The diversity of microparasites of rodents: a comparative analysis that helps in identifying rodent-borne rich habitats in Southeast Asia

Abstract

Background: Predicting habitats prone to favor disease transmission is challenging due to confounding information on habitats, reservoirs, and diseases. Comparative analysis, which aims at investigating ecological and evolutionary patterns among species, is a tool that may help. The emergence of zoonotic pathogens is a major health concern and is closely linked to habitat modifications by human activities. Risk assessment requires a better knowledge of the interactions between hosts, parasites, and the landscape.

Methods: We used information from a field spatial study that investigated the distribution of murid rodents, in various habitats of three countries in Southeast Asia, in combination with their status of infection by 10 taxa of microparasites obtained from the literature. Microparasite species richness was calculated by rodent species on 20,272 rodents of 13 species. Regression tree models and generalized linear models were used to explain microparasite diversity by the average distance between the trapping site and five categories of land cover: forest, steep agriculture land, flat agriculture land, water, and built-up surfaces. Another variable taken into account was the slope.

Results: We found that microparasite diversity was positively associated with flat agriculture land, in this context mainly rice fields, and negatively associated with slope. Microparasite diversity decreased sharply a 100 m or less from flat agriculture land.

Conclusion: We conclude that there is high microparasite circulation in rodents of flooded farmlands, meaning possibly a higher risk of disease for human inhabitants.

Keywords: comparative analysis; landscape; rodent-borne diseases; transmission ecology.

Fig. 1

Distribution of rodent species according to habitat types: paddy fields (lowland rain-fed), non-flooded lands, forests, households and settlement) on the two first axes of a principal component analysis. The axis 1 and 2 accounted for 85% of the variance. (B.ind: Bandicota; B.sav: Bandicota savilei; B.berd: Berrylmys berdmorei; B.bow: Berrylmys bowersi; L.edw: Leopodamys edwarsi; M.sur: Maxomys surifer; M.car: Mus caroli; M.cer: Mus cervicolor; M.coo: Mus cooki; N.fulv: Niviventer fulvescens; R.arg: Rattus argentiventer; R.exu: Rattus exulans; R.los=Rattus losea; R.norv=Rattus norvegcius; R.tan=R. tanezumi).

Fig. 2

Regression tree model explaining distribution of rodents in relation to distance to main habitats: forest, steep agriculture, flat agriculture, settlement, and with slope and elevation.

Fig. 3

Relationship between microparasite species richness and distance to flat agriculture (i.e. irrigated/flooded, paddy rice fields) (A) using raw data (the distribution is fitted to a polynomial regression of second order, R2=0.63, F2,11=8.50, P=0.007) and (B) using independent contrasts (the distribution is fitted to a polynomial regression of second order without intercept, R2=0.41, F2,10=3.40, P=0.07).