Identification of candidate genes and testing for association with tuberculosis in humans

[摘要] ENGLISH ABSTRACT: This research investigated human candidate genes for susceptibility to tuberculosis and the effectof various factors on time to sputum conversion in the admixed South African Coloured (SAC)population. Population stratification was formally tested and excluded. Population based casecontrolstudies were the primary analysis method with a variety of genotyping methods.Candidate polymorphisms in RANTES, CCR5, CCR2 and SDF1, were not associated withtuberculosis susceptibility. Initially the RANTES polymorphism -403 was found to be associatedwith tuberculosis susceptibility but after the testing of additional samples the association was lost,illustrating the challenges with association studies.The C-type lectins DC-SIGN, encoded by the gene CD209, and L-SIGN are importantpathogen-recognition receptors of the human innate immune response. Both lectins have beenshown to interact with Mycobacterium tuberculosis. CD209 promoter polymorphisms, -336 and -871, were both found to be associated with tuberculosis susceptibility. The haplotype containingCD209 -871G and -336A was strongly associated with the control group. The CD209 -336Aallele has been found to be associated with increased DC-SIGN expression, which may be theunderlying reason for an increased efficiency of host phagocytes.Susceptibility to tuberculosis in mice has recently been attributed to the Ipr1 gene. Eightpolymorphisms in the human homologue, SP110, were investigated, including two novelpolymorphisms. No significant associations were found with any of the polymorphismsinvestigated, including two polymorphisms that were previously found to be associated withtuberculosis susceptibility in West African populations.A cohort of 249 cases from a longitudinal study of first time pulmonary tuberculosis patients wasavailable. The cohort was used to investigate if the vitamin D receptor gene (VDR)polymorphisms FokI, ApaI and TaqI were associated with tuberculosis susceptibility or time tosputum conversion, and to investigate other clinical and demographic factors affecting the rate ofresponse to treatment. No association between the VDR genotype and tuberculosis was found inthe case-control study. The cohort allowed for a reliable conversion time to be determined forsmear (n=220) and culture (n=222). Analysis was carried out to determine which factors,including VDR FokI, ApaI, and TaqI genotypes, contribute to faster mycobacterial resolution insputum. This was done by survival curves and Cox regression models. The results indicate thatthe extent of disease at diagnosis was predictive of both smear and culture conversion times inthe final models. Smoking status and VDR genotype contributed independently to smearconversion time, with ApaI 'AA' and TaqI 'T' containing genotypes being predictive of a fasterresponse to tuberculosis therapy. We can conclude that the time taken for an individual toconvert to sputum negativity while on DOTS therapy, can be independently predicted by theVDR genotype. This may have implications for future immunomodulatory therapies.Identifying what contributes to susceptibility to tuberculosis will provide us with a betterunderstanding of the human immune response to tuberculosis which may lead to thedevelopment of accurately targeted therapeutics and vaccines.

[发布机构] Stellenbosch University