Abstract Body

Background: Indoleamine 2,3-dioxygenase-1 (IDO) activity, as assessed by plasma kynurenine/tryptophan (KT) ratio, is induced by immune activation and microbial translocation, causes T cell proliferative defects and Th17 depletion, and strongly predicts mortality in treated HIV infection. Whether host genetic variation contributes to IDO activity is unknown.

Methods: We assessed common genetic polymorphisms in association with plasma KT ratio in HIV-infected Ugandans in the UARTO and ARKS cohorts at months 6 and 12 of confirmed antiretroviral therapy (ART)-mediated viral suppression using genome (Illumina OmniExpress) and exome (Affymetrix Axiom) array genotyping. Linear mixed models adjusted for cohort, gender, pregnancy status, and population ancestry were employed for candidate gene and discovery-based analyses. Genome-wide imputation was performed using the 1000 Genomes reference panel. For the candidate gene analysis, genes encoding known factors that induce IDO were included.

Results: Participants (N=597) were mostly female (62%) with a median age of 35, and had advanced disease (median pre-ART CD4+ count 135 cells/mm3 and HIV RNA 5.1 log10copies/mL). For the candidate gene analysis, single nucleotide polymorphisms (SNPs) in TNF (rs17200810a, rs34451538, rs114064880, rs11575838), IFNGR1 (rs276565a), and TLR4 (rs2770148b) genes were significant at Bonferroni P <5E10-5. For the GWAS analysis, one intergenic SNP (between CSPG5 and ELP6) achieved genomewide significance (rs56185965, P=8.7E-09), while several other SNPs achieved near-genomewide significance at P <5E10-7, including genes encoding 2 different protein tyrosine phosphatases involved in reversible phosphorylation during cell signaling, PTPRM (rs75257475b, rs115059620) and PTPRN2 (rs6950107), as well as a SNP downstream of CYP24A1 (rs13041834b), which encodes an enzyme that plays a major role in vitamin D metabolism and lies in an H3K27Ac histone mark-enriched region.

Conclusions: Our candidate gene analysis suggests that host genetic variation may modify the impact of IFN-g, TNF-a, and microbial products (via toll-like receptor 4) on IDO induction. The GWAS analysis also identified potential novel mechanisms by which host genetics may modify IDO induction via protein tyrosine phosphatase or vitamin D pathways. Additional studies are needed to confirm these findings in other populations and to validate their functional impact on IDO induction.

a Genotyped SNP; b In linkage disequilibrium (LD) with a genotyped SNP