Depression, also referred to as major depressive disorder (MDD), or major depression, is characterized by a range of clinical symptoms based on DSM-V criteria. In order to be diagnosed with MDD, patients must experience a depressed mood that lasts at least two weeks. Depressive manifestations include feelings of low self-esteem and a loss of interest or pleasure in activities that the patient previously enjoyed. Behaviorally, depression can be evident as reduced vigor, circadian dysfunction, eating disorders, or difficulty concentrating on important tasks. These challenges can constitute a condition that has devastating impacts directly for the patient and indirectly for others who have close relationships with the patient.

Impaired social functioning is a hallmark of depression and is correlated with the severity of depression (Hirschfeld et al., 2000; Rhebergen et al., 2010). Major depressive disorder patients tend to spend less time interacting socially and develop less depth within friendships than healthy individuals (Elmer and Stadtfeld, 2020). Additionally, social isolation and social deficits can contribute to the emergence of depressive symptomology (Jose and Lim, 2014). Consistent with this relationship, patients’ reporting of feelings of loneliness is strongly associated with behavioral symptoms of depression during social isolation in young adults (Matthews et al., 2016).

The global rise in the number of MDD diagnoses points to the need for effective therapies. Over the past 50 years, many new therapies have emerged, such as pharmacological approaches that have yielded 5 classes of pharmacological antidepressants consisting of selective serotonin reuptake inhibitors (SSRIs), serotonin and norepinephrine reuptake inhibitors (SNRIs), tricyclic antidepressants (TCAs), monoamine oxidase inhibitors (MAOIs) and atypical antidepressants drugs. These pharmacological treatments have clinical benefit (Hillhouse and Porter, 2015), but are therapeutically effective in less than 50% of patients and this have not improved significantly over time (Berton and Nestler, 2006). It is estimated that 10–30% of MDD patients completely fail to respond to these treatments which is considered treatment-resistant MDD (Rush et al., 2006; Conway et al., 2017). These broad therapeutic strategies also have a latency to provide benefit and can cause dire side effects that lead patients to withdraw from treatments (Fava, 2000; Uher et al., 2011; Cartwright et al., 2016; Wang et al., 2018; Sobieraj et al., 2019).

Notably, therapeutic effects vary greatly between the sexes. For example, SSRIs, SNRIs and other pharmacological antidepressants such as mood stabilizers, have a significantly greater effects on women compared to men (Kornstein et al., 2000; Khan et al., 2005; Berlanga and Flores-Ramos, 2006; Seney and Sibille, 2014; Charlotte et al., 2015). These sex-based differences highlight the necessity to understand the sex differences in molecular and circuit mechanisms of depression. In this mini review we summarize recent progress in epigenetic research revealing epigenetic target sites corresponding to depressive behaviors. Specifically, we focus on the sexually dimorphic epigenetic factors controlling social deficits involved in MDD.

For about 50 years, it has been consistently reported that women are diagnosed with MDD at almost 2-fold greater rate than men (Weissman and Klerman, 1977; Ford and Erlinger, 2004). Yet, men who suffer with depression have a 10-fold greater rate of suicide attempts than women (Blair-West et al., 1999). Many psychosocial theories are proffered as to why sexual dimorphisms exist in the diagnosis of depression (Jorm, 1987; Mehl-Madrona et al., 2019). However, such stark sexual differences in depression also highlight the need to understand the neurophysiological differences between the sexes.

There are many baseline differences between female and male social behaviors (Eagly and Steffen, 1986; Halpern et al., 2007; Gur et al., 2012). Hormonal and chemosensory signals are integrated in specific brain regions that control social behavior (Newman, 1999; Rolls, 2004; Amodio and Frith, 2006). Neuroimaging studies on human neurology have suggested these behavioral differences may be influenced by sex differences in neuroanatomy and structural connectome (Gur et al., 1999; Goldstein et al., 2001; Cosgrove et al., 2007; Ingalhalikar et al., 2014). A recent study using single-cell RNA sequencing of the mouse ventrolateral subdivision of the ventromedial hypothalamus, revealed some transcriptomic types exhibiting differential expressions in males and females, providing the first evidence of the existence of sex specific neurons in the mammalian CNS (Kim et al., 2019). In addition, it has also been suggested that genetic, epigenetic and environmental factors all have impacts on sex differences in social behavior (Manuck et al., 2000; Hammock and Young, 2004; Shepard et al., 2009; Aspe-Sanchez et al., 2015; Dumais and Veenema, 2016).

Epigenetics refers to changes in gene expression in the absence of alterations of the genome sequence (McCarthy et al., 2009). Primary mechanisms of epigenetic regulation include DNA methylation and demethylation, histone modifications, and non-coding RNAs known as microRNAs [for reviews, please see (Peters and Schubeler, 2005; Klose and Zhang, 2007; Li et al., 2007; Bannister and Kouzarides, 2011; Ha and Kim, 2014)]. These mechanisms are important for regulations of transcription profiles and non-coding RNA expression, whose disruptions have significant impacts on cellular functions and therefore lead to diseases (Portela and Esteller, 2010). Recent breakthroughs demonstrated that stress-induced epigenetic modifications not only have been implicated in the development of MDD (van der Doelen et al., 2014; Jawahar et al., 2015), but have also been reported to be transmitted across generations (Franklin et al., 2010; Dietz et al., 2011; Gapp et al., 2014, 2020; Short et al., 2016; Pang et al., 2017; Jawaid et al., 2018; Cunningham et al., 2021).

Studies in rodent models of depression have revealed that specific depressive behaviors relate to specific dysfunctions of neural mechanisms (Yoon et al., 2014). It is also suggested that different causes of one’s depression, such as different early life traumas, could prompt variations of neural mechanistic issues (Pacak and Palkovits, 2001; Goldstein, 2010; Smith and Pollak, 2020). In light of these many possible mechanistic variations, it is not surprising that manifestations of depression can vary greatly between individuals. If we are to achieve a complete understanding of the mechanisms that underlie MDD, it will be necessary to consider not only epigenetic mechanisms, but also how they are modulated in different ways in different social contexts. To perform those investigations, researchers have developed a variety of animal models of depression. Table 1 lists comparisons of studies on a variety of rodent models of depression. Here we focus on epigenetics studies in rodent models either exhibiting social deficits or implementing social stressors.

Studies in epigenetic modifications of human MDD have recently begun to grow, but investigations into sex differences are still rare. Recent studies have found sexually different methylation patterns in the dorsolateral PFC of late-life depression (Huls et al., 2020). Ealy life stress studies have revealed sex differences in methylation changes at the promoter of NR3C1 and the regulatory region of the FKBP5 locus (Hill et al., 2019; Wiechmann et al., 2019). Another study discovered a significant sex difference in methylation of the promoter of oxytocin gene in MDD patients (Sanwald et al., 2020). Transcriptional studies have also shown sex-specific transcriptional signatures in human MDD which might be due to epigenetic modifications. For example, Labonte et al. (2017) explored the differential expression and weighted gene coexpression network analyses between male and female MDD patients across the ventromedial and dorsolateral PFC, the anterior insula, NAc, and the ventral subiculum then compared the results with CVS mouse profiles (Labonte et al., 2017). They were able to identify sex-specific gene coexpression modules significantly associated with MDD and hub genes that carry important functional roles such as DUSP6 downregulation in females and EMX1 overexpression in males (Labonte et al., 2017).

MDD may arise in part from the differential expressions and actions of the same gene across brain regions. Ciuculete et al. (2020) discovered higher methylation levels within the hepatocyte growth factor receptor (MET) gene associated with higher depression scores and susceptibility for suicidal symptoms, along with an inverse relationship to mRNA levels of both hepatocyte growth factor (HGF) expression and MET expression in the hippocampus (Ciuculete et al., 2020). Misztak et al. (2020) discovered significant decreases in H3K9/14ac expression, BDNF protein levels and p-S421-MeCP2/MeCP2 protein ratio in both the frontal cortex (FCx) and hippocampus, along with significant increases in HDAC3 protein levels and H3K27me2 expression in both the FCx and hippocampus, as well as increases in Sin3a in the hippocampus in suicide victims (Misztak et al., 2020). This suggests that the lowered BDNF protein levels in suicide victims were most likely due to decreases in histone acetylation and increased levels of factors related to deacetylation and methylation along with MeCP2 factor which may act bidirectionally (Misztak et al., 2020).

Importantly, systematic comparisons in transcriptional profiles between human MDD and three different mouse models, including CVS, social isolation and CSDS, observed the shared transcriptional signatures between human and mouse models in two brain regions, the medial PFC and NAc (Scarpa et al., 2020). Specifically, CVS and social isolation each replicated ∼20% of the transcriptional changes in humans MDD in the PFC and NAc whereas, CSDS recapitulated ∼4% changes in gene expression. These results not only reveal significant overlaps in human MDD and mouse models, but also highlight different mouse models recapturing distinct aspects of human MDD.

Interestingly, a recent study investigating transgenerational epigenetic changes of depressive disorders associated reductions in microRNA-449 and microRNA-34 in sperm of both men and mice exposed to chronic ESI (Dickson et al., 2018). These microRNA deficits persisting in sperm promoted anxiety and social deficits in their offspring across generations (Dickson et al., 2018). This study supports the notion that epigenetic mechanism controls gene expression in a heritable way.

Most transcriptional and epigenic studies in human MDD use bulk homogenates of tissues, concealing potentially distinct changes in gene expression from individual cell types. Advanced examination using single-nucleus RNA-sequencing (snRNA-seq) on transcriptomics of the dorsolateral PFC in MDD patients, identified 26 cellular clusters of which, 60% revealed differential gene expression from controls, with the greatest dysregulation found in deep layer excitatory neurons and immature oligodendrocyte precursor cells associated with altered expression of PRNP and KAZN genes (Nagy et al., 2020). These results highlight the importance of exploring cell-type specific mechanisms in the development of MDD.

It is evident that current pharmacological treatments cause widespread/non-specific side effects in many MDD patients, and patient testimonials suggest these effects do not bring them back to a baseline of normalcy or bring stability of mind (Wang et al., 2018). Limited efficacy and tremendous side effects of existing treatments should compel the scientific community to develop better targeted and individualized interventions.

Two obstacles in exploring pathogenic mechanisms of MDD are the extremely broad clinical manifestations and the unique individual etiological triggers. Epigenetics is an exciting approach to bridge environments, genes and behaviors, as it investigates environmental influences on gene expressions that may produce individual and sex-based discrepancies in behavior (West and Greenberg, 2011).

Human epigenetic studies are rapidly expanding. However, most studies explore general epigenetic abnormalities in MDD, lacking the sexual and behavioral specificity. Various rodent models provide powerful tools to ameliorate specific aspects of different depression-like phenotypes (Catale et al., 2020; Chang et al., 2020; Lei et al., 2020; Kronman et al., 2021). For example, social deficits in MDD are easily and superiorly recapitulated in the ELS models with maternal separation and social isolation paradigms. ELS, as well as, CSDS/vCSDS models, have great potential to reveal sex differences within the epigenetic and behavioral responses to stress.

Major depressive disorder (MDD) research has only begun transitioning to explore the underlying mechanisms for evident sexual dimorphisms in MDD. With the rapid maturation and increasing commercial availability for new techniques such as snRNA-seq, future research hopefully will reveal epigenetic machinery toward accurate understanding of the sexual dimorphic and specific aspects of behavioral deficits in MDD.

This table compiles results from most epigenetics studies on rodent models and human MDD. In this table, we endeavor to show the progress seen in recent decades from early research that only included male subjects transitioning into exploring the sexually different epigenetic modifications. (A) Chronic social defeat stress and vicarious chronic social defeat stress is a model in which subjected rodents are repeatedly physically defeated by a larger, more aggressive strain or repeatedly view the physical defeat of another rodent of the same strain via a larger, more aggressive strain for psychosocial stress. (B) The chronic variable stress model utilizes a variety of stressors in order to stress subjects in order to express depressive phenotypes. Given the selection of stressors, researchers attempt to express specific depressive phenotypes such as social deficits. Articles utilizing social stressors are in blue. (C) The early life stress model has been the most popular model for MDD from our findings. The early life stress model administers one of various stressors, the most common being maternal separation before weaning is supposed to occur. Some of these stressors can be administered to the dam before the birth of her litter. This model has provided the most robust depressive phenotypes as of recent and can reliably express the social deficit phenotypes when social paradigms are used that are seen in MDD. (D) The FSL model is a line of rodents that physiologically express many neural dysfunctions as well as some behavioral dysfunctions seen in MDD like decreased serotonin synthesis, reduced BDNF expression and anxious social interactions. (E) Human MDD study utilizes the post-mortem tissues of MDD patients. Most cases do not explore the specific stress phenotypes that patients experienced because testimonial accounts are limited and thus, it is hard to study MDD associated stressful behaviors associated with mechanistic and genetic influences.

BC, JP, and YL contributed to the preparation of the manuscript. BC and YL wrote the manuscript. JP, SP-W, and YL edited the manuscript. All authors contributed to the article and approved the submitted version.

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.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.