Tuberculosis (TB) is caused by an infectious pathogen, Mycobacterium tuberculosis (M.tb) in susceptible individuals, but we cannot yet classify or predict outcomes in those prone to pulmonary TB disease versus those prone to resistance. In part, this reflects knowledge gaps regarding genotypes that may increase susceptibility, and in validated disease correlates (e.g. serum of lung protein biomarkers) measured individually, or combined signatures. We address these knowledge gaps by using Diversity Outbred (DO) mice, a population with abundant genetic diversity and heterozygosity, like the human population. Also, like humans, a low dose M.tb infection of DO mice produces a spectrum of outcomes, from highly susceptible to highly resistant, and many intermediate outcomes. In this proposal, we use the DO population to: 1) Identify and test the capacity of genotypic (alleles and statistically significant loci) to predict outcomes such as diagnostic category (class); and 2) To identify and test lung and serum biomarker (protein) and granuloma signatures to determine diagnostic category (class); and 3) To identify and test serum biomarker (protein) signatures that can forecast disease onset, within a 3-week window before illness manifests clinically. The best performing signatures will be tested using samples from humans. Collectively, results from these studies will generate new translatable knowledge regarding correlates of pulmonary TB (useful for diagnostics), and genotypic and serum protein signatures (useful for prognostics).
Mycobacterium tuberculosis (M.tb) causes tuberculosis (TB) in millions of susceptible humans each year. It is well known that humans respond variably to M.tb infection, yet we are unable to predict outcomes with accuracy. Here, we use the Diversity Outbred (DO) mouse population to identify and test genotypic, serum, and lung biomarker signatures to accurately predict outcomes. Findings are also validated in samples from humans.
