Genetics of Tuberculosis Resistance

Abstract

One third of the world’s population has been infected with Mycobacterium tuberculosis (MTB). Most of those exposed develop an asymptomatic latent infection. In the absence of co-morbidities, only 5-10% of people progress to active tuberculosis disease (TB) post-exposure, defined as primary TB. However, immunosuppression resulting from malnutrition or human immunodeficiency virus (HIV) co-infection increases the likelihood of both primary TB and activation of a latent infection. As a result, tuberculosis remains a major global health problem. Worldwide, TB is the second-leading cause of mortality from a single infectious agent, after HIV. In 2013, 9 million new cases of clinical tuberculosis were diagnosed and 1.5 million deaths were attributed to the disease. An estimated 360,000 deaths occurred in people co-infected with HIV, and 75% of these cases occurred in sub-Saharan Africa. MTB infection and tuberculosis disease have a substantial heritable component. In this project, we studied a genetic resistance phenotype to TB disease and MTB infection, as opposed to more standard approaches tailored towards finding loci associated with susceptibility. We recruited HIV-positive patients from three recently concluded prospective cohorts of TB from Uganda and Tanzania. We hypothesized that HIV-positive individuals living in MTB hyperendemic areas who do not develop TB disease represent an extreme resistance phenotype. A study of 175,906 variants in 267 cases and 314 controls revealed a genome-wide significant association of a common TB resistance single nucleotide polymorphism, rs4921437, in the regulatory region of IL12, a gene previously associated with susceptibility. We also hypothesized that HIV-positive individuals who do not establish MTB infection despite living in hyperendemic regions are genetically resistant. In a study of 162,228 variants in 244 cases and 235 controls, we discovered a genome-wide significant association of a resistance variant rs877356 near IL9, a gene previously associated with bronchial hyperresponsiveness and asthma. Furthermore, we evaluated the effects of multi-locus interactions in a candidate gene approach and found that epistasis also plays a significant role in risk of MTB infection and TB disease. We present a novel approach to genome-wide association studies, also applicable to whole genome sequencing studies, where the use of an extreme resistance phenotype allowed us to identify large effect sizes and attain genome-wide significance in cohorts of a relatively small sample size. The use of HIV status as a central feature of our hypothesis as opposed to a confounder or exclusion criterion is also unique.