Food web structure and microenvironment affect Chagas disease vector infection and abundance in a rural landscape

Abstract

AbstractVector‐borne pathogen transmission is shaped by multiple abiotic and biotic factors. Understanding the relative importance of these factors on vector abundance and infection is important for developing vector‐borne disease control strategies. The crown of the Attalea butyracea palm provides a natural arboreal mesocosm suitable for studying how food web relations and microclimate affect Chagas disease vectors of the genus Rhodnius, which feed on vertebrate blood and interact with many vertebrate and invertebrate species, vegetation, and detritus within the palm crown. We performed a cross‐sectional, observational study of A. butyracea crowns using a community ecology network approach to evaluate abiotic and biotic conditions associated with Rhodnius pallescens abundance and infection with Trypanosoma cruzi, the Chagas disease etiologic agent. We collected 1098 R. pallescens from 105 A. butyracea crowns in rural landscapes of Panama. In a palm subset (N = 49), we recorded microclimate and habitat conditions and counted vertebrate and invertebrate species in order to characterize palm crown food webs. We used food web metrics, namely Generality (average number of prey per predator) and Vulnerability (average number of predators per prey) to evaluate associations between palm community trophic structure, vector predators, vector blood meal species composition, vector abundance, and vector T. cruzi infection. Field data analyzed with generalized linear models showed that vector abundance and infection in a given palm crown were influenced by a combination of geographic location, land‐use type, palm crown animal community composition, and microhabitat conditions. Vector abundance was negatively associated with increased overall palm crown community diversity. However, vector abundance was positively associated with the invertebrate predator and mammal community. R. pallescens Vulnerability score, which measures predation pressure on the vector within the palm crown, was positively associated with vector abundance. Vector infection with T. cruzi was positively associated with mammal presence; vectors in palms with mammals were about 25 times more likely to be infected with T. cruzi compared with vectors in palms without mammals. Our findings suggest that a holistic study of food webs and environmental variables at the local scale can provide useful information for vector‐borne disease management across a wide habitat range.