A comparison of bats and rodents as hosts for zoonotic viruses: What characteristics make a good reservoir host?

Background/Question/Methods

With the recent discoveries of bats as important reservoir hosts of several emerging infectious viruses, such as SARS coronavirus, Ebola, Marburg, Nipah, and Hendra viruses, we are led to ask the questions: Are bats special in some way? Are they better reservoir hosts than other taxa? If so, which characteristics make bats better reservoirs for zoonotic viruses? In order to explore these questions, we compare bats (Order Chiroptera) and rodents (Order Rodentia) as reservoirs of zoonotic viruses. Using generalized linear models (GLMs), we examined correlates of the number of zoonotic viruses hosted by a genus or species. Traits examined included number of litters per year, litter size, longevity, body mass, and torpor use. We performed the GLMs on bats alone, on rodents alone, and on both taxa together.  Models were compared using AIC.

Results/Conclusions

Bats were found to be more likely to have a greater number of zoonotic viruses than rodents. The best model for the number of zoonotic viruses in bats included torpor use (a categorical variable for no torpor use, some torpor, or true hibernation), maximum longevity, and number of litters per year, explaining roughly half of the variation seen. The best model for rodents included number of litters per year, explaining 20% of the variation seen. The best model for bats and rodents together included order, torpor use, litters per year, and maximum longevity. These results suggest that bats indeed may be special, but what makes a good reservoir host for zoonotic viruses may differ between bats and rodents. Simple mathematical models are presented to explore how these host characteristics impact the disease dynamics.