Post-traumatic stress disorder (PTSD) is a debilitating, stress-related psychiatric condition, which occurs among persons exposed to traumatic events involving life threats, serious injury, or death (Matosin et al., 2017). It develops as a result of failure to contain the normal stress response hence dysregulation of the hypothalamic–pituitary–adrenal (HPA) axis which is one of the body’s major stress response systems (Wong, 2002; Yehuda and LeDoux, 2007). Although nearly 90% of individuals are exposed to a traumatic event during their lifetime (Breslau et al., 1998), only a small proportion go on to develop PTSD (Matosin et al., 2017). There is significant variation in the risk of PTSD among individuals experiencing the same trauma; this is likely determined by genetic predisposition and epigenetic mechanisms (Kessler et al., 2017). One of the best understood epigenetic mechanisms is DNA methylation through the addition of a methyl chemical bond to the cytosine C5 in the cytosine-phosphate-guanine (CpG) dinucleotides of the DNA, influencing gene activity and considered to be a stable epigenetic mark in post-mitotic cells (Guo et al., 2011). Generally, DNA methylation, especially if occurring close to the promoter region of the gene is known to inhibit gene transcription.

Epigenome-wide Association Studies (EWAS) provide an unbiased approach to identify DNA methylation loci associated with a disease independently of what is known about the pathophysiology (Almli et al., 2014). Most EWAS examining epigenetic effects of trauma exposure use DNA methylation profiles in peripheral tissues like blood and saliva, which are useful for biomarker identification (Almli et al., 2014). Several genes previously associated with stress or epigenetic regulation of neuronal function have been identified through EWAS of PTSD (Melroy-Greif et al., 2017; Nievergelt et al., 2018; Howie et al., 2019). Additionally, multiple genome-wide studies have also uncovered genetic contributions to PTSD risk and symptomatology through candidate genes which have encoded for varied proteins (Gormley et al., 2016; Kilaru et al., 2016; Blacker et al., 2019). EWAS show that there are significant epigenetic differences between individuals with PTSD compared to healthy controls (Uddin et al., 2010; Smith et al., 2011), particularly in genes involved in inflammation, immune and nervous system function (Uddin et al., 2010; Smith et al., 2011; Rusiecki et al., 2012). Psychosocial stress may alter global and gene-specific DNA methylation patterns potentially associated with peripheral immune dysregulation (Uddin et al., 2010; Smith et al., 2011; Rusiecki et al., 2012).

A larger number of studies have investigated DNA methylation of specific candidate genes, including genes involved in the regulation of the HPA axis (Almli et al., 2014). Epigenetic DNA methylation changes may accompany lifetime experiences and alter gene expression profiles (Yehuda et al., 2015). Significant methylation changes in early life, specifically in genes implicated in developing severe psychiatric conditions including DLG4, DRD2, NOS1, NRXN1, and SOX10 also indicate vulnerability to the effects of stress and psychiatric disorders via epigenetic mechanisms (Numata et al., 2012). Large longitudinal twin and molecular genetic cohort studies suggest that the impact of adverse life events is probably moderated by genetic variants through genetic and environmental interactions (Meaney and Szyf, 2005; Probst et al., 2009).

Post-traumatic stress disorder is associated with the occurrence of multiple comorbidities including depression and coronary heart disease which have been well studied (Sareen, 2014). PTSD is also highly comorbid with chronic pain conditions that often co-occur such as migraine headaches, tension headaches, temporomandibular disorder, irritable bowel syndrome, fibromyalgia, chronic fatigue syndrome and chronic prostatitis/chronic pelvic pain syndrome (Gasperi et al., 2021). Patients with PTSD have a risk of developing pain disorders, which may produce long-lasting changes in the threshold for migraine attacks by inducing epigenetic modifications throughout the brain (Eising et al., 2013). Although several epidemiological studies have reported that PTSD is a predictor of migraine and is much more prevalent in patients with migraine than in the general population (Peterlin et al., 2008; Smitherman et al., 2009; Minen et al., 2016; Zarei et al., 2016), migraine is an understudied comorbidity of PTSD.

Migraine may be aggravated by stress, exercise, sleep deficiency, hormonal changes, head traumas, major depression, PTSD and environmental cues (Theeler et al., 2012; Guglielmetti et al., 2020). In addition, migraine-related pain may cause sensitisation of certain pain pathways via inflammation-induced changes in epigenetic gene regulation (Eising et al., 2013). A recent GWAS reported that migraine showed a higher genetic correlation with psychiatric disorders when compared to other neurological disorders, suggesting common genetic basis or pathways (Anttila et al., 2018). Little is understood about the link between PTSD and migraine disorders, however, it has been suggested that several systems such as the immune system are likely to be involved in the co-occurrence of these disorders (Peterlin et al., 2011). Biological, environmental and genetic risk factors may converge to produce a brain state which predisposes an individual to both PTSD and migraine (Antonaci et al., 2011). Causal pathways shared between migraine and its comorbid disorders may be modulated by epigenetic mechanisms (Eising et al., 2013); these have been suggested to play a role in development of both disorders (McGowan, 2013).

It is important to understand the role genetic, environmental and epigenetic factors play in determining an individual’s susceptibility to PTSD and other co-occurring symptoms (Blacker et al., 2019). In this study we used the disease-discordant monozygotic (MZ) twin design to investigate PTSD-migraine comorbidity. This is a powerful strategy in genetic and epigenetic epidemiology as participants are genetically identical and well-matched by age, sex, maternal environment, population cohort effects and exposure to many shared environmental factors. Recent studies have used this design and uncovered considerable epigenetic (methylation) variation between MZ twins for several complex phenotypic traits which are detectable in blood DNA samples (Bell and Spector, 2011; Bolund et al., 2017; Kaut et al., 2017; Gerring et al., 2018; Peng et al., 2018; Gasperi et al., 2021).

The study aimed to identify PTSD associated genes overlapping with migraine and evaluate similarity in biological pathways between these disorders. To the best of our knowledge this is the first study to use the monozygotic twin design to determine DNA methylation differences between twins discordant for PTSD and migraine.

PTSD often co-occurs with other disorders and migraine being one of the less-studied comorbidities of PTSD. In the current study, we analyzed epigenome-wide DNA methylation data in MZ twins discordant for PTSD and migraine to identify common genes and pathways in PTSD and migraine. DNA methylation can be affected by both genetic and environmental factors. Therefore, the MZ twin’s disease-discordant design is a powerful approach to dissect overlap between disorders as the participants are genetically and demographically matched (Bell and Spector, 2011; Tan et al., 2015).

Using this unique study design, we first investigated six pairs of MZ twins that were all exposed to stress but were discordant for PTSD. Genome-wide analyses of DNA methylation differences across the twins identified 7 CpG loci that were significantly associated with PTSD even after stringent Bonferroni correction for multiple testing. The top CpG was in an intergenic region on chromosome 2, located near gene IL37. This gene has not been directly implicated in PTSD and is an anti-inflammatory cytokine that has been found to be increased in the amygdala and dorsolateral prefrontal cortex of children with autism spectrum disorder (Tsilioni et al., 2019). The cg26433527 site is upstream of the CSF1 gene, a cytokine that controls the production, differentiation, and function of macrophages.

Other genome-wide significant genes included WNT3, ADNP2, SLFN11 and HTT. WNT3 has been associated with stress-induced depression-like behaviors (Zhou et al., 2016) while ADNP2 has been suggested to cause changes in cellular viability under oxidative stress (Kushnir et al., 2008). SLFN11 acts as a global regulator of chromatin structure with the potential to engage the innate immune activation in response to replicative stress (Murai et al., 2020). The HTT is the Huntingtin gene involved in Huntington’s disease, a neurodegenerative disorder characterized by loss of striatal neurons. This is in line with our previous finding of an involvement of DOCK2 in PTSD, a gene which has also been implicated in the formation of amyloid plaques in the brain in Alzheimer’s disease (Mehta et al., 2017), suggesting the role of genes common to both PTSD and neurodegenerative disorders.

When assessing 45 candidate genes known to be involved in PTSD through other studies, we found that DNA methylation of these genes were significantly more likely to be associated with PTSD in the current study than expected by random chance. Of the PTSD candidate genes, we found 11 genes were also associated with PTSD in the current twin study after gene-wise Bonferroni correction (see Table 2). The candidate genes included DOCK2, SLC6A3, DICER1, DRD2, ADCYAP1, ADCYAP1R1, SKA2, OXTR, STMN1, SLC6A4, DBH, ZNF626, TRAIP, TSNAIRE1, and IMMP2L. Our team has previously found DOCK2, an amyloid-plaque associated gene in Alzheimer’s, to be associated with PTSD in an epigenome-wide study in Australian veterans (Mehta et al., 2017) and here have validated this gene in the MZ PTSD twins. Another PTSD candidate gene SLC6A3 encodes the dopamine transporter, individuals with PTSD carrying the SLC6A3 9-repeat allele were found to be at higher risk for PTSD when also having higher methylation in the SLC6A3 promoter locus (Chang et al., 2012). SLC6A4 encodes the serotonin transporter. A study reported increased methylation of SLC6A4 in bullied twins at age 10 compared to their non-bullied monozygotic twins (Ouellet-Morin et al., 2013). In genome wide studies, reduced SLC6A4 methylation levels were associated with more traumatic events and increased risk for PTSD in individuals carrying a specific SLC6A4 risk allele genotype while higher SLC6A4 methylation appeared protective against the development of PTSD (Koenen et al., 2011). The contrary results of SLC6A4 are likely driven by a gene-by-environment interaction whereby the particular genotype interacts with environment (trauma) to induce DNA methylation changes and confer susceptibility towards PTSD. Other stress-related candidate genes include SKA2 and ADCYAP1R1; a study showed associations between the methylation status, a polymorphic site in the 3′UTR of the SKA2 gene—involved in mitosis—with reduced thickness of several cortical areas and symptom severity in PTSD (Daskalakis and Yehuda, 2014). Methylation in peripheral blood samples of ADCYAP1 (PACAP protein) and its receptor ADCYAP1R1 (PAC1 protein), genes involved in regulating the cellular stress response, were associated with PTSD diagnosis and symptom severity, specifically in females (Ressler et al., 2011). There is emerging evidence of a link between PACAP and migraine, with clinical trials targeting PACAP as a likely prevention for migraine (Sundrum and Walker, 2018; Rustichelli et al., 2020).

Posttraumatic stress disorder and chronic pain conditions often occur together, and the underlying mechanisms are diverse and multifactorial (Smitherman et al., 2009; Minen et al., 2016; Gasperi et al., 2021). To investigate genes involved in both PTSD and migraine, we compared genes associated with PTSD in the current study and tested whether the same genes were also associated with migraine in an independent sample of 15 MZ twins discordant for migraine. We identified DAPK2 and TM6SF2 as two of the top overlapping genes between the two disorders. DAPK2 is a calmodulin-regulated protein kinase, it has been implicated in the intracellular degradation process essential for adaptation to metabolic stress (autophagy) (Shiloh et al., 2018). TM6SF2 is associated with cardiovascular disease and plays a role in oxidative stress (Taliento et al., 2019). These findings suggest that epigenetic changes in response to different types of stress may “mediate” stress phenotypes. Stress is a possible underlying cause of both PTSD and migraine and can impact the epigenome (Fuchikami et al., 2009; Schouten et al., 2013; Chakravarty et al., 2014; Kenworthy et al., 2014) and damage the brain in severe cases (McEwen, 2006, 2013). For instance, a single immobilization stress alters hippocampal brain-derived neurotrophic factor (BDNF) gene expression and histone acetylation at BDNF gene promoters (Fuchikami et al., 2009). BDNF is involved in the neural plasticity underlying the extinction of fear (Chhatwal et al., 2006; Heldt et al., 2007; Soliman et al., 2010) and recovery from stress. The BDNF gene has been studied in relation to anxiety disorders such as PTSD (Ressler et al., 2011; Andero and Ressler, 2012). Increased methylation of a CpG site in BDNF and peripheral blood of adults with PTSD have been reported (Smith et al., 2011). Studies have shown that BDNF is increased during migraine attacks, and in cluster headache (Fischer et al., 2012), and there is some genetic evidence suggesting a role of BDNF in migraine (Sutherland et al., 2014).

Another well-studied PTSD candidate is Catechol-O-methyltransferase (COMT), which encodes enzymes that degrade neurotransmitters such as dopamine and the serotonin transporter SERT. Increased COMT promoter methylation was associated with impaired fear inhibition in individuals with PTSD carrying the COMT met/met genotype (Meyer-Lindenberg and Weinberger, 2006). Polymorphisms in the BDNF, COMT, and SERT lead to disturbances in the normal brain pathway neurotransmitters making patients more susceptible to several neuropsychiatric disorders (Andersen and Skorpen, 2009; Fischer and Jan, 2020). Interestingly, COMT has been associated with headache medication overuse (Fischer and Jan, 2020). These studies further supporting the involvement of similar stress-related genes in the pathophysiology of both PTSD and migraine. Other proposed neurobiological mechanisms underlying PTSD-migraine comorbidity includes dysfunction of the autonomic nervous system, Hypothalamic Pituitary Adrenal (HPA) axis and brain serotonergic dysfunction (Smitherman et al., 2009; Minen et al., 2016).

When assessing biological pathways implicated in genes associated with PTSD and migraine, we found that several different pathways were involved in genes implicated in PTSD only versus those implicated in PTSD and migraine. For instance, vascular smooth muscle, axon guidance and oxytocin signaling pathways were overrepresented for genes associated with PTSD; all these are plausible and well-known pathways for PTSD. Interestingly, migraine GWAS studies have also identified an abundance of vascular genes associated with migraine (Mason and Russo, 2018; van den Maagdenberg et al., 2019; Guo et al., 2020). For genes associated with both migraine and PTSD, AMPK signaling and longevity regulating pathways were overrepresented. Adenosine monophosphate-activated protein kinase (AMPK) pathway activation might be a therapeutic target for PTSD (Wang et al., 2017) and AMPK activators might also be effective for treatment of chronic pain disorders by inhibiting signaling pathways that promote changes in the function of peripheral nociceptive neurons (Asiedu et al., 2016). There is not much known about the role of longevity pathways in PTSD or migraine.

Epigenetic regulation of gene expression is a dynamic and reversible process and therefore a good pathway target with drugs (Eising et al., 2013). In our study we used the Drug Gene Interaction Database (DGIdb) for all genes associated with PTSD to check if they were known to be affected by a known drug (clinically actionable) or are genes or encode gene products that are known or predicted to be targets for new drugs (i.e., a druggable genome). Of genes significantly associated with PTSD at 10% FDR, we found that 7.4% of these genes were known and clinically actionable targets for drugs while a further 22% of genes had a druggable genome. It has been suggested that defining the exact nature of this association and the pathophysiological mechanisms underlying the comorbidity are relevant in clinical practice as it might influence both the response to treatment and likelihood to achieve remission (Dresler et al., 2019; Seng and Seng, 2016). Current acute and prophylactic treatments are effective in less than half of the patients (Goadsby et al., 2002), indicating the need for more effective drugs. There are currently no epigenetic drugs that have gone into clinical trials for the treatment of neuropsychiatric disorders or migraine with the exception of valproate (Peedicayil and Kumar, 2012; Eising et al., 2013). Identifying factors that predispose to migraine attacks is therefore crucial to provide specific molecular targets to design novel migraine drugs. Understanding which genes are involved in co-morbid disorders such as PTSD and migraine will support the development of more advanced psychological and/or pharmacological interventions that target key pathways underlying both disorders.

There are several limitations of this study including small sample size (42 samples) which has limited power to detect small DNA methylation changes. Given the unique study design of genetically and demographically matched samples, this is a small yet homogenous sample to detect changes in disease. Another limitation of the study is the use of peripheral blood to assess DNA methylation in PTSD and migraine and while this has been a topic for discussion for many years, it is now evident that whole blood acts as a relevant and easily accessible surrogate to investigate brain-disorders. We also used an online tool by Hannon et al. to investigate the correlation of DNA methylation in blood for the significant genes with four brain regions (Ligthart et al., 2016). The modest overlap in genes and pathways between PTSD and migraine might be due to several factors including limited power and differences across the two MZ twin samples compared. Given the cross-sectional nature of the DNA methylation measurements, it is difficult to disentangle cause from effect. It is recommended that future studies should use large cohorts and at various time points to identify biological underpinnings of PTSD and migraine.

Nevertheless, we were able to validate and replicate several previously reported findings from PTSD in this study and this is the first study of its kind to use a MZ-twin design to disentangle the genes overlapping in PTSD and migraine. Using the disease discordant MZ twin design is a powerful approach in EWAS as the participants are genetically matched with similar environmental exposure and lifestyle especially in the earlier years (Tan et al., 2015). Twin studies provide a useful reference for hypotheses to be replicated and validated to understand epigenetics of complex diseases (Bell and Spector, 2011).

There are several proposed mechanisms underlying the psychiatric comorbidities and migraine relationship, but their exact etiology and biological mechanisms are not entirely known (Smitherman et al., 2009; Eising et al., 2013). Although there have been promising developments in genome-wide studies on psychopathologies in recent years, data on the biological basis of PTSD and migraine are limited. Few studies with small populations have been conducted to examine genetic and epigenetic (methylation) associations in PTSD (Uddin et al., 2010; Mehta et al., 2018). There is need for comprehensive evaluation and integrated model of care of psychiatric disorders in migraine (Dresler et al., 2019). More research is required to identify epigenetic targets that affect migraine pathophysiology and drugs that specifically act to modulate chromatin structure at migraine pathways (Eising et al., 2013; Minen et al., 2016).

To the best of our knowledge, this is the first study to examine overlapping genes and biological processes in PTSD and migraine comorbidity using the monozygotic co-twin design. These results are important and suggest that common genes and pathways might be associated with PTSD and migraine, with implications for diagnosis of comorbidities in PTSD and common treatments for these co-morbid disorders.

The datasets presented in this study can be found in online repositories. The names of the repository and accession number can be found below: Gene Expression Omnibus GSE172464.

The studies involving human participants were reviewed and approved by QIMR Berghofer Medical Research Institute and Queensland University of Technology (QUT) Human Research Ethics Committee. The patients/participants provided their written informed consent to participate in this study.

CB analyzed the data and wrote the first draft of the manuscript with DM support and guidance. NM, EN, AH, IH, and MW led studies and provided data from QIMR Berghofer Medical Research Institute. HS and RP conducted the laboratory sample analysis. LH and LG contributed to head of the Genomics Research Centre at QUT and supervised all experimental protocols. AM provided statistical advice and input. DM and DN conceived the study and supervised the project. All authors approved the manuscript and provided critical comments on the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.