Mycobacterium tuberculosis-dependent expression quantitative trait loci in monocytes are associated with cytokine signaling and resistance to TST/IGRA conversion
Hong, H.1,2*, Dill-McFarland, K.A.1, Benchek, P.3, Simmons, J.D.1, Peterson, G.J.1, Mayanja-Kizza, H.5, Boom., W.H.4, Stein, C.M.3,4, Hawn, T.R.1
1) TB Research & Training Center, Department of Medicine, University of Washington, Seattle, WA, USA.
2) Biobehavioral Nursing and Health Informatics, School of Nursing, University of Washington, Seattle, WA, USA.
3) Department of Population & Quantitative Health Sciences, Case Western Reserve University, Cleveland, OH, USA.
4) Department of Medicine, Case Western Reserve University, Cleveland, OH, USA.
5) Department of Medicine, School of Medicine, Makerere University, Kampala, Uganda.
Mycobacterium tuberculosis (Mtb) and myeloid cells have co-evolved for thousands of years with genetic pressure that selects genes and variants which modulate host immunity and may regulate clinical outcomes. We hypothesized that monocytes have Mtb-dependent expression quantitative trait loci (eQTL) in genes with single nucleotide polymorphisms (SNP) that are also associated with Mtb-induced cytokine expression and clinical outcomes. We collected monocytes from 61 Ugandan household contacts (HHCs) of pulmonary TB index cases, infected with Mtb (H37Rv), and measured transcriptional profiles by RNAseq at 6 hours. We defined cis-eQTLs against SNPs from MEGAEX genotyping that are associated with gene expression from RNAseq and found 1191 eQTLs in uninfected and 1011 eQTLs in Mtb-infected monocytes (FDR<0.01). We further identified Mtb-dependent eQTLs associated with gene expression in Mtb-infected, but not uninfected monocytes by evaluating the interaction effect of SNPs between uninfected and infected monocytes. Of those 1011 eQTLs in Mtb-infected monocytes, we found that 26 Mtb-dependent eQTLs were associated with the expression of 14 genes in response to Mtb infection (Mtb-dependent eQTL genes, FDR <0.1). Although most people heavily exposed to Mtb become infected and develop latent TB infection (LTBI), we previously found a minority of HHCs who have a persistently negative TST/IGRA (labeled “resisters” or RSTR) in Uganda. To assess the clinical significance of Mtb-dependent eQTL genes, we used a candidate gene association study with RSTR (n=74) or LTBI (n=189) subjects. We found that 10 Mtb-dependent eQTL genes contained 14 SNPs associated with the clinical phenotype of RSTR vs. LTBI status (p <0.05, adjusted for age, sex, and kinship). Furthermore, for 5 of these Mtb-dependent eQTLs, the uncommon allele was associated with decreased levels of Mtb-infected cytokine expression in monocytes (linear model, TNF, IFNβ, IL6, and IL1β). Together, these data suggest a small number of monocyte genes have eQTLs that are genetically programmed to respond to Mtb and associated with clinical susceptibility phenotypes (RSTR vs. LTBI). Of these, 5 Mtb-dependent eQTL genes (SIRPB1, RTCA, CHMP1B, PEX7, and MED27) may regulate cytokine signaling pathways. Such functional variation may underlie inter-individual variation in susceptibility to Mtb infection, providing insight into targets for host-directed therapies.