Infection-interactions in Ethiopian village chickens

Abstract

Chickens raised under village production systems are exposed to a wide variety of pathogens, and current or previous infections may affect their susceptibility to further infections with another parasite, and/or can alter the manifestation of each infection. It is possible that co-infections may be as important as environmental risk factors. However, in cross-sectional studies, where the timing of infection is unknown, apparent associations between infections may be observed due to parasites sharing common risk factors. This study measured antibody titres to 3 viral (Newcastle disease, Marek's disease and infectious bursal disease) and 2 bacterial (Pasteurella multocida and Salmonella) diseases, and the infection prevalence of 3 families of endo- and ecto-parasites (Ascaridida, Eimeria and lice) in 1056 village chickens from two geographically distinct populations in Ethiopia. Samples were collected during 4 cross-sectional surveys, each approximately 6 months apart. Constrained ordination, a technique for analysis of ecological community data, was used to explore this complex dataset and enabled potential relationships to be uncovered and tested despite the different measurements used for the different parasites. It was found that only a small proportion of variation in the data could be explained by the risk factors measured. Very few birds (9/1280) were found to be seropositive to Newcastle disease. Positive relationships were identified between Pasteurella and Salmonella titres; and between Marek's disease and parasitic infections, and these two groups of diseases were correlated with females and males, respectively. This may suggest differences in the way that the immune systems of male and female chickens interact with these parasites. In conclusion, we find that a number of infectious pathogens and their interactions are likely to impact village chicken health and production. Control of these infections is likely to be of importance in future development planning.

Keywords: Co-infection; Constrained ordination; Ethiopia; Host-parasite interaction; Village chicken.

Fig. 1

Biplots of the first three principal components (scaling 2) showing correlations between variables. Correlations are represented by the angles between the vectors rather than the apices of the lines. Labels a, b and c correspond to the three-dimensional views shown above.

Fig. 2

Triplots showing the relationships between response and explanatory variables between the first two canonical axes before and after incorporating household as a fixed effect in the RDA analysis. Angles between variables reflect their correlations. Solid lines represent pathogens (outcomes); dashed lines represent the explanatory variables.