Sex differences in epigenetic mechanisms may underlie risk and resilience for mental health disorders

Abstract

Alterations in the epigenetic programming of sex differences in the brain may underlie sexually dimorphic neurodevelopmental disorders. Sex differences have been found in DNA methyltransferases 3a, DNA methylation patterns, MeCP2, and nuclear corepressor within the developing brain. Natural variations in these epigenetic mechanisms have profound consequences on gene expression and brain function. Exogenous or endogenous perturbations during development may impact these epigenetic processes and alter the trajectory of the developing brain and confer sexually dimorphic risk and resilience for developing a neurological or mental health disorder.

Sexual differentiation of the neonatal rat brain is regulated by dynamic processes occurring at the level of DNA, resulting in sexually dimorphic gene expression. Steroid hormone receptors act partly in the developing brain by recruiting co-regulatory proteins and other transcription factors that lead to altered gene expression. Recent data indicate that epigenetic processes play an important role in sexual differentiation of the brain. Methylation of DNA is an epigenetic process that can alter gene expression without altering the original DNA sequence. Methylation marks on DNA can recruit methyl-binding proteins (MBPs), which in turn may recruit co-repressor proteins and histone deacetylase complexes (HDACs) to modify chromatin and ultimately alter gene transcription. The dynamic interaction between epigenetic factors and chromatin leads to functional plasticity within the brain that may explain the diversity of responses to endogenous and exogenous signals. Subtle variations in these epigenetic factors may also partially explain individual differences in the development of neurological and mental health disorders. We propose that sex differences in these epigenetic factors not only contribute to sexual differentiation of the brain and social behavior, but that they may confer sexually dimorphic risk and resilience for developing neurological and mental health disorders later in life (Fig. 1).

Sex Differences in Epigenetic Mechanisms

DNA methyltransferases and brain sex differences.

Methylation of DNA occurs when a methyl group attaches to a cytosine within a 5′-CpN-3′ dinucleotide site through an enzymatic reaction that is catalyzed by DNA cytosine-5-methyltransferases (DNMTs). While this bond can be rather stable, it has also been shown to be reversible.1 DNA methylation occurs mostly at CpG sites, with lower levels at CpA, CpC and CpT sites.2,3 DNMTs include DNMT1, DNMT2, DNMT3a, DNMT3b, and DNMT3L, which appear to differ in function. DNMT1 is considered important in maintaining DNA methylation patterns due to its high preference for hemimethylated DNA,4,5 whereas DNMT3a and DNMT3b are important in de novo methylation.6,7 Some data suggest that DNMT3a protein expression increases during the first three weeks of postnatal brain development and then declines, whereas DNMT3b protein expression is undetectable during this time period,8 suggesting that DNMT3a may be more important in inducing de novo methylation in postnatal brain.

Interestingly, there appears to be a sex difference in the levels of DNMT3a within the developing amygdala.9 Specifically, females express significantly more DNMT3a at postnatal day 1 (PN1) in the amygdala, but not within the preoptic area or medial basal hypothalamus. This sex difference is eliminated by the second week of life, suggesting a small window in which some genes may be differentially marked between males versus females. Sex differences in hormone exposure during early postnatal development appear partly responsible for the sex difference in DNMT3a levels, as treatment with either estradiol benzoate (EB) or dihydrotestosterone (DHT) decreases DNMT3a levels within the amygdala. This is particularly relevant as differences in hormones levels due to endogenous or exogenous sources, such as early-life adversity, may program lasting differences in brain function by modifying the expression levels of DNMTs within the developing amygdala.

As the amygdala is known to be important for modulating juvenile social play,10 aggression,11 and anxiety,12,13 programming or possibly reprogramming gene expression within the amygdala is likely to impact numerous socio-emotional functions. While the role of DNMT3a on programming sex differences in the amygdala has not been directly tested, recent studies support a role for DNMTs in regulating emotional behavior and synapse function, as well as DNA methylation patterns.14,15 Changes seen in DNMT3a, but not DNMT1, expression during the early postnatal period suggest that DNMT3a may be a key factor in producing sex differences in methylation patterns during brain development and may induce methylation patterns that alter the probability of a neurodevelopment disorder.

DNA methylation: ERα promoter methylation and brain sex differences.

As DNMTs levels are sexually dimorphic and responsive to changes in endogenous steroid hormone levels, it is not surprising that recent data indicate that DNA methylation itself is sexually dimorphic. Previous data have shown that variations in maternal care can lead to altered DNA methylation patterns.16,17 In particular, differences in maternal care alter ERα promoter methylation patterns in the developing offspring's brain and these differences correlate with changes in ERα expression.16 Interestingly, mother rats differentially groom male versus female offspring,18,19 and thereby may be modifying the DNA methylation patterns in a sex-specific manner. Indeed, it was recently found that there is a sex difference in DNA methylation of the ERα promoter region within the developing rat preoptic area20 and amygdala.21 Specifically, we found that males exhibit higher levels of methylation within the 5′ flanking region of ERα exon 1b promoter region contrasted to females during the first week of postnatal life.20 These data are consistent with previous findings that males express lower levels of ERα mRNA compared to females within the developing brain.22–24 Perhaps more importantly, treatment of newborn female rats with estradiol reprogrammed ERα promoter methylation patterns, suggesting that differences in hormone exposure within the first few days of life can have lasting consequences on gene expression levels by altering DNA methylation marks within the developing brain.20

To further clarify the contributions of maternal care to sex difference in ERα promoter methylation and expression, we modeled a component of maternal care using a paradigm called simulated maternal grooming (SMG) to mimic the somatosensory stimuli associated with maternal grooming. This paradigm controls for other factors of maternal care and focuses on the tactile stimulation of licking and grooming. We found that this subtle somatosensory stimulus altered DNA methylation patterns within the developing brain, in particular within the developing preoptic area20 and amygdala.21 That is, we find sex differences in DNA methylation patterns within the developing brain, and these differences, in part, appear to result from differences in maternal care directed at male vs. female offspring. We find that providing females with additional somatosensory stimuli associated with maternal grooming produces male-typical patterns of ERα mRNA levels and methylation of certain CpG sites in the exon 1b ERα promoter region. As the amygdala is a critical site for socio-emotional behavior, these data suggest that differences in mother-infant interactions may modify social or emotional processes by reprogramming DNA methylation patterns within the developing amygdala.

Intriguingly, the modification of ERα promoter methylation patterns by maternal stimuli is not immediate and appears to be established gradually over the first ten days of neonatal life. That is, we found that SMG-induced ERα methylation is established between P2 and P10. It is not clear if there is a critical threshold in the amount of environmental stimulus received to adjust DNA methylation patterns or if it may be time dependent; however, these data do suggest that the programming of ERα methylation is tightly regulated during development. Another possibility is that the connections relaying somatosensory stimuli from the periphery to the brain may not be fully established during the first few days of life and, therefore, those connections cannot trigger processes controlling DNA methylation. Any of these possibilities suggest a critical window during early neonatal development in which subtle variations in maternal care could reprogram DNA methylation patterns; however, it is also plausible that an accumulation of methylation marks in a number of cells are needed to detect a change in methylation pattern within that brain region. We suggest a critical window as the functional meaning of somatosensory stimuli is likely to decrease with age and therefore, are no longer able to reprogram ERα promoter methylation. However, this is not to suggest that other factors are constrained to this same time period or cannot override this window. For example, it is likely that the consequences of maternal neglect are less confined to this potential developmental window. It is also likely that other signals can later reverse these patterns later in life. A recent report indicates that some of the same CpG sites programmed during development can be altered by circulating steroid hormone levels in the adult brain.25 This suggests some DNA markings are not as stable as previously thought, and that some of these patterns need to be maintained by an endogenous or exogenous signal later in life. This observed reversibility of DNA methylation questions the stability of methylation patterns within particular regions of DNA, at least on single CpG sites located within a promoter region.

Methyl-CpG binding proteins: MeCP2 brain sex differences and function.

It is likely that binding of methyl-CpG binding proteins to methylated DNA is essential in reading methylation marks that alter gene expression. A family of methyl-CpG-binding proteins was first discovered after the characterization of the methyl-CpG-binding domain (MBD) that binds to the methylated DNA.26 Members of methyl-CpG-binding proteins include Kaiso, MBD1, MBD2, MBD3, MBD4 and Methyl-CpG-binding protein 2 (MeCP2). Upon binding to methylated DNA, methyl-binding proteins (MBPs) recruit co-repressor proteins and HDACs to modify chromatin and repress gene transcription.

MeCP2 was the first methyl-CpG binding protein to be described that binds methylated CpG sites.27 Importantly, mutations in MECP2, an X-linked gene, are believed to cause Rett syndrome, a progressive neurodevelopmental disorder.28 Rett syndrome is diagnosed mainly in females, as it is generally lethal in males. Mutations of MECP2 in mice appear to recapitulate many of the neurological symptoms occurring in Rett syndrome,29,30 and a recent study reports that re-expression of MeCP2 in these mice results in a reversal of many of these symptoms.31 These data indicate the importance of MeCP2 in the etiology of Rett syndrome. More recent reports have suggested that subtle reductions in MeCP2 expression along with aberrant MECP2 methylation patterns occur in individuals with autism,32 another sexually dimorphic neurodevelopmental disorder. In contrast to Rett syndrome, autism is known to occur at higher rates in boys. Together, this could suggest that the male brain is less able to compensate for reductions or loss of MeCP2.

In the rodent brain, females express higher levels of MeCP2 mRNA within the developing amygdala compared to males.33 It is possible that there may be a critical threshold of MeCP2 expression, and that reduction under this threshold may be detrimental to brain development. Indeed, transiently reducing MeCP2 expression within the amygdala disrupted the organization of juvenile social play behavior in males, but not females.34 It is important to note that disruption of MeCP2 in males did not alter their performance in a sociability task, suggesting that these males were less apt to engage in active social interactions that are typical of juvenile male rodents. These data not only support the concept that epigenetic factors, such as methyl-binding proteins, contribute to sexual differentiation of the brain and social behavior, but that disruption in its expression may induce sexually dimorphic behavioral disruptions later in life. These data could suggest a potential role for MeCP2 in the etiology of autism spectrum disorders, and do support the idea that the male brain is less able to adapt to reductions in MeCP2 levels.

Co-repressor complexes: NCor sex differences and function.

Methyl-CpG binding proteins bind to methylated DNA and can initiate the recruitment of nuclear co-repressors and histone deacetylase to form repressor complexes. These repressor complexes can inhibit gene expression by removing acetyl groups from histones leading to condensation of the chromatin and gene repression.35 Co-repressor complexes recruited by methyl-binding proteins include multi-protein Sin3, NuRD, CoREST, and the NCoR/SMRT repressor complexes.36 It is likely that different combinations of these co-repressor complexes allows for increased diversification of function and thereby phenotypical variations.

NCoR was one of the first corepressors to be identified through its interaction with thyroid hormone receptors;37 however, it was also reported to interact with androgen receptors,38,39 estrogen receptors40 and progestin receptors.41 NCoR has also been shown recently to interact directly, as well as indirectly, with methyl-binding proteins.42–45 In particular, NCoR is reported to interact with Kaiso, MeCP2 and other methyl-binding proteins42–44 and may interact indirectly with methyl-binding proteins through its association with the Sin3 co-repressor complex.45

While the functional role of co-repressors in the developing brain is still being elucidated, recent data indicate that there is a sex difference in NCoR levels within the developing rat amygdala. Specifically, females express higher levels of NCoR mRNA than do males. These differences may be partly due to differences in hormone exposure, as treatment with EB decreased the expression of NCoR in females to male like levels.46 This is interesting as differences in hormones levels due to endogenous or exogenous sources, such as early-life adversity, may alter NCoR expression levels within the developing amygdala and thereby change the cell or tissue's response to further cues. A transient, targeted disruption of NCoR within the developing amygdala was found to have lasting consequences on both juvenile social play and anxiety-like behaviors. While the alterations in juvenile social play occurred only in males, the transient reduction of NCoR had lasting consequences on increasing anxiety-like behavior in both juvenile males and females. These data indicate that transient disruptions in corepressors complexes within the developing brain can have sexually dimorphic consequences on juvenile social behavior and have alterations in anxiety-like behavior in both sexes. It is important to consider that sex differences in epigenetic processes may not always induce sex differences, but may be there to reduce sex differences. This is best illustrated with the NCoR data in which lowering NCoR expression levels within the developing amygdala further enhanced male-typical juvenile play behavior in males, but to atypical levels.46 Therefore, NCoR may play a role in reducing sex differences in areas of the brain where sex differences could be detrimental.

Conclusions

Our laboratory has found that the amygdala exhibits sex differences in multiple epigenetic factors during brain development, and these factors impact later behavior. Specifically, we found that neonatal females express significantly higher levels of MeCP2, NCoR and DNMT3a than males within the amygdala. We have also found that transiently decreasing the expression of genes that are recruited to methylated DNA, such as MeCP2 and NCoR, within the amygdala during the neonatal period results in altered social behavior later in life.34,46 Interestingly, transient reductions in MeCP2 within the developing amygdala decreased juvenile social play in males to female levels, while transient reductions in NCoR hyper-masculinized juvenile social play in males. These data suggest that different epigenetic factors or their combinations modify divergent phenotypes. It is intriguing that males express significantly less DNMT3a, MeCP2 and NCoR within the developing amygdala. While it is not clear why this pattern of lowered repression exist, it could suggest that the male amygdala is undergoing active differentiation induced by steroid hormone exposure during that time period. This potential increased window of differentiation of the male amygdala could put males at risk if a perturbation occurs during this time of brain organization.

Understanding epigenetic mechanisms, which underlie sexually dimorphic gene expression and function, is critical for further understanding both typical and atypical neurodevelopment. Furthermore, sexual differentiation of the brain is a powerful model to study epigenetic mechanisms, as brief exposure to steroid hormones can reorganize the brain and have lasting consequences on behavior. Perhaps more important, studying the male versus female brain provides a natural model of risk or resilience for some mental health disorders. For example, autism spectrum disorders, Rett syndrome, attention deficit hyperactivity disorder and schizophrenia, have sex differences in prevalence, time of onset and/or severity. As these disorders are believed to have an epigenetic component, sex differences in epigenetic processes may alter ones risk or resilience to develop a particular disorder. Therefore, it will be very interesting to further examine sex differences in epigenetic mechanisms and how these differences contribute to typical differentiation of the brain and social behavior, but also how they may confer sexually dimorphic risk or resilience for developing neurological and mental health disorders.

Figures and Tables

Figure 1 Sex differences in epigenetic mechanisms underlie risk and resilience for mental health disorders. Gene and environmental interactions remodel chromatin structure by recruiting co-regulatory proteins and other transcription factors resulting in either increased or decreased gene transcription rates. These interactions can lead to methylation or demethylation of DNA, as well as acetylation or deacetylation of histones. Lasting chromosomal alterations that impact gene and brain function can lead to gender specific altered risk or resilience to neurodevelopmental disorders later in life.