Preliminary Insights Into the Relationship Between the Gut Microbiome and Host Genome in Posttraumatic Stress Disorder

Abstract

Posttraumatic stress disorder (PTSD) may develop following trauma exposure; however, not all trauma-exposed individuals develop PTSD, suggesting the presence of susceptibility and resilience factors. The gut microbiome and host genome, which are interconnected, have been implicated in the aetiology of PTSD. However, their interaction has yet to be investigated in a South African population. Using genome-wide genotype data and 16S rRNA (V4) gene amplicon sequencing data from 53 trauma-exposed controls and 74 PTSD cases, we observed no significant association between the host genome and summed abundance of Mitsuokella, Odoribacter, Catenibacterium and Olsenella, previously reported as associated with PTSD status in this cohort. However, PROM2 rs2278067 T-allele was significantly positively associated with the summed relative abundance of these genera, but only in individuals with PTSD and not trauma-exposed controls (p < 0.014). Polygenic risk scores generated using genome-wide association study summary statistics from the PGC-PTSD Overall Freeze 2 were not predictive of gut microbial composition in this cohort. These preliminary results suggest a potential role for the interaction between genetic variation and gut microbial composition in the context of PTSD, underscoring the need for further investigation.

Keywords: PROM2; PTSD; gastrointestinal microbiome; host genome‐microbiome interaction; polygenic risk scores.

FIGURE 1

The relationship between the summed relative abundance of four genus‐level taxa and PTSD case/control status, CAPS‐5 total scores and CTQ scores in N = 127 participants (PTSD = 74, TEC = 53). This figure is comparable to Figure 3 of Malan‐Müller et al. [13] (Figure S2; N = 137), although only a subset (N = 127) with both microbiome data and host genome data was used in the current analysis. (A) The summed relative abundance of the four genera was higher in individuals with PTSD compared to TEC (Wilcoxon, p = 0.001) and (B) was positively correlated with the overall CAPS‐5 total score (Spearman r _s  = 0.28; p = 0.001 [purple]), although not when stratified into PTSD cases (Spearman r _s  = 0.124; p = 0.293 [red]) and controls (Spearman r _s  = −0.049; p = 0.725 [blue]). (C) The summed relative abundance was not significantly correlated with CTQ score (Spearman r _s  = 0.14; p = 0.09). CAPS‐5, Clinician‐Administered Posttraumatic Stress Disorder Scale for DSM‐5; CTQ, Childhood Trauma Questionnaire; IQR, interquartile range; TEC, trauma‐exposed control.

FIGURE 2

The minor allele of 10 SNPs was associated with the total summed relative abundance of four genus‐level taxa (p < 0.05). (A) rs2710119 (CNTNAP2) (CC = 35, CT = 73, TT = 19). (B) rs2278067 (PROM2) (GG = 56, GT = 58, TT = 13). (C) rs61906997 (AA = 99, AG = 25, GG = 3). (D) rs3775467 (MMRN1) (CC = 76, CA = 47, AA = 4). (E) rs13249293 (GG = 124, GC = 3). (F) rs2630788 (ZNF385D) (AA = 108, GA = 18, GG = 1). (G) rs986417 (CC = 100, CT = 26, TT = 1). (H) rs115795847 (CC = 107, CT = 19, TT = 1). (I) rs11236806 (LINC02757) (AA = 82, AG = 41, GG = 4). (J) rs17806643 (AA = 98, AG = 27, GG = 2). Significance was not retained following Bonferroni correction for multiple testing (p > 0.0003). Data are represented by box plots overlayed by individual data points. SNPs without names are unmapped.

FIGURE 3

PROM2 rs2278067, but not CNTNAP2 rs2710119, interacted with PTSD case/control status to influence the total relative abundance of four genus‐level taxa. (A) rs2710119 (CNTNAP2) did not interact with PTSD status to influence the sum of the four genus‐level taxa (p > 0.05) (Control [blue], CC = 16, CT = 30, TT = 7; PTSD [red], CC = 19, CT = 43, TT = 12). (B) The summed relative abundance of the four genus‐level taxa increased with every copy of the rs2278067 (PROM2) T allele in PTSD cases but not in TEC (β = 1.72, SE = 0.69, t = 2.49, p _interaction = 0.0138, F _(9,117) = 3.73, p _model = 0.0004) (Control [blue], GG = 30, GT = 18, TT = 5; PTSD [red], GG = 26, GT = 40, TT = 8). Data are represented by box plots overlayed by individual data points.

FIGURE 4

The PRS for PTSD was not associated with PTSD case/control status or the relative abundance sum of four genus‐level taxa. PTSD‐PRS ~ case/control status, p = 0.309 (left‐hand panel); PTSD‐PRS ~ sum of four genera, p = 0.377 (right‐hand panel).