Genetics of resilience: Implications from genome‐wide association studies and candidate genes of the stress response system in posttraumatic stress disorder and depression

Resilience is the ability to cope with critical situations through the use of personal and socially mediated resources. Since a lack of resilience increases the risk of developing stress‐related psychiatric disorders such as posttraumatic stress disorder (PTSD) and major depressive disorder (MDD), a better understanding of the biological background is of great value to provide better prevention and treatment options. Resilience is undeniably influenced by genetic factors, but very little is known about the exact underlying mechanisms. A recently published genome‐wide association study (GWAS) on resilience has identified three new susceptibility loci, DCLK2, KLHL36, and SLC15A5. Further interesting results can be found in association analyses of gene variants of the stress response system, which is closely related to resilience, and PTSD and MDD. Several promising genes, such as the COMT (catechol‐O‐methyltransferase) gene, the serotonin transporter gene (SLC6A4), and neuropeptide Y (NPY) suggest gene × environment interaction between genetic variants, childhood adversity, and the occurrence of PTSD and MDD, indicating an impact of these genes on resilience. GWAS on PTSD and MDD provide another approach to identifying new disease‐associated loci and, although the functional significance for disease development for most of these risk genes is still unknown, they are potential candidates due to the overlap of stress‐related psychiatric disorders and resilience. In the future, it will be important for genetic studies to focus more on resilience than on pathological phenotypes, to develop reasonable concepts for measuring resilience, and to establish international cooperations to generate sufficiently large samples.

relationship problems, serious health problems or workplace and financial stressors. It means 'bouncing back' from difficult experiences" (APA, 2018). Since all individuals are at some point exposed to stressful life events or traumas, understanding of how some of us can cope with such experiences and others not, is crucial to maintaining or regaining mental health in society. In this context, a better understanding of the genetic mechanisms underlying resilience is important to improve treatment and prevention strategies and to implement personalized medicine.
In the past 20 years there have been enormous developments in the discovery of genetic factors associated with complex psychiatric diseases such as schizophrenia (Giegling et al., 2017) and Alzheimer's disease (Kunkle et al., 2019), but also with personality traits (Sanchez-Roige, Gray, MacKillop, Chen, & Palmer, 2018) and intelligence (Savage et al., 2018). However, there are very few studies that have investigated the genetic impact on resilience. An important reason for this is the large number of resilience-related indicators, so that the measurement of resilience is neither clearly operationalized (Rodriguez-Llanes, Vos, & Guha-Sapir, 2013) nor a gold standard has been defined (Windle, Bennett, & Noyes, 2011). Moreover, the focus has so far been less on health-promoting factors than on diseaseassociated and deficit-oriented aspects. One way to counter this problem, at least in part, and still being able to draw conclusions about the underlying genetic mechanisms of resilience, is to consider studies in which vulnerable phenotypes have been investigated. Why this is a reasonable approach becomes apparent when one considers resilience and vulnerability as the poles of a continuum (Haddadi & Besharat, 2010;Kim-Cohen & Turkewitz, 2012). In addition, there is an overlap of indicators between vulnerable phenotypes, especially posttraumatic stress disorder (PTSD) and major depressive disorder (MDD), and psychological resilience, which is reflected by the fact that after a trauma or an adverse life event, a lack of resilience can contribute to the development of PTSD or MDD (Ahmadpanah et al., 2017;Mattson, James, & Engdahl, 2018). Thus, genetic case-control studies comparing individuals who have developed a mental disorder after stress exposure with those who have not developed mental problems provide a way to identify genetic factors associated with resilience, since these studies compare resilient and nonresilient phenotypes.
Moreover, there is evidence for mechanisms that predict vulnerability to stress and susceptibility to PTSD and MDD in the face of stress and trauma (Southwick & Charney, 2012;Wu et al., 2013).
Based on these preliminary considerations, this review is structured as follows: The first section focuses on the heritability of resilience. As there are few studies on this issue, it is necessary to use other resources to gain a deeper insight into the genetic background of resilience. Therefore, the second section gives an overview of studies that have investigated associations of vulnerable phenotypes with genetic variants of the neuroendocrine stress response system.
It is assumed that the stress response system plays a key role for resilience (Feder, Nestler, & Charney, 2009), so that the focus in this section is on the serotonergic, noradrenergic, and dopaminergic systems as well as the hypothalamic-pituitary-adrenal axis (HPA axis), neuropeptide Y (NPY), and brain-derived neurotrophic factor (BDNF).
In particular, results will be presented that have revealed a gene × environment interaction in the development of mental disorders and thus suggest a connection with resilience. In the third section, results of genome-wide association studies (GWASs) on resilience, PTSD and MDD will be presented, as they offer a relatively new approach to the identification of hypothesis-free phenotype-associated genetic variants and thus an opportunity to gain direct insights into the genetics of resilience. Finally, the discussion section contains a summary of the most important results, a conclusion on the current state of knowledge and an outlook for the future.

| METHODS
A MEDLINE (PubMed) research was conducted for this review. First of all, studies were considered in which genetics and heritability of resilience were addressed. Since the literature in this field is limited, we have included studies that have investigated the association of genetic variants of the stress response system with psychiatric disorders and have therefore considered PTSD and MDD as outcome variables in terms of a lack of resilience. It should be noted that the focus was on studies from the last 10 years and that not all studies were included, in particular those with very small sample sizes and those from which no relationship to resilience could be derived. Finally, a systematic search for GWAS on resilience, PTSD and MDD was conducted to use this new and promising approach, which has led to a significant development in genetic research in recent years.

| HERITABILITY OF RESILIENCE
Most of the knowledge about the heritability of resilience derives mainly from twin studies. In a study of more than 1,000 pairs of twins in childhood, genetic and environmental factors affecting resilience were investigated, with 46% of the variance of cognitive and 70% of the variance of behavioral resilience being explained by genetic effects (Kim-Cohen, Moffitt, Caspi, & Taylor, 2004). A study carried out by Wolf et al. (2018) on 3,318 male twin pairs from the Vietnam Era Twin Registry, which included analyses of genetic and environmental influences on the severity of PTSD symptoms as measured by the PTSD Checklist (Weathers et al., 2017) and an assessment of resilience, measured with the Connor-Davidson Resilience Scale-10 (Connor & Davidson, 2003), revealed a heritability of resilience of 25% and PTSD of 49%. Resilience and PTSD were negatively correlated at r = −.59, and 59% of this correlation was attributable to a single genetic factor, whereas the remainder was due to a single nonshared environmental factor (Wolf et al., 2018). Another study investigating the genetic contribution to resilience in a genome-wide approach with 8,734 participants from the GS:SHFS study (Generation Scotland:Scottish Family Health Study) confirmed the heritability of resilience, but the estimated phenotypic variance of 8% attributable to genetic factors was significantly lower than in the aforementioned studies (Navrady et al., 2018). This study also investigated the influence of genetic factors on different coping styles (task-oriented, emotion-oriented, avoidance-oriented coping), which are closely related to resilience (Iacoviello & Charney, 2014). Interestingly, a large genetic correlation between emotion-oriented coping and resilience was found, which indicates a common genetic background of these traits (Navrady et al., 2018). Amstadter, Maes, Sheerin, Myers, and Kendler (2016) found in patients with MDD and generalized anxiety disorder (GAD) that 42% of MDD heritability and 60% of GAD heritability are due to genetic factors influencing resilience, suggesting shared heritability of these diseases and resilience. These findings support an impact of genetics on resilience, whereby the studies differ in the extent of heritability. There is also evidence that the investigation of PTSD and MDD may allow conclusions to be drawn about the genetic background of resilience, as there is at least a partial overlap between resilience and these psychiatric disorders.

| CANDIDATE GENES OF THE NEUROENDOCRINE STRESS RESPONSE SYSTEM
Several neurotransmitter systems contribute to resilient responses to stress and are implicated in the development of PTSD and MDD.
Genetic variants of the noradrenergic, dopaminergic, and serotonergic systems, as well as genes encoding for neurotrophic factors or genes related to the HPA axis have been most extensively studied (Sheerin, Lind, Bountress, Nugent, & Amstadter, 2017;Wu et al., 2013). The following sections provide an overview of the main results of association studies on genetic variants of the stress response system with PTSD and MDD in the context of resilience.

| Serotonergic system
The serotonergic system is connected to the function of two key stress response systems: the HPA axis (Leonard, 2005) and the locus coeruleus (LC)-norepinephrine (NE) system (Goddard et al., 2010).
A promising gene from this neurotransmitter system is the SLC6A4 gene (solute carrier family 6 member 4), encoding the serotonin transporter (SERT). Within the promotor region of SLC6A4, there is a polymorphism (serotonin transporter-linked polymorphic region; 5-HTTLPR) with short (S) and long (L) repeats, with the S allele leading to decreased SERT expression compared to the L allele (Lesch et al., 1996). A meta-analysis showed that the S allele is associated with increased stress sensitivity (M. Zhao et al., 2017) and furthermore, S allele carriers are more likely to develop MDD, which has already been proven in several studies (López-León et al., 2008). Overall, there seems to be an association between the promoter polymorphism of the SLC6A4 gene, depression and environmental interactions, as carriers of the low-active S allele had a markedly elevated risk of developing depression under stress exposure, which was demonstrated in a meta-analysis of 54 studies (Karg, Burmeister, Shedden, & Sen, 2011).
This study also found evidence for the association of the S allele with stress sensitivity and depression in maltreated children. A connection of the S allele was also shown in an increased risk for PTSD in patients with childhood adversity and adult traumatic events (Xie et al., 2009).
A dose-dependent relationship between SLC6A4 variants and emotional resilience was additionally demonstrated in a study on 423 psychology students, with lower resilience scores found in S allele carriers (Stein, Campbell-Sills, & Gelernter, 2009). However, a number of meta-analyses investigating the SLC6A4 × environment interaction revealed mixed results, and the effect, if present, is modest and unlikely to be generalized (Culverhouse et al., 2018;Karg et al., 2011;Munafò, Durrant, Lewis, & Flint, 2009;Risch et al., 2009;van der Auwera et al., 2018). Taken together, S allele carriers are more likely to develop stress-related psychiatric disorders, such as PTSD and MDD, which may be due to lower resilience in S allele carriers.
In addition to the SLC6A4 gene, serotonin receptors and enzymes of the serotonin metabolism have been investigated. The mitochondrial enzyme monoamine oxidase A (MAOA) is responsible for the degradation of serotonin as well as epinephrine and NE and a metaanalysis found an association between a variable number of tandem repeats polymorphism (uVNTR) in the MAOA promoter region and MDD, but limited to Asians (Fan et al., 2010). In addition, epigenetic modifications by DNA methylation of the MAOA gene have been associated with PTSD (increased methylation status) and panic disorder (decreased methylation status) as well as the occurrence of positive and negative life events (Domschke et al., 2012;Ziegler et al., 2017). Another enzyme in the serotonin metabolism is tryptophan hydroxylase 2 (TPH2), the rate-limiting enzyme in the synthesizing pathway for brain serotonin (Invernizzi, 2007). A higher risk for MDD has been reported for two independent SNPs of TPH2 (Gao et al., 2012), with the T allele of rs4570625 being associated with smaller volumes of bilateral amygdala and hippocampus, a typical finding in emotion-related psychiatric disorders (Inoue et al., 2010). Genetic variants of the genes HTR1A (5-hydroxytryptamine receptor 1A; Kishi et al., 2013) and HTR2A (X. Zhao et al., 2014) appear to be associated with depression and of HTR2C with depressive symptoms in women and elevated cortisol levels induced by acute mental stress, implying a direct link between HTR2C and HPA axis activation (Brummett et al., 2012;Brummett, Babyak, Kuhn, Siegler, & Williams, 2014).

| Dopaminergic and noradrenergic systems
Dopamine emerges in several, relatively confined groups of neurons projecting to various brain areas including the prefrontal cortex, nucleus accumbens (NAcc), hippocampus, and amygdala. Differences in striatal dopamine transporter (DAT) density in PTSD patients compared to healthy, traumatized individuals, suggest an influence of the dopaminergic system on vulnerable phenotypes and resilience (Hoexter et al., 2012). In a meta-analysis by Li et al. (2016), two genetic variants in genes of the dopaminergic system with increased susceptibility to PTSD were detected, namely the VNTR polymorphism in the promoter region of the human DAT gene (SCL6A3) and a polymorphism (rs1800497) in the dopamine receptor D2 gene (DRD2). DRD2 has also been shown to regulate synaptic modification in response to stress (Perreault, Hasbi, O'Dowd, & George, 2014;Sim et al., 2013). In addition, both genes, SCL6A3 and DRD2, are associated with MDD, whereby the association of DRD2 has been demonstrated in a large GWAS with 130,664 cases and 330,470 controls (López-León et al., 2008;Wray & Sullivan, 2017). Also an influence on resilience could have variants of the DRD4 gene (dopamine receptor D4), where carriers of seven or more copies of a VNTR polymorphism in the third exon had a seemingly protective effect and thus an increase of resilience if they suffered adversity during childhood.
Conversely, this effect was not observed when no childhood trauma occurred.
The catecholamine NE is released from its main production sitethe LC in the pons-upon stress-induced activation of the noradrenergic system and transported to its various projection sites, including amygdala, hippocampus, hypothalamus, and prefrontal cortex (Bandelow et al., 2017). β-adrenergic receptors as well as α-adrenergic receptors and the NE transporter are considered to be involved or affected in various psychiatric disorders and resilience (Borodovitsyna, Flamini, & Chandler, 2017;Krystal & Neumeister, 2009). So far, however, there are no conclusive results on genetic variants of the NE system related to resilience.
One potential candidate affecting both the dopaminergic and noradrenergic systems is the enzyme catechol-O-methyltransferase (COMT). The SNP rs4680 (Val 158 Met), which affects the activity of encoded COMT, is probably the most replicated disease-relevant polymorphism of this system. The Met allele is associated with a decreased COMT enzyme activity and thus higher NE and dopamine levels (Chen et al., 2004). Homozygous carriers of the Met allele show lower emotional resilience against negative mood states in humans (Smolka et al., 2005) and exaggerated stress reactivity in mice (Papaleo et al., 2008). The Met allele was found to be associated with decreased inhibition-related activation in the hippocampus, which in turn was associated with PTSD and depression symptoms in patients with childhood trauma (van Rooij et al., 2016). An accumulation of the Met allele was also found in individuals who developed PTSD after being exposed to urban violence (Valente et al., 2011). A study on genocide survivors showed, that Val allele carriers exhibited an elevated risk for PTSD, depending on the number of lifetime traumatic events, while Met/Met homozygotes were at high risk for PTSD regardless of the traumatic load (Kolassa, Kolassa, Ertl, Papassotiropoulos, & de Quervain, 2010). The presence of the COMT Met allele also leads to a stronger cortisol stress response in children (Armbruster et al., 2012). These results imply an interaction of the COMT variants with stress and thus suggest an influence on resilience.
However, it should not go unmentioned that the study data on COMT and PTSD are inconsistent and that a meta-analysis of five studies did not show any significant effect (Li et al., 2016).

| Hypothalamic-pituitary-adrenal axis
The HPA axis is a major neuroendocrine system that affects various organ systems and plays a fundamental role in mediating stress With regard to the HPA axis, several genes and their potential impact on vulnerable phenotypes have been studied, but there are few studies that have investigated the link between genes of this hormone system and resilience. However, a connection between the HPA system and resilience processes is supported, for example, by the observation of an altered HPA reactivity in later life depending on the presence of adverse life events in early life (Romeo, 2015). For the corticotropin-releasing hormone receptor CRHR1, several polymorphisms are associated with a reduced risk of depressive symptoms after being exposed to early life stress (for review see Laryea, Arnett, & Muglia, 2012). And another study on gene × environment interactions in children revealed an association between CRHR1 haplotypes with resilience depending on their maltreatment status (Cicchetti & Rogosch, 2012). A similar gene × environment effect has been found in two studies that investigated maltreatment during childhood, with CRHR1 variants appearing to moderate the risk of depressive symptoms in adulthood Polanczyk et al., 2009). Such gene × environment interactions are a strong indication of a genetic impact on resilience, as variations in resilient behavior after adversity or stress may be caused by a different genetic composition. In addition, significant associations of genetic variants in the CRHR1 gene have been detected in PTSD patients (Boscarino, Erlich, Hoffman, & Zhang, 2012;White et al., 2013;Wolf et al., 2013).
Studies focusing on the relationship between variants of the glucocorticoid receptor gene (NR3C1) and resilience have not yet been conducted. However, epigenetic modifications by DNA methylation related to trauma exposure have been shown, although the results of these studies were inconsistent (Watkeys, Kremerskothen, Quidé, Fullerton, & Green, 2018). There is also evidence that NR3C1 polymorphisms are associated with PTSD symptoms and depression (Hauer et al., 2011;Lian et al., 2014;Peng, Yan, Wen, Lai, & Shi, 2018).
Another gene of the HPA axis is the FK506-binding protein 5 gene (FKBP5), which interacts with the glucocorticoid receptor binding heat-shock protein 90 (HSP90). Elevated FKBP5 levels lead to a decreased negative feedback regulation of the HPA axis and glucocorticoid receptor resistance, which is probably responsible for a dysregulated stress response . In several association studies, genetic variations in the FKBP5 gene were associated with PTSD occurrence and severity, depending on the presence of childhood trauma Buchmann et al., 2014;Comasco et al., 2015;Watkins et al., 2016). These results were substantiated in a recently published study showing a gene × environment interaction between FKBP5 polymorphisms and childhood abuse to predict the risk for PTSD (Tamman et al., 2019). Such findings can help to identify patients with an increased risk of mental disorders and to implement personalized medicine in the future. Moreover, common allelic variants in the FKBP5 gene are associated with an increased risk of developing affective disorders like anxiety, depression, and PTSD (Criado-Marrero et al., 2018).
A higher risk for depression susceptibility after maltreatment in childhood was also found for haplotypes of the mineralocorticoid receptor (NR3C2), whereby a relationship between NR3C2 variants and current depressive symptoms and lifelong MDD diagnosis has been demonstrated in two samples (Vinkers et al., 2015). Since the HPA axis is the most important physiological stress response system (Silverman & Deuster, 2014), genetic variations in this system are likely to influence resilience and contribute to psychiatric disorders in vulnerable phenotypes.

| Neuropeptide Y
Neuropeptide Y is a biologically active peptide and acts as a neuromodulator in the brain. In several brain regions (hippocampus, hypothalamus, LC, and amygdala) corticotropin-releasing hormone mediated anxiogenic effects are counteracted by NPY, which is necessary for the compensation of stress reaction and homeostasis (Thorsell et al., 2000).
Polymorphisms within the NPY locus affect NPY expression and it has been reported that NPY haplotypes that mediate lower NPY expression are associated with diminished resilience to stress Z. Zhou et al., 2008). In addition, several polymorphisms in the NPY gene have been described in connection with anxietyrelated disorders, early childhood adversity, and early life stress. Various studies on gene × environment interactions of the NPY promotor variant rs16147 in traumatized subjects revealed promising results.
One study showed that the C allele of this polymorphism is associated with anxiety and depressive symptoms depending on childhood adversity (Sommer et al., 2010), while T allele homozygotes were at higher risk of developing a GAD after high hurricane exposure (Amstadter et al., 2010). A gene × environment interaction study of the same SNP for a divergent stress-induced response of cortisol and adrenocorticotropic hormone levels depending on adversity exposure of the participants during childhood was also demonstrated (Witt et al., 2011). And in two cohorts of traumatized participants, T allele carriers of rs16147 adopted better traumatic stress than C homozygotes and developed a higher positive future focus, which is a relevant aspect of resilience (Gan, Chen, Han, Yu, & Wang, 2019). Based on these studies, an influence of this promoter polymorphism in interaction with environmental factors on resilience is likely, which could possibly be mediated by differential expression of the protein.

| Neuronal and synaptic plasticity
According to the neurotrophic hypothesis of MDD, the disease may be associated with impaired structural plasticity and cellular resilience, with a key role of BDNF, a neurotrophin highly expressed in the hippocampus and involved in the regulation of synaptic plasticity, neurogenesis, neuronal survival, and differentiation (Ferrari & Villa, 2017). It has been repeatedly demonstrated that BDNF is a contributing factor to a variety of psychiatric disorders, and it is known that BDNF levels are affected by stress in PTSD and MDD patients (Casey et al., 2009;Duman, 2009;Duman & Monteggia, 2006).
Association studies on the functional BDNF Val 66 Met polymorphism (rs6265) revealed inconsistent results regarding the influence on stress response and resilience. Although there were studies that found no significant association between the polymorphism Val 66 Met and PTSD diagnosis (Rakofsky, Ressler, & Dunlop, 2012), further studies, including a meta-analysis, discovered an increased risk for PTSD and the severity of PTSD symptoms in Met allele carriers (Bruenig et al., 2016;Dai et al., 2017). An interesting approach, which explored possible causes of this connection was followed in a study by Felmingham et al. (2018), which showed that Met allele carriers presented more severe PTSD symptoms in addition to poorer fear extinction learning, which is crucial for PTSD treatment. An overlap with resilience is possible, because disturbed fear extinction can lead to the development or maintenance of mental illnesses and a lack of resilience (Shansky, 2015). Other genes that are relevant for neuronal and synaptic plasticity and that are also linked to nonresilient phenotypes are CREB1 and CACNA1C. CREB1 (cyclic adenosine monophosphate response element-binding protein 1) encodes a downstream effector of BNDF that increases the expression of BDNF target genes (Juhasz et al., 2011). Polymorphisms in CREB1 have been reported to modulate the risk of different major psychiatric disorders including MDD (Xiao et al., 2017), while no association has been found with PTSD (Serretti et al., 2013). The CACNA1C gene (calcium voltage-gated channel alpha 1C subunit) is involved in the regulation of calcium-mediated membrane depolarization and modulates intracellular signaling, gene transcription, and synaptic plasticity (Bhat et al., 2012). CACNA1C has been proposed as a susceptibility gene for various psychiatric disorders (Cross-Disorder Group of the Psychiatric Genomics Consortium, 2013). The effect of CACNA1C polymorphisms on MDD susceptibility was confirmed by a meta-analysis that extracted genotypic data from available GWAS and performed a candidate gene study in an independent sample (Rao et al., 2016).

| GENOME-WIDE ASSOCIATION STUDIES
Genome-wide association studies represent the methodological answer to the observation of the highly polygenic component of psychiatric traits, including MDD and PTSD, and of course resilience (Peterson et al., 2017). In addition, GWAS enable the detection of genetic variants associated with specific phenotypes that could not be discovered with conventional hypothesis-based strategies. This provides a completely new starting point for a better understanding of pathophysiological mechanisms and factors that influence disease development, as well as for the investigation of complex traits or constructs such as resilience.
To date there is only one GWAS on resilience, which was published recently by Stein et al. (2019). Since PTSD in particular, but also MDD, can occur frequently due to trauma, stress a result of a lack of resilience, these phenotypes are useful to identify new potential loci that can then be further investigated to assess possible effects on resilience.
For this reason, the next section summarizes the first GWAS on resilience on the one hand and the most important GWAS results on PTSD and MDD on the other. Tables 1 and 2 additionally provide an overview of all GWA studies on PTSD and MDD carried out so far.
In the only GWAS on resilience to date, US soldiers of European descent were studied, and resilience was measured using a five-item self-report questionnaire and by measuring the outcome using the There are significantly more GWAS on posttraumatic stress disorder, although most of them do not have well-powered samples (Table 1). The first GWAS by Logue et al. (2013), involving military veterans, identified the retinoid-related orphan receptor alpha (RORA) as best association with PTSD. Another study detected the Tolloid-like 1 gene (Xie et al., 2013) and LINC01090 as a risk factor for PTSD . A study on 3,394 US Marines reported genomewide association for PRTFDC1 (phosphoribosyl transferase domain containing 1 gene) as a potential predictor of combat stress vulnerability and resilience (rs6482463; OR = 1.47, p = 2.04 × 10 −9 ; Nievergelt et al., 2015). In a study (New Soldier Study) combining 3,167 PTSD patients and 4,607 trauma-exposed controls, a genome-wide significant locus was found in ANKRD55 on chromosome 5 (rs159572; OR = 1.62; p = 2.34 × 10 −8 ), which persisted after adjustment for cumulative trauma exposure (OR = 1.64; p = 1.18 × 10 −8 ) in the African-American samples (Stein et al., 2016). ANKRD55 has previously been associated with diabetes mellitus type 2 (Harder et al., 2013) and various autoimmune diseases, such as rheumatoid arthritis (Viatte et al., 2012) and multiple sclerosis (Alloza et al., 2012), suggesting a genetic overlap of these diseases, as PTSD is also associated with autoimmune diseases and diabetes. Restricted to the European ancestry subgroup, a genomewide significant association near zinc finger protein 626 gene (ZNF626) on chromosome 19 (rs11085374; OR = 0.77; p = 4.59 × 10 −8 ) was detected. The Psychiatric Genomics Consortium-PTSD continues to encourage the further discovery of genes involved in the pathology and susceptibility to PTSD (Banerjee, Morrison, & Ressler, 2017). The largest GWAS on PTSD so far (including 20,730 samples: 15,548 controls, 5,182 cases) revealed no genome-wide significant association with the disease in a multiethnic PGC-PTSD cohort, but suggested a robust genetic overlap with bipolar disorder and schizophrenia (Duncan et al., 2018). A previously found overlap of PTSD with MDD could not be confirmed, but this as well as the failure to detect genome-wide significant associations was attributed to the relatively low power of the PTSD and MDD studies. Nevertheless, the top pathway was the neurotrophic factor-mediated Trk receptor signaling pathway, which includes BDNF and which also showed overlaps to resilience (see section "Neuronal and synaptic plasticity").
Although sample sizes were much higher than in PTSD, the identification of MDD-associated loci that reached genome-wide significance in GWAS was challenging, in particular because of the high genetic heterogeneity and high prevalence of MDD (Table 2) (Bicaudal C homolog 1) and downregulation of BDNF/TrkB signaling were observed in both hippocampus and cortex after application of chronic unpredictable stress in a mouse model of depression . In addition, treatment with antidepressants reduced the expression of BICC1, and the knockdown of this gene in the hippocampus also prevented anhedonia, a key feature of depression in the same model in rats (Ota, Andres, Lewis, Stockmeier, & Duman, 2015).
Moreover, there is evidence that BICC1 associated polymorphisms affect the capability of the BICC1 promoter to respond to PKA (protein kinase A) signaling in amygdala neurons (Davidson et al., 2016).  Among the replicated genome-wide associations, NEGR1 (neuronal growth regulator 1) shows a role in synaptic plasticity in MDDrelevant brain regions such as the cortex, hypothalamus, and hippocampus (Hashimoto, Maekawa, & Miyata, 2009;Sanz, Ferraro, & Fournier, 2015;Schäfer, Bräuer, Savaskan, Rathjen, & Brümmendorf, 2005). DCC (Netrin 1 receptor) also looks promising as it is one of the most relevant genes contributing to the association between the NETRIN signaling pathway and MDD in different samples . No genome-wide significant hit in the meta-analysis of the combined sample. One genome-wide hit in the AA sample located in the KLHL1 gene on chromosome 13 (rs139558732, p = 3.33e−08). Genetic overlap with schizophrenia and bipolar disorder Wilker et al. (2018) 924 Cases 371 Cases African Association tests with lifetime PTSD risk revealed suggestive significance for one SNP on chromosome 2, two SNPs on chromosome 3, two SNPs on chromosome 5, one SNP on chromosome 6, and one SNP on chromosome 13. Replication of one SNP (rs3852144) on chromosome 5 Abbreviations: AA, African American; AI, American Indian.
T A B L E 2 Summary of genome-wide association studies (GWASs) that investigated the genetics of major depressive disorder (MDD) resilience was repeatedly demonstrated in twin studies, although the proportion of the genetic impact between these studies varied markedly (Connor & Davidson, 2003;Kim-Cohen et al., 2004), as well as in the so far only GWAS on resilience in which SNP-based heritability was estimated at 16% (Stein et al., 2019).
In this first GWAS on resilience, some interesting genome-wide significant hits were obtained, although the sample, especially the outcome-based analysis, was small. An interesting candidate among the significant hits was DCLK2, a member of the doublecortin family of kinases that promote survival and regeneration of neurons (Nawabi In the discovery meta-analysis, rs9540720 in the PCDH9 gene was associated with MDD (p = 1.69e−08) and the result was confirmed in the meta-analysis including two additional data sets (p = 1.20e−08) Hall et al. (2018)   occurrence of PTSD and MDD can be used as outcome variables, which can at least indirectly give a hint to possible genetic resilience factors. This is underlined by the fact that when a polymorphism is associated with a stress-related mental illness, one allele of this polymorphism is associated with a higher and the other with a lower disease risk. In other words, one allele is associated with the resilient phenotype and the other allele with the nonresilient phenotype.
In connection with resilience, the neuroendocrine stress response system in particular is attributed a major role (Feder et al., 2009). A promising candidate is the SLC6A4 gene, which encodes the serotonin transporter (SERT). Several studies have shown an association between the S allele variant of this gene and PTSD and MDD in relation to experienced stress and adversity (Karg et al., 2011;Xie et al., 2009) as well as a lower level of resilience in S allele carriers . Another promising candidate of the catecholaminergic system is COMT, whose variants also show a gene × environment interaction effect, with Met allele carriers who experienced trauma or adversity in childhood exhibiting a greater risk for the development of PTSD and depression and thus appearing to be less resilient (Valente et al., 2011;van Rooij et al., 2016). The HPA axis also appears to have an influence on resilience, particularly for the CRHR1 and FKBP5 genes, with interesting results suggesting a link between genetic variants and maltreatment during childhood and the development of PTSD and depression Polanczyk et al., 2009;Tamman et al., 2019). Although the HPA axis is such an important part of the stress response system, there are relatively few studies that address resilience. Table 3 gives an overview of the most promising genes implicated in resilience.
Many other susceptibility genes have been discovered for PTSD and MDD (Tables 1 and 2), but the exact function of the respective genes and the corresponding proteins is often unclear. Whether these genes also have an impact on resilience must be clarified in future research projects. However, GWAS offer a promising approach to discover new common genetic variants, as a hypothesis-driven methodology is not necessary. This development is also facilitated by the fact that GWAS have been increasingly implemented since 2008, as the costs for genome sequencing began to decrease dramatically and became more feasible in large samples. This made it possible to identify previously unknown interacting genetic factors by investigating large cohorts of PTSD and MDD patients. Furthermore, a growing number of genome sequencing projects on large samples from the general population are expected to provide new and notable findings about the genetics of psychiatric disorders in the near future (e.g., "Genomic Aggregation Project" in Sweden (Bergen & Sullivan, 2017) and "All of Us" in the United States (https://allofus.nih.gov/).
Although there are several studies that suggest a genetic influence on resilience processes, investigations on large samples, possibly also in a longitudinal approach, are necessary in order to shed light on the underlying genetic processes of resilience. Collaboration in consortia, such as the Psychiatric Genomics Consortium (PGC), has helped to expand the sample sizes for psychiatric disorders research. This might also be an approach for gathering sufficiently large samples to study resilience in the future. Within this context, it will also be necessary to operationalize resilience uniformly and not only to investigate disease-associated phenotypes, which is almost exclusively the case so far. This could also involve focusing on resilience-related features such as coping styles, cognitive assessment, emotionality, and cognitive self-regulation, which can be helpful to address the problem from the nondisease-related side.
Further research into resilience is of great importance, also to better understand the healthy functioning of the human mind and to identify factors that could prevent the occurrence of mental disorders. This is also necessary in order to develop precise psychotherapeutic interventions and pharmacological treatments that selectively target resilience associated signaling pathways, in order to specifically promote resilience, avert consequential damage, and strengthen prevention.