasite interactions. We emphasize the need to test assumptions and quantify the effect of avoidance independent of other defenses. Importantly, many open questions may be most tractable in host systems that have not been the focus of traditional behavioral avoidance research, such as plants and invertebrates.

Circadian rhythms mediate infection risk in Daphnia dentifera

Biological rhythms mediate important within-host processes such as metabolism, immunity, and behavior which are often linked to combating disease exposure. For many hosts, exposure to pathogens occurs while feeding. However, the link between feeding rhythms and infection risk is unclear because feeding behavior is tightly coupled with immune and metabolic processes which may decrease susceptibility to infection. Here, we use the Daphnia dentifera-Metschnikowia bicuspidata host-pathogen system to determine how rhythms in feeding rate and immune function mediate infection risk. The host is known to have a nocturnal circadian rhythm in feeding rate, yet we found that they do not exhibit a circadian rhythm in phenoloxidase activity. We found that the time of day when individuals are exposed to pathogens affects the probability of infection with higher infection prevalence at night, indicating that infection risk is driven by a host's circadian rhythm in feeding behavior. These results suggest that the natural circadian rhythm of the host should be considered when addressing epidemiological dynamics.

Timescale reverses the relationship between host density and infection risk

Host density shapes infection risk through two opposing phenomena. First, when infective stages are subdivided among multiple hosts, greater host densities decrease infection risk through 'safety in numbers'. Hosts, however, represent resources for parasites, and greater host availability also fuels parasite reproduction. Hence, host density increases infection risk through 'density-dependent transmission'. Theory proposes that these phenomena are not disparate outcomes but occur over different timescales. That is, higher host densities may reduce short-term infection risk, but because they support parasite reproduction, may increase long-term risk. We tested this theory in a zooplankton-disease system with laboratory experiments and field observations. Supporting theory, we found that negative density-risk relationships (safety in numbers) sometimes emerged over short timescales, but these relationships reversed to 'density-dependent transmission' within two generations. By allowing parasite numerical responses to play out, time can shift the consequences of host density, from reduced immediate risk to amplified future risk.

Host Controls of Within-Host Disease Dynamics: Insight from an Invertebrate System

AbstractWithin-host processes (representing the entry, establishment, growth, and development of a parasite inside its host) may play a key role in parasite transmission but remain challenging to observe and quantify. We develop a general model for measuring host defenses and within-host disease dynamics. Our stochastic model breaks the infection process down into the stages of parasite exposure, entry, and establishment and provides associated probabilities for a host's ability to resist infections with barriers and clear internal infections. We tested our model on Daphnia dentifera and the parasitic fungus Metschnikowia bicuspidata and found that when faced with identical levels of parasite exposure, Daphnia patent (transmitting) infections depended on the strength of internal clearance. Applying a Gillespie algorithm to the model-estimated probabilities allowed us to visualize within-host dynamics, within which signatures of host defense could be clearly observed. We also found that early within-host stages were the most vulnerable to internal clearance, suggesting that hosts have a limited window during which recovery can occur. Our study demonstrates how pairing longitudinal infection data with a simple model can reveal new insight into within-host dynamics and mechanisms of host defense. Our model and methodological approach may be a powerful tool for exploring these properties in understudied host-parasite interactions.

Dose relationships can exacerbate, mute, or reverse the impact of heterospecific host density on infection prevalence

The likelihood an individual becomes infected depends on the community in which it is embedded. For environmentally transmitted parasites, host community composition can alter host density, the density of parasites that hosts encounter in the environment, and the dose to which hosts are subsequently exposed. While some multi-host theory incorporates some of these factors (e.g., competition among hosts), it does not currently consider the nonlinear relationships between parasite exposure dose and per-propagule infectivity (dose-infectivity relationships), between exposure dose and infected host mortality (dose-mortality relationships), and between exposure dose and parasite propagule excretion (dose-excretion relationships). This makes it difficult to predict the impact of host species on one another's likelihood of infection. To understand the implications of these nonlinear dose relationships for multi-host communities, we first performed a meta-analysis on published dose-infectivity experiments to quantify the proportion of accelerating, linear, or decelerating dose-infectivity relationships; we found that most experiments demonstrated decelerating dose-infectivity relationships. We then explored how dose-infectivity, dose-mortality, and dose-excretion relationships might alter the impact of heterospecific host density on infectious propagule density, infection prevalence, and density of a focal host using two-host, one-parasite models. We found that dose relationships either decreased the magnitude of the impact of heterospecific host density on propagule density and infection prevalence via negative feedback loops (decelerating dose-infectivity relationships, positive dose-mortality relationships, and negative dose-excretion relationships), or increased the magnitude of the impact of heterospecific host density on infection prevalence via positive feedback loops (accelerating dose-infectivity relationships and positive dose-excretion relationships). Further, positive dose-mortality relationships resulted in hosts that traditionally decrease disease (e.g., low competence, strong competitors) increasing infection prevalence, and vice versa. Finally, we found that dose relationships can create positive feedback loops that facilitate friendly competition (i.e., increased heterospecific density has a positive effect on focal host density because the reduction in disease outweighs the negative effects of interspecific competition). This suggests that without taking dose relationships into account, we may incorrectly predict the effect of heterospecific host interactions, and thus host community composition, on environmentally transmitted parasites.

A high-throughput method to quantify feeding rates in aquatic organisms: A case study with Daphnia

Food ingestion is one of the most basic features of all organisms. However, obtaining precise-and high-throughput-estimates of feeding rates remains challenging, particularly for small, aquatic herbivores such as zooplankton, snails, and tadpoles. These animals typically consume low volumes of food that are time-consuming to accurately measure.We extend a standard high-throughput fluorometry technique, which uses a microplate reader and 96-well plates, as a practical tool for studies in ecology, evolution, and disease biology. We outline technical and methodological details to optimize quantification of individual feeding rates, improve accuracy, and minimize sampling error.This high-throughput assay offers several advantages over previous methods, including i) substantially reduced time allotments per sample to facilitate larger, more efficient experiments; ii) technical replicates; and iii) conversion of in vivo measurements to units (mL-1 hr-1 ind-1) which enables broad-scale comparisons across an array of taxa and studies.To evaluate the accuracy and feasibility of our approach, we use the zooplankton, Daphnia dentifera, as a case study. Our results indicate that this procedure accurately quantifies feeding rates and highlights differences among seven genotypes.The method detailed here has broad applicability to a diverse array of aquatic taxa, their resources, environmental contaminants (e.g., plastics), and infectious agents. We discuss simple extensions to quantify epidemiologically relevant traits, such as pathogen exposure and transmission rates, for infectious agents with oral or trophic transmission.

Genotypic variation in parasite avoidance behaviour and other mechanistic, nonlinear components of transmission

Traditional epidemiological models assume that transmission increases proportionally to the density of parasites. However, empirical data frequently contradict this assumption. General yet mechanistic models can explain why transmission depends nonlinearly on parasite density and thereby identify potential defensive strategies of hosts. For example, hosts could decrease their exposure rates at higher parasite densities (via behavioural avoidance) or decrease their per-parasite susceptibility when encountering more parasites (e.g. via stronger immune responses). To illustrate, we fitted mechanistic transmission models to 19 genotypes of Daphnia dentifera hosts over gradients of the trophically acquired parasite, Metschnikowia bicuspidata. Exposure rate (foraging, F) frequently decreased with parasite density (Z), and per-parasite susceptibility (U) frequently decreased with parasite encounters (F × Z). Consequently, infection rates (F × U × Z) often peaked at intermediate parasite densities. Moreover, host genotypes varied substantially in these responses. Exposure rates remained constant for some genotypes but decreased sensitively with parasite density for others (up to 78%). Furthermore, genotypes with more sensitive foraging/exposure also foraged faster in the absence of parasites (suggesting 'fast and sensitive' versus 'slow and steady' strategies). These relationships suggest that high densities of parasites can inhibit transmission by decreasing exposure rates and/or per-parasite susceptibility, and identify several intriguing axes for the evolution of host defence.