The contribution of wild mammals to the epidemiology of tuberculosis (Mycobacterium bovis) in New Zealand : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy at Massey University
The objective of these studies was twofold. The primary aim was to gain a better understanding of the role of free-living mammalian species, other than possums, in the epidemiology of wildlife tuberculosis in New Zealand. The other objective was to continue the operation of the Castlepoint longitudinal so that hypotheses regarding the epidemiology of M. bovis infection in possums could be further refined and clarified. Of the wild carnivores found in New Zealand, the disease persists at high prevalence only in ferrets, and is probably maintained principally by ingestion of tuberculous carrion. Although a moderate number of ferrets excrete M. bovis orally, there appears to be only minor intraspecific transmission by bite wounding. Although cats and stoats can also become infected through scavenging, they appear to be less susceptible to oral infection than ferrets. There is no substantial evidence to suggest that any of New Zealand's free-living carnivores are likely to be reservoir hosts of M. bovis. Observational studies involving twelve domestic red deer suggested that cervids probably become infected through close inspection and investigation of moribund tuberculous possums, and that the likelihood of exposure to M. bovis was related to the curiosity and social ranking of the deer. Necropsies conducted principally on wild red deer and involving 152 animals provided evidence to suggest that significant bacillary excretion from infected deer was uncommon, and that only the few with advanced disease had the potential to be highly infectious. However, behavioural phenomena and disease characteristics preclude the ready transmission of disease amongst cohorts. There is now strong evidence to suggest that a high prevalence of tuberculosis infection in wild deer can only be maintained through contact with infected possums. However, deer may still be able to maintain the disease amongst themselves, albeit at a low prevalence, in the absence of infection in possums. This study also confirmed the importance of lymphoepithelial tissues, such as the oropharyngeal and nasopharyngeal tonsils, as primary sites for the establishment of M. bovis infection, and the subsequent excretion of organisms in deer. The gross and histopathological appearance of the lesions found in six infected hedgehogs are described. It is likely that infection arose from the scavenging behaviour of hedgehogs. The moderate prevalence (3.9%) of tuberculosis in these animals, combined with their small home ranges may allow them to be used successfully in wildlife surveys to pinpoint the locality in which tuberculous possums have died. To gain an understanding of the potential role of wild pigs, goats, sheep, rabbits, hares, rats and mice in the dynamics of Mycobacterium bovis infection in free-ranging animals, numbers of these species were examined for evidence of infection. Of these, only the pig appears to have sufficient potential for intraspecific transmission to be of concern in tuberculosis control programmes. Sheep and goats appear to be simply spillover hosts, which may have a limited role in disease amplification following possible, but limited, intraspecific transmission. Rodents and lagomorphs are most unlikely to play any substantial role in the epidemiology of tuberculosis in New Zealand, under current circumstances. A longitudinal study was established in 1989 to examine the disease behaviour in an infected possum population on a farm in the southern North Island of New Zealand, by trapping, using a fixed set of 295 traps for at least 3 days per month. Animals captured were examined at 2 monthly intervals for evidence of tuberculosis. During the first 5.5 years of this project over 900 individual possums were captured and tagged. Blood was collected from each possum examined, and the sera retained were stored frozen. Using these stored sera, three indirect ELISAs were evaluated as diagnostic tests for tuberculosis in possums. All ELISAs had low sensitivity when a cutoff selected to maximise the specificity was chosen. None of the ELISAs reliably detected possums infected with tuberculosis and they therefore have limited value for epidemiological studies. The lymphocyte transformation assays performed on blood taken from possums was estimated to have a sensitivity for detection of tuberculosis of approximately 80%, when the specificity was set at 99%. The lymphocyte transformation assay was the best of the in vivo tests evaluated, with the moderate sensitivity allowing it to be used with a degree of confidence to retrospectively diagnose disease, and aid the development of hypotheses regarding the epidemiology of tuberculosis in possums. The evaluated tests were applied retrospectively to sera and blood samples from possums from the Castlepoint longitudinal study. The additional data arising from these assays suggested that perhaps as few as one fifth of study site possums which had contact with M. bovis had been previously detected as infected by clinical examination. A proportion of these test positive/examination negative animals may have been exhibiting resistance to M. bovis infection, and/or had resolved lesions or cryptic infection. Such animals may have formed a pool of possums in which future reactivation of tuberculosis was possible. The time from earliest evidence of infection till death, in those possums which showed clinical disease, varied from months to several years. Cortisol assays performed on stored sera, and monitoring of trends in body weight, were used to investigate the role of stressful environmental phenomena in the epidemiology of tuberculosis in possums. Major stressful periods involving inadequate nutrition, heat, cold and moisture stress appear to precipitate severe tuberculosis outbreaks, which are believed to have their origins in the reactivation of subclinical/latent infection in the population. As the period of pre-clinical disease varies substantially, and can be as long as several years, this epidemic of tuberculosis takes several years to subside. Thereafter a small number of clinically diseased possums are likely to be restricted to "hot spots" conducive to transmission of M. bovis. Isolates of M. bovis recovered from a variety of species, both wild and domestic, in the Castlepoint environs, and in particular the Castlepoint study site, were subjected to restriction endonuclease analysis to DNA fingerprint the strains present, and hence gain a better understanding of the inter- and intraspecific epidemiology of tuberculosis. The results do not challenge the accepted view of possums being the major reservoir hosts of tuberculosis in the Wairarapa. There was also no evidence to suggest that host adaptation of M. bovis has occurred, except in the case of possums, where they appear to be able to maintain clusters of individuals infected with particular restriction types, in microhabitats for at least 5 year periods. The occurrence of newly introduced restriction types has made possible new observations on the epidemiology of infection, including the documentation of the occurrence of latent infections, duration of primary progressive disease in newly infected possums (7-8 months), and the likely occurrence of post-primary reactivation of tuberculosis.