Parasite diversity and coinfection determine pathogen infection success and host fitness

Abstract

While the importance of changes in host biodiversity for disease risk continues to gain empirical support, the influence of natural variation in parasite diversity on epidemiological outcomes remains largely overlooked. Here, we combined field infection data from 2,191 amphibian hosts representing 158 parasite assemblages with mechanistic experiments to evaluate the influence of parasite richness on both parasite transmission and host fitness. Using a guild of larval trematode parasites (six species) and an amphibian host, our experiments contrasted the effects of parasite richness vs. composition, observed vs. randomized assemblages, and additive vs. replacement designs. Consistent with the dilution effect hypothesis extended to intrahost diversity, increases in parasite richness reduced overall infection success, including infections by the most virulent parasite. However, the effects of parasite richness on host growth and survival were context dependent; pathology increased when parasites were administered additively, even when the presence of the most pathogenic species was held constant, but decreased when added species replaced or reduced virulent species, emphasizing the importance of community composition and assembly. These results were similar or stronger when community structures were weighted by their observed frequencies in nature. The field data also revealed the highly nested structure of parasite assemblages, with virulent species generally occupying basal positions, suggesting that increases in parasite richness and antagonism in nature will decrease virulent infections. Our findings emphasize the importance of parasite biodiversity and coinfection in affecting epidemiological responses and highlight the value of integrating research on biodiversity and community ecology for understanding infectious diseases.

Fig. 1.

(A) Total parasite persistence and (B) persistence of each parasite species in metamorphosed Pseudacris regilla exposed to different levels of parasite richness. Persistence values (±1 SE) represent residuals from regression analyses with time to metamorphosis as a covariate. Infectious cercariae of Ribeiroia ondatrae (C), Echinostoma trivolvis (D), Alaria sp. 2 (E), Cephalogonimus americanus (F), Clinostomum attenuatum (G), and an undescribed echinostome magnacauda (H).

Fig. 2.

Survival of Pseudacris regilla exposed to (A) different levels of parasite richness, (B) increasing dosages of three parasites (Ribeiroia, Echinostoma, and Alaria), and (C) parasite treatments representing an additive vs. substitutive design. Parasite exposure dosage interacted with parasite identity to determine mortality (Firth-adjusted χ² = 62.27, df = 5, P < 0.0001; exposure χ² = 10.67, P = 0.0011, Ribeiroia χ² = 12.23, P = 0.0005, exposure*Ribeiroia χ² = 9.32, P = 0.0023). In C, tadpoles were exposed to differing dosages (30 or 120) and combinations of Ribeiroia (R) Echinostoma (E), Alaria (A), and Clinostomum (C). The REAC treatment received 30 of each parasite. Data represent proportional survival within each treatment with 95% binomial confidence intervals.

Fig. 3.

(A) Relationship between mean parasite load (+1 SE) in recently metamorphosed Pseudacris regilla and parasite species richness at the wetland. (B) Parasite community compositions from recently metamorphosed P. regilla representing 134 sampled wetlands; the number of wetlands with one through five parasite species were 38, 53, 39, 23, and 1, respectively. For a given level of parasite richness, the percentage of ponds containing each species is represented by the size of the circle. The parasite communities were highly nested with an observed matrix temperature of 9.44 °C.