An investigation of disease reservoirs in complex ecosystems: rabies and canine distemper in the Serengeti

Rabies virus (RABV) and canine distemper virus (CDV) have been implicated as the cause of dramatic declines in wildlife populations worldwide and rabies is considered a growing public health problem throughout much of the developing world. In Tanzania, recent severe epidemics of both diseases in Serengeti carnivores have been associated with the large population of domestic dogs (Canis familiaris) living in proximity to the protected areas, but many questions remain about their epidemiology, particularly with respect to the understanding of reservoir infection dynamics and patterns of viral maintenance and transmission in multi-host communities. This study examines all the available evidence for reservoirs of rabies in the Serengeti and presents data on the temporal and spatial dynamics of CDV in domestic and wild carnivores (dogs and lions [Panthera leo]) to help understand long-term patterns of infection in the two populations.Practical difficulties in detecting rabies in areas such as the Serengeti, where surveillance and laboratory confirmation of disease are severely constrained, limit the collection of epidemiological data, a critical step in identifying reservoirs of infection. A novel direct rapid immunohistochemical test (dRIT) as a field test for rabies surveillance was therefore investigated. Preliminary evaluation on frozen and glycerolated field brain samples under field and laboratory conditions showed a sensitivity and specificity equivalent to those of the direct fluorescent antibody (DFA) test, the gold standard in rabies diagnosis. Examination by molecular phylogenies of the genetic characteristics of RABVs isolated from a range of species revealed one single major variant belonging to the group of southern Africa canidassociated viruses (Africa lb), a high degree of genetic relatedness among viruses with no evidence for distinct virus-host associations, and patterns consistent with temporal direction of evolutionary change from dogs to other species. Overall, these analyses point to the domestic dog being responsible for supporting the cycle of a single virus variant in the ecosystem. Rabies incidence data available from the Serengeti (1991-2005) and data on the genetic characteristics of the virus were then used to draw conclusions on reservoirs of infection: domestic dog populations occurring at high densities were the only population essential for persistence, whereas other carnivores contributed to the reservoir as non-maintenance components. Serological and case morbidity/mortality data on CDV indicated that infection patterns in unvaccinated dog and lion populations were consistent with periodic re-introductions with no evidence for persistent infection, re-introduction of infection in vaccinated dog populations coincided with declines in vaccination coverage, but patterns of exposure were not different in vaccinated and unvaccinated populations, and circulation in lions did not appear to occur in the absence of infection in dogs, suggesting that dogs may be the only source of infection. The lack of evidence for long-term persistence in any of the populations suggests that no single population may be capable of independent maintenance, but a network of populations may constitute a maintenance community.Finally, the overall results are discussed in relation to the surveillance of rabies and the design of appropriate control strategies for rabies and CDV for the Serengeti ecosystem and areas of sub-Saharan Africa where both diseases occur. The qualities of the dRIT for rabies surveillance in field conditions and countries with limited diagnostic infrastructures and the potential benefits of its wider application in developing country settings are highlighted. Efforts directed at controlling infection in domestic dogs through mass vaccination programmes are expected to have the most significant impact on reducing or eliminating disease in all the other species. Elimination of CDV in dog populations would also provide definite insights into their role in disease persistence within a potentially complex reservoir system but questions remain about the cost-effectiveness of such an approach as a long-term management strategy for African wild carnivore populations