Post-traumatic stress disorder (PTSD) is a heterogeneous disorder with complex causality, including gene–environment (G × E) interactions.Reference Koenen1 A dissociative subtype of PTSD has been included in DSM-5.2 Identifying mechanisms to better understand the biology of this dissociative subtype could advance treatments.
One plausible biological mechanism of dissociation is dysregulation of the oxytocin system,Reference Seng3 which cascade theory posits to be vulnerable to dysregulation in infancy and early childhood via relational traumaReference Teicher, Andersen, Polcari, Anderson and Navalta4 and which is a biological system related to many survival functions affected by trauma.Reference Porges5 Oxytocin is associated with caregiving and attachment,Reference Brown and Michael Brown6 and has parasympathetic roles in recovery from stress.Reference Porges5,Reference Moberg7 The oxytocin system is one of the three pillars of self-regulation posited by cascade theory which adapts in the context of trauma associated with childhood maltreatment.Reference Teicher, Andersen, Polcari, Anderson and Navalta4 Oxytocin makes theoretical sense as a common, cross-cutting element of the PTSD dissociative subtype. A small number of studies suggest that the oxytocin receptor gene (OXTR) may be relevant to PTSD and dissociation;Reference Apazoglou, Adouan, Aubry, Dayer and Aybek8,Reference Reiner, Frieling, Beutel and Michal9 one links a single nucleotide polymorphism (SNP) in OXTR (rs53576) with an unresolved attachment status and higher depersonalisation and derealisation symptoms,Reference Reiner, Frieling, Beutel and Michal9 and another finds increased methylation in the OXTR promoter in persons with functional neurological disorders.Reference Apazoglou, Adouan, Aubry, Dayer and Aybek8 However, further study of dissociation and OXTR is warranted.
Here we present an exploratory analysis from a community sample of pregnant women of European ancestry to examine the association of OXTR polymorphisms with dissociative symptoms and PTSD diagnosis and to further investigate the G × E interaction of OXTR SNPs with a history of childhood maltreatment.
Method
Design
This was a candidate gene association study of pregnant women, either PTSD-positive or trauma-exposed and PTSD-negative, recruited from a study of the effect of PTSD on perinatal outcomes (US National Institutes of Health, J.S.S. and A.P.K., R01 NR 008767-S2). The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. All procedures involving human participants were approved by the University of Michigan Institutional Review Board (HUM00010610). Written informed consent was obtained from all participants. Details of the parent study methods, including psychometrics, are reported elsewhere.Reference Seng, Low, Sperlich, Ronis and Liberzon10
Sample
This analysis was conducted with salivary DNA from 228 European American women from the PTSD-diagnosed and trauma-exposed but resilient cohorts. Of these, 135 had lifetime PTSD and 93 had been exposed to trauma but did not develop PTSD.
Data collection
The mid-trimester diagnostic interview used validated, reliable measures.Reference Seng, Low, Sperlich, Ronis and Liberzon10 Lifetime PTSD was determined using the National Women's Study DSM-IV PTSD module. Maltreatment was determined by reports on the Life Stressor Checklist of any emotional abuse or neglect, physical abuse or neglect, and contact or penetrative sexual abuse as the index trauma. Dissociation was assessed using the Dissociative Experiences Scale Taxon version (DES-T); a report of ‘sometimes or more often than sometimes’ on the DES-T items for depersonalisation or derealisation was a proxy for the DSM-5 dissociative subtype. The Oragene kit was used for saliva specimen collection (DNA Genotek Inc., Ottawa, Canada).
Genotyping
Genomic DNA was extracted using a semi-automated filter system (QuickGene, Kurabo Ltd, Osaka, Japan), quantified using PicoGreen (Invitrogen, Carlsbad, California, USA) and genotyped using a custom 4800 SNP Infinium bead-array microarray (Illumina, San Diego, California, USA), at the University of Michigan Advanced Genomics Core. Our microarray included nine OXTR SNPs previously associated with PTSD (rs1042778, rs237885, rs237887, rs237888, rs2254298, rs237889, rs53576, rs237895 and rs237897). We used Haploview to determine OXTR SNP linkage disequilibrium structure (www.broadinstitute.org). All OXTR SNPs were found to be in Hardy–Weinberg equilibrium (all P > 0.05). Multiple testing correction accounting for linkage disequilibrium between SNPs was calculated per Nyholt (α < 0.0089 for a type I error rate at 5% for these nine SNPs).
Data analysis
To test associations of OXTR SNPs with the symptom variables, we conducted quantitative trait gene association tests using the symptom counts for PTSD and dissociation for one-way analysis of variance (ANOVA) with a Bonferroni adjustment for post hoc comparisons. Independent samples t-tests determined which genotype was associated with the dissociation phenotype. We used univariate ANOVA to examine the variance explained by the SNP in comparison with childhood maltreatment history and then to assess whether the risk accrued additively or via G × E interaction.
Results
Sample characteristics
Participants had a mean age of 30.9 years. The majority had post-secondary education and household income larger than $15 000. Among these 228 women with trauma exposure, 93 (40.8%) met lifetime PTSD diagnostic criteria. The mean number of lifetime PTSD symptoms was 5.3 (s.d. = 4.96). Dissociation mean score was 1.0 (s.d. = 2.08). Fifteen (16.1%) of the women with PTSD had the dissociative subtype (supplementary Table 1, available at https://doi.org/10.1192/bjo.2022.74).
Gene associations with dissociation
Results of genotypic association with dissociation score are in Table 1. For rs237895 and rs237897, major C allele carriers reported higher levels of dissociation.
Dissociation score stratified by OXTR single nucleotide polymorphism (SNP) genotypes (n = 228)a
a. Results were not corrected for multiple testing, and instead the adjusted P-value of 0.0089 was used at the threshold for significance.
b. The minor allele homozygote (C/C) of rs237888 occurs only once in the sample; thus, the standard deviation was not available.
c. Groups with the same letter represent significant differences as tested by one-way analysis of variance with post hoc analysis with Bonferroni correction.
To put the genotype into biopsychosocial context, we conducted univariate ANOVA models to compare variance in dissociation explained by the genotypes with variance explained by maltreatment (supplementary Table 2). The major alleles and genotypes account for less variance (ε2) than the maltreatment variable. All of these alleles and genotypes appeared to be protective when the individual had a maltreatment history. For dissociation, the major G allele in rs2254298 was protective (P < 0.001) when interacting with maltreatment (ε2 = 6%). The CC homozygote in rs237897 also appeared protective for dissociation, although not significant with the P < 0.0089 multiple testing correction, especially when having more than 1 type of child maltreatment (ε2 = 3%).
Discussion
In this small community sample of European American women, we found a potentially biologically meaningful association of OXTR SNPs rs237895 and rs237897 with dissociative symptoms, and also of the OXTR SNP rs2254298 with dissociation when interacting with maltreatment history. These findings indicate that the oxytocin receptor may be responsive to psychological and/or environmental stressors, a hypothesis also supported by epigenetic studies where DNA methylation in the OXTR gene occurred after both acute psychological stressReference Unternaehrer, Luers, Mill, Dempster, Meyer and Staehli11 and early-life adversity.Reference Apazoglou, Adouan, Aubry, Dayer and Aybek8 Our finding that a high level of dissociative symptoms was associated with OXTR also lends support to theories that alterations in the oxytocin system might play a role in the phenomenon of dissociation characterised by depersonalisation and derealisation symptoms (e.g. cascade theoryReference Teicher, Andersen, Polcari, Anderson and Navalta4 and polyvagal theoryReference Porges5,Reference Tucci, Weller, Mitchell, Porges and Dana12 ).
Findings are similar to other studies of gene associations with PTSD.Reference Ressler, Mercer, Bradley, Jovanovic, Mahan and Kerley13 Our preliminary finding of a gene variant in the oxytocin system is consistent with results on hormone levels observed in previous pilot studies, in which PTSD was associated with low plasma concentration of oxytocinReference Seng, Miller, Sperlich, van de Ven, Brown and Carter14 and the PTSD dissociative subtype was associated with very high levels of oxytocin.Reference Seng, Miller, Sperlich, van de Ven, Brown and Carter14,Reference Munro, Brown, Pournajafi-Nazarloo, Carter, Lopez and Seng15
There are limitations to this non-clinical exploratory study. Recall bias due to self-reporting of maltreatment may have introduced a source of error variance that would decrease power to detect significant relationships. The exclusively European American and exclusively pregnant sample may affect the generalisability. Replication in clinical populations is needed with large samples that include adequate ancestry and gender representation. Given the oxytocin system's multiple functions relevant to dissociation and early-life trauma, studies that consider oxytocin dysregulation as a cross-cutting biological mechanism involved in childhood maltreatment sequelae are warranted.
Supplementary material
Supplementary material is available online at http://doi.org/10.1192/bjo.2022.74.
Data availability
The data that support the findings of this study are available from the corresponding author on reasonable request.
Acknowledgements
The authors acknowledge the work of Sandra M. Villafuere Monje, PhD, in the early phase of this project.
Author contributions
J.S.S. and A.P.K. formulated the research question, and designed and conducted the project. A.P.K. designed the microarray and oversaw laboratory work. H.L. analysed the initial clinical data, genetic data and G × E interactions, and reported the results. Y.L. conducted additional literature review and analyses per current criteria for dissociation. All authors wrote and revised the manuscript.
Funding
This study was funded by the National Institutes of Health, National Institute for Nursing Research grant number NR008767-S2, a genotyping supplement to ‘Psychobiology of PTSD and adverse outcomes of childbearing’.
Declaration of interest
None.
Open access
eLetters
No eLetters have been published for this article.