Using host species traits to understand the consequences of resource provisioning for host-parasite interactions.

Daniel J Becker,Sonia Altizer,Daniel G Streicker,Elizabeth Derryberry

doi:10.1111/1365-2656.12765

Abstract

Supplemental food provided to wildlife by human activities can be more abundant and predictable than natural resources, and subsequent changes in wildlife ecology can have profound impacts on host–parasite interactions. Identifying traits of species associated with increases or decreases in infection outcomes with resource provisioning could improve assessments of wildlife most prone to disease risks in changing environments.We conducted a phylogenetic meta‐analysis of 342 host–parasite interactions across 56 wildlife species and three broad taxonomic groups of parasites to identify host‐level traits that influence whether provisioning is associated with increases or decreases in infection.We predicted dietary generalists that capitalize on novel food would show greater infection in provisioned habitats owing to population growth and food‐borne exposure to contaminants and parasite infectious stages. Similarly, species with fast life histories could experience stronger demographic and immunological benefits from provisioning that affect parasite transmission. We also predicted that wide‐ranging and migratory behaviours could increase infection risks with provisioning if concentrated and non‐seasonal foods promote dense aggregations that increase exposure to parasites.We found that provisioning increased infection with bacteria, viruses, fungi and protozoa (i.e. microparasites) most for wide‐ranging, dietary generalist host species. Effect sizes for ectoparasites were also highest for host species with large home ranges but were instead lowest for dietary generalists. In contrast, the type of provisioning was a stronger correlate of infection outcomes for helminths than host species traits.Our analysis highlights host traits related to movement and feeding behaviour as important determinants of whether species experience greater infection with supplemental feeding. These results could help prioritize monitoring wildlife with particular trait profiles in anthropogenic habitats to reduce infectious disease risks in provisioned populations.

Highlights

Resource availability affects population and community dynamics by affecting reproduction, dispersal and trophic interactions (Yang, Bastow, Spence, & Wright, 2008)
Effect sizes for ectoparasites were highest for host species with large home ranges but were instead lowest for dietary generalists
We considered univariate mixed-effects models (MEMs) of each trait, biologically meaningful interactions between traits where collinearity was weak and that were supported by the sample size, and interactions between traits and supplemental feeding type (Appendix S4, Tables S3 and S4)

Summary

| INTRODUCTION

Resource availability affects population and community dynamics by affecting reproduction, dispersal and trophic interactions (Yang, Bastow, Spence, & Wright, 2008). Greater resource access and better nutrition can improve host immune defence by increasing pathogen clearance or resistance to infection, which can reduce transmission (Forbes et al, 2016; Wilcoxen et al, 2015) Understanding when these different mechanisms will dominate is important for predicting how provisioning will affect infection risks for wildlife, humans and domestic animals (Epstein et al, 2008; Lawson et al, 2012). By accounting for effects of host and parasite biology, we provide important steps to establish a framework for predicting which wildlife species experience greater infection risks with resource provisioning and supplemental feeding. We used the r package metafor to convert effect sizes (R Core Team, 2013; Viechtbauer, 2010)

| MATERIALS AND METHODS

Findings

| DISCUSSION

| CONCLUSIONS