A genome-wide identified risk variant for PTSD is a methylation quantitative trait locus and confers decreased cortical activation to fearful faces

Abstract

Genetic factors appear to be highly relevant to predicting differential risk for the development of post-traumatic stress disorder (PTSD). In a discovery sample, we conducted a genome-wide association study (GWAS) for PTSD using a small military cohort (Systems Biology PTSD Biomarkers Consortium; SBPBC, N = 147) that was designed as a case-controlled sample of highly exposed, recently returning veterans with and without combat-related PTSD. A genome-wide significant single nucleotide polymorphism (SNP), rs717947, at chromosome 4p15 (N = 147, β = 31.34, P = 1.28 × 10(-8) ) was found to associate with the gold-standard diagnostic measure for PTSD (the Clinician Administered PTSD Scale). We conducted replication and follow-up studies in an external sample, a larger urban community cohort (Grady Trauma Project, GTP, N = 2006), to determine the robustness and putative functionality of this risk variant. In the GTP replication sample, SNP rs717947 associated with PTSD diagnosis in females (N = 2006, P = 0.005), but not males. SNP rs717947 was also found to be a methylation quantitative trait locus (meQTL) in the GTP replication sample (N = 157, P = 0.002). Further, the risk allele of rs717947 was associated with decreased medial and dorsolateral cortical activation to fearful faces (N = 53, P < 0.05) in the GTP replication sample. These data identify a genome-wide significant polymorphism conferring risk for PTSD, which was associated with differential epigenetic regulation and with differential cortical responses to fear in a replication sample. These results may provide new insight into understanding genetic and epigenetic regulation of PTSD and intermediate phenotypes that contribute to this disorder.

Keywords: GWAS; PTSD; epigenetic; fMRI; meQTL.

FIG. 1

Chromosome 4 SNP is Associated with PTSD at a Genome-Wide Level. (A) Manhattan plot for PTSD symptoms as measured by the CAPS in the SBPBC cohort (N = 147) showing a peak on chromosome 4, with top SNP rs717947 (β (SE) = 31.34 (5.19), P = 1.28×10⁻⁸). (B) Regional plot of chromosome 4 showing a peak of SNPs with low p-values near the genome-wide significant hit (rs717947 shown as a blue diamond) and non-coding RNA BC036345 (chr4: 33,897,961–34,041,515). Plot generated using LocusZoom (http://csg.sph.umich.edu/locuszoom/).

FIG. 2

Replication of Genome-wide PTSD-associated SNP and Evidence for an meQTL. Replication results for associations of genome-wide significant SNP, rs717947, with PTSD diagnosis in the (A) discovery cohort (N = 147, OR = 8.6 CI = 3.5–21.6, P = 3.84×10⁻⁶). and (B) replication cohort (females only, N = 2006, OR = 1.25, CI = 1.1–1.5, P = 0.005). (C) Of the 11 CpG within 1 MB of the genome-wide SNP, rs717947 affects expression levels (probe cg09242288) with the risk genotype (TT) showing decreased proportion of methylation (N = 157, β (SE) = −0.02 (0.007), P = 0.002].

FIG. 3

Dorsomedial and Dorsolateral PFC Differentially Respond to Fearful Faces as a function of Genotype. Functional MRI was used to examine whether the PTSD-associated SNP was related to differential brain regional activation with fearful cues. (A) Example of neutral and fearful face stimuli from the fearful faces task. (B) Axial section showing medial and lateral prefrontal signal associated with genotype. Significant clusters associated with genotype are shown at a whole-brain corrected threshold of P < 0.05, N = 53 (33 CC, 16 TC, 4 TT). (C) Dorsolateral and (D) Dorsomedial regions surviving whole brain correction. Images are shown in neurological orientation, on template slices in Montreal Neurological Institute (MNI) space. Correction for multiple comparisons used a combined height-extent threshold calculated using monte carlo simulation in Alphasim, with 1,000 iterations and a cluster-forming threshold of P < 0.01.