Typing of Mycobacterium bovis in formalin-fixed, paraffin-embedded tissues from selected wildlife species in the Kruger National Park, South Africa

Abstract

Mycobaterium bovis is the causative agent of bovine tuberculosis (BTB) and it is a member of the Mycobacterium tuberculosis complex (MTBC). This bacterium has a wide host range of which, cattle is considered as the maintenance host. Humans, goats, wildlife, cats, dogs and lions are also susceptible to the bacterium and are considered putative spillover hosts as infection is not confined in these hosts. Mycobacterium bovis is prevalent in developing countries especially in farmed animals. This presents a problem since BTB is a zoonosis. People living in close contact with infected cattle or those who drink unpasteurized milk are at risk of infection. About 10% of cases of human tuberculosis are thought to be caused by M. bovis. In some instances, wildlife provides a reservoir for the pathogen and transmits it to cattle in farms and poses further risk to humans at the wildlife/livestock/human interface. Certain countries like the United Kingdom where BTB was previously eradicated are experiencing substantial increase in BTB infection. This is thought to be a result of wildlife reservoirs that infect farmed animals, especially cattle. Such reservoirs make eradication of the disease extremely difficult and require programmes to be put in place to control spread of the disease. This makes M. bovis a pathogen of economic importance since the programmes may be costly. In addition, wildlife that is infected cannot be exported and this further affects the economy negatively. In order to control the spread of the pathogen, it is essential to determine the source of infection. However, it is difficult to determine the source or to track the spread of BTB especially in wildlife where animals have unrestricted movement. The inability to conduct epidemiological studies of BTB may be a result of the lack of molecular typing methods that allow bacteria to be identified to strain level rapidly and fairly simpler than culture, thus providing much needed information about the pathogen. In recent years, typing of M. bovis isolates to strain level has been made possible by the development of PCR-based technologies such as IS6110 typing and spoligotyping. These technologies were however, found to be unsuitable for differentiating certain species in the MTBC. Newer technologies based on the variable number of tandem repeats (VNTRs) in organisms have been developed and allow for the differentiation of members in the MTBC, which have a high level of genome homology. These technologies include multiple-locus variable number tandem repeat analysis (MLVA) and mycobacterial interspersed repetitive unit (MIRU)-VNTR analysis. It was also discovered that mycobacteria have genomic regions of difference (RD) that could be used to identify the different species of bacteria in the MTBC. Retrospective studies may play a key role in tracing the source of diseases and following the pattern of transmission. However, in most instances, no fresh samples are available for such studies. For this reason, formalin-fixed paraffin-embedded (FFPE) tissue from wildlife in the Kruger National Park (KNP) was used for conducting a retrospective study aimed at determining the epidemiology of M. bovis in the KNP. However, amplification of DNA derived from FFPE tissue for PCR based techniques has been found to be a difficult exercise and not many standard protocols have been developed and validated for the use of such DNA. In this study, different methods of extraction were used to obtain DNA from FFPE tissue since it is difficult to obtain high quality DNA from such tissue, which is degraded. Formaldehyde, the main component of formalin which is used to fix tissue samples, causes degradation and cross-linking of DNA. In addition, previous studies are inconsistent with regards to the best method to use when extracting DNA from FFPE tissue. Three PCR-based techniques were used to type or identify the isolates in order to standardize a protocol for use in typing isolates from FFPE tissue. These techniques included analysis of the RDs, VNTR based methods i.e. MLVA and MIRU-VNTR and spoligotyping. Since there are many factors that influence the quality of FFPE tissue, samples confirmed BTB positive by VNTR analysis, spoligotyping and IS6110 analysis were used in order to optimize a PCR for FFPE tissue. Furthermore, in order to serve as control samples for spoligotyping and analysis of the RDs, DNA obtained from fresh tissue was also used in the study. Despite the various methods used to extract and to type DNA, the DNA from FFPE tissue provided unspecific results that did not allow for an informative retrospective study of M. bovis. This may be due to the fact that the DNA used had a high degree of degradation from prolonged fixation in formalin. Although M. bovis could not be typed in FFPE tissues, it could be identified by analysis of the regions of difference, more specifically the RD9 region. Amplification of RD9 is thus recommended for use in retrospective studies for diagnostic purposes, especially in cases where highly degraded DNA is used. This region (RD9) should however, only be used as a presumptive diagnosis since RD9 also identifies M. africanum, M. microti, M. pinnipedii, M. caprrae and M. bovis BCG. However, RD9 specifically excludes M. tuberculosis. In the SA context, particularly in the KNP, this allows for some sound inferences since the animals are likely to be infected with M. bovis as opposed to M. tuberculosis. This study highlighted statements in previous studies where it was stated that fixation of tissue in formalin should be done in such a way to reduce degradation of DNA in FFPE tissue in order to allow for its use in retrospective molecular studies which may be very insightful in determining the epidemiology of diseases that are difficult to track and/or control. Copyright