Role of Epigenetics in Mental Disorders

Abstract

The purpose of this paper was to selectively review the literature on the role of epigenetics in mental illnesses. Aberrant epigenetic regulation has been clearly implicated in the aetiology of some human illnesses. In recent years a growing body of evidence has highlighted the possibility that epigenetics may also play a key role in the origins and expression of mental disorders. Epigenetic phenomena may help explain some of the complexity of mental illnesses and provide a basis for discovering novel pharmacological targets to treat these disorders.

Keywords

differential allele expression epigenetics gene regulation

The aetiology of major mental disorders involves a complex interplay of nature and nurture. Simplistic nature versus nurture arguments are no longer tenable [1], especially when nurture is limited to the post-uterine psychosocial environment. Developments in molecular genetics have shown that non-genetic variables impact on gene expression (epigenetics), which is the subject of this review paper. However, genes may also influence environmental measures, including stressful life events, parenting behaviour, family environment, social support, peer interactions and marital quality to a moderate extent [2]. Furthermore, environmental effects may be dependent upon (interact with) genetic variability, with the best-documented example being how polymorphisms of the serotonin transporter influence vulnerability to depression in the presence of adverse life events [3].

This review, which focuses on epigenetics, highlights how reversible and heritable modifications of gene expression may occur, without a change in the DNA sequence itself. Such modifications may involve methylation, acetylation and phosphorylation of the DNA and/or other post-translational modifications that alter chromatin structure and gene activity. These changes then alter the production of mRNA and thence protein production.

Gene expression is a multi-step process that is controlled by many different mechanisms, some of which are well understood (such as transcription factors), but we are only beginning to comprehend the function of epigenetics in gene regulation. For instance, every human carries two copies (known as alleles) of every gene, one of which is inherited from the father and one from the mother. In many cases these alleles are both expressed, so that they contribute equally to the mRNA transcript and protein pool (biallelic expression). However, exceptions are known to exist. For example, X-chromosome inactivation in female cells involves the transcriptional silencing of one of the two X-chromosomes to achieve dosage compensation equal to the normal male chromosomal composition of only one X-chromosome. Another example is genomic imprinting where one allele is completely silenced in a parent-of-origin-dependent manner resulting in monoallelic gene expression (for a catalogue of imprinted genes see: http://igc.otago.ac.nz/home.html). It is epigenetic regulation that is responsible for these observed differences of expression levels between alleles.

Several lines of evidence confirm that such allelic variation in gene expression exists in humans and mice [4–7]. It is unknown to what degree allelic expression differences are biologically relevant and this is likely to vary between genes and in a tissue- and temporal-specific manner. Nonetheless, it has become apparent that differential allele expression is indeed a common phenomenon and may be an important factor contributing to the frequently observed human phenotypic diversity, including differences in disease susceptibility. In recent years, progress has been made in understanding some of the key epigenetic mechanisms, namely the function of post-translational modifications on histones and the role of DNA methylation, and their effect on gene activity.

This review will first explain the two best studied epigenetic mechanisms, second, discuss environmental influences on these epigenetic mechanisms and finally present support for a role of epigenetics in behaviour. Examples relating to psychiatric conditions will be included at each stage of the discussion, illustrating how environment and genetics can both contribute to the expression of behaviours, such as mental illness. In doing so this review will highlight the importance of integrating an epigenetic perspective into psychiatric research.

Epigenetic mechanisms

Epigenetics and the histone code of chromatin

In eukaryotic cells DNA is almost entirely associated with proteins and organized in the form of chromatin. It has become apparent that chromatin plays a significant role in regulating gene expression. Chromatin is composed mainly of DNA, histones and so-called non-histone chromatin proteins that facilitate the packing of the DNA double helix into higher order structures so that it can be stored in the nucleus (Figure 1) [8]. Two forms of chromatin can be distinguished, namely heterochromatin and euchromatin. Euchromatin is more loosely packed, which allows the access of transcription factors and other components to promoter regions, thereby enabling genes to be transcribed. Conversely, heterochromatin is more compactly arranged, leading to transcriptional inhibition due to the inaccessibility of promoter elements.

Figure 1.

Chromatin condensation. This model shows the hierarchical order of chromatin packaging within a chromosome. Reprinted with permission from Macmillan Publishers, from Felsenfeld G, Groudine M. Controlling the double helix. Nature 2003; 421:448–453.

The basic structural unit of the chromatin fibre is the nucleosome, which consists of DNA wrapped around a histone octamer consisting of two copies each of the so-called core histones H2A, H2B, H3, and H4 (Figure 2a) [9], [10]. Core histones are proteins that tightly bind with the negatively charged phosphate backbone of the DNA, thereby stabilizing the chromatin. Each histone also contains a flexible domain called the histone tail, which remains outside of the nucleosome [11]. These exposed tails are targets for reversible post-translational modifications, such as methylation (on arginine and lysine residues), acetylation (on lysine residues), phosphorylation (on serine and threonine residues) and ubiquitination (on lysine residues), which alter chromatin structure and consequently influence gene activity [12], [13]. The large number of possible modification sites results in the potential for numerous combinations of modifications. By generating distinct modification patterns epigenetic information can thus be stored within the histone tails (Figure 2b). Therefore, it has been hypothesized that these covalent modifications constitute a ‘histone code’ and function as epigenetic signals from the DNA to the cellular machinery for various processes including gene regulation [14], chromosome condensation (mitosis) [15], [16] and DNA repair [17].

Figure 2.

Schematic diagram of the structure of histones arranged in nucleosomes. (a) Each nucleosome consists of two copies each of H2A, H2B, H3 and H4 (core histones) forming an octamer around which the DNA is wrapped. The amino-terminal tails of core histones are exposed on the outside of the nucleosomes. (b) Detailed representation of the amino-terminal tails of the core histones with potential post-translational modification sites for histone deacetylases (HDACs) and histone acetyltransferases (HATs). A, acetyl; C, carboxyl terminus; E, glutamic acid; M, methyl; N, amino terminus; P, phosphate; S, serine; Ub, ubiquitin. Reprinted with permission from Macmillan Publishers, from Marks P, Rifkind RA, Richon VM, Breslow R, Miller T, Kelly WK. Histone deacetylases and cancer: causes and therapies. Nat Rev Cancer 2001; 1:194–202.

Epigenetics and DNA methylation of CpG sites

DNA methylation also plays a key role in gene regulation. It is a common method of gene silencing that can be inherited without changing the nucleotide sequence of the involved gene and thus is part of the epigenetic code. In adult human somatic tissues DNA methylation usually occurs at sites where the bases C (cytosine) and G (guanine) are located adjacent on the DNA backbone (called CpG sites). DNA methylation involves the addition of a methyl group to the cytosine residues (Figure 3) [18], [19]. The level of DNA methylation is closely associated with the level of gene silencing, where hypermethylation results in gene silencing and hypomethylation in gene transcriptional activity. CpG methylation is not evenly spread throughout the mammalian genome. In eukaryotes >80% of all CpGs are methylated while notably high concentrations of unmethylated CpG sites are found at the promoters of actively transcribed genes [20], [21]. In humans CpG methylation is catalysed by three enzymes, namely, DNA methyltransferases 1, 3A and 3B (DNMT1, DNMT3A, DNMT3B) [21–24]. Knock-out mice deficient for any of these enzymes either do not survive embryogenesis or die within days of birth, highlighting the important role of DNA methyltransferases in development [24–26].

Figure 3.

Methylation of cytosine. DNA methylation involves the addition of a methyl group to the 5-carbon position of cytosine residues.

Epigenetics: environment meets genome

Environmental stimuli can alter epigenetic modifications and influence gene expression

Changes in epigenetic modifications can be induced via environmental stimuli such as nutrition, maternal care/behaviour, hormones and drugs. For example, nutrition can influence gene expression via DNA methylation. The metabolic production of methyl groups used in all biological methylation reactions is extremely dependent on diet-derived methyl donors (e.g. methionine) and critical cofactors (e.g. folic acid, vitamin B12) [27]. In pre-implantation embryos the genome undergoes widespread demethylation resulting in a loss of the CpG methylation pattern [28]. The requirement to re-establish the genomic methylation pattern following implantation and the necessity to maintain this pattern during the many cycles of rapid cell proliferation in early development make the fetal and early postnatal stages of life more critically reliant on the availability of appropriate levels of dietary methyl donors and cofactors, more so than later in life [29], [30]. Therefore, overall availability of critical amino acids and micronutrients may alter DNA methylation and thus influence gene expression.

Nutrition influences epigenetics

There are several lines of evidence supporting the notion of nutritional effects on CpG methylation, and therefore on the epigenetic status of genes and even whole chromosomal regions. First, tissue culture experiments with mouse embryos clearly show that aberrant expression of the imprinted gene H19 is dependent on the type of culture media used [31]. Second, imprinting disorders such as Prader-Willi syndrome occur more frequently in children who have been conceived with the help of assisted reproductive technologies, presumably because embryos grown outside the womb in artificial media are in a suboptimal nutritional environment [32–34]. Ke et al. demonstrated that uteroplacental insufficiency also affects epigenetic determinants of chromatin structure, such as histone H3 hyperacetylation and DNA hypomethylation, in the brains of neonatal and juvenile rats [35]. This leads to persistent changes in cerebral mRNA levels of various genes including DNMT1 [35]. MacLennan et al. showed that genome-wide DNA hypomethylation was present in the liver of postnatal intrauterine growth restriction (IUGR) rats [36]. Similarly, Pham et al. found that IUGR leads to hypomethylation of the p53 gene in postnatal rat kidneys, along with increased renal apoptosis [37].

These results from animal research are consistent with human health studies. Health records collected during the Dutch Hunger Winter of 1944–1945, which occurred at a precisely confined time and place, allowed the analysis of long-term consequences of prenatal famine. It was found that severe prenatal nutritional deficiency is specifically associated with schizophrenia [38], [39], schizophrenia spectrum personality disorders [40], [41] and congenital anomalies of the central nervous system [42]. The increased risk for schizophrenia following prenatal exposure to famine was replicated by the analysis of health data collected in the wake of another famine in China in 1959–1961 [43]. Other consequences of famine exposure during gestation include impaired glucose tolerance, obesity, coronary heart disease, hypertension and affective disorders [44–46]. Nutritional deprivation during childhood has been linked to changes in reproductive function, earlier menopause and an increased risk for breast cancer [45]. This suggests that early nutritional influences on gene expression may have a significant impact on adult human health, which has resulted in the hypothesis of fetal origins of adult diseases [47].

Maternal behaviour influences epigenetics

Behaviour is another environmental factor that can trigger changes in epigenetic modification patterns. For example, an animal study has elucidated the specific effects of maternal care on the epigenetic regulation of a steroid receptor gene. Szyf et al. showed that alterations of the DNA methylation and histone acetylation status of the glucocorticoid receptor gene occur in the hippocampus of offspring whose mothers displayed higher than average pup licking/grooming and arched-back nursing (all of which are naturally occurring traits in maternal behaviour) [48]. These observed epigenetic changes persisted into adulthood but were reversible if pups were cross-fostered. Interestingly, it was also observed that rats that had a lot of attention from their mothers as pups handled stressful situations, as shown by their hypothalamic–pituitary–adrenal (HPA) response to stress, much better later in life compared to rats that were reared by mothers who displayed lower than average maternal care [48]. These findings indicate an association between early maternal nurture, the epigenetic state of the glucocorticoid receptor and HPA stress response in offspring [48], [49]. Therefore, epigenetic gene expression can indeed be determined by maternal behavioural programming and although the established epigenetic mark can be stable, it is also often dynamic and reversible.

Interestingly, many imprinted genes influence maternal behaviour such as paternally-expressed-gene 1 (Peg1). The promoter region of Peg1 displays a strict parent-of-origin-specific methylation pattern whereby the maternal allele is completely methylated at CpG sites and thus silenced [50]. Only the unmethylated paternal allele is expressed in individual tissues of embryos, neonates and adults [51]. Peg1 was found to be highly expressed in the developing and adult mouse brain [51], [52]. Of particular interest is the finding that Peg1 is involved in postnatal behaviour of adult mice, because female mice deficient for Peg1 display abnormal maternal behaviour, including impaired placentophagia (a distinctive mammalian behaviour in which the mother consumes the placenta after giving birth) [52]. Additionally, these mice lack an appropriate immediate response to their offspring, including failure to build nests for their offspring as well as delayed retrieval and feeding of the pups. In view of the significance of olfactory cues in maternal behaviour and the fact that Peg1 is expressed in the olfactory bulbs, Peg1-deficient female mice were tested for olfactory function and found not to be impaired [52]. This made Peg1 the first imprinted gene to be shown to have a role in the control of adult behaviour.

Epigenetics and human diseases

Direct link between epigenetics and certain human disorders

Aberrant epigenetic mechanisms have been directly implicated in certain human diseases. For instance mutations in genes encoding methylating enzymes involved in the DNA methylation process can cause severe human illnesses. One such illness is a rare autosomal recessive developmental disorder called immune deficiency, centromeric instability and facial anomalies (ICF) syndrome [53]. Patients with this disorder were found to carry mutations in the C-terminal domain of the DNMT3B gene, which results in hypomethylation and decondensation of the heterochromatin structure in pericentromeric repeats, as well as alterations in gene expression [54]. Another human disorder caused by aberrant DNA methylation is Rett syndrome. This X-linked dominant disorder mainly affects the female and is marked by progressive neurological deterioration, gait abnormalities, loss of purposeful hand usage, seizures, cognitive impairment and loss of speech [55]. The vast majority of sufferers carry mutations in the methyl-CpG-binding protein 2 (MECP2) gene. The MECP2 protein is able to bind selectively to methylated CpG sites. This protein forms a link between DNA methylation and chromatin structure as it recruits histone deacetylases and other chromatin remodelling proteins that modify core histones, resulting in heterochromatin formation and thus gene silencing [55].

Epigenetics and psychiatric illnesses

While epigenetics has been directly implicated in certain conditions, it has been hypothesized to be also involved in psychiatric illness. It is a plausible association based on how epigenetic variation parallels observed variation in psychiatric phenotypes. Such phenotypic variation includes monozygotic (MZ) discordance, parent-of-origin effects (POE), changes in disease course and drug response. There is growing evidence that epigenetics could be responsible for differences between individuals, as well as responsible for variation within an individual.

Epigenetic modifications have been recently seen to vary throughout an individual's life. This observation has been attributed to a concept called ‘metastable epialleles’. The resulting variability over time may explain how an individual's disease course and drug response can differ through their life. It has been noted that some mammalian alleles exhibit an unusual stochastic variability in their expression in the absence of genetic heterogeneity. This variability is due to the lability of the allele's epigenetic state and results in phenotypic mosaicism between cells and among individuals [56]. Rakyan et al. recommended that the term ‘metastable epialleles’ be used to signify such alleles, where ‘metastable’ alludes to the labile nature of the epigenetic state and ‘epialleles’ to the alleles’ ability to retain their epigenetic state across generations [56]. All metastable epialleles characterized to date contain a transposon (a defined DNA sequence that can move around the genome), emphasizing the ability of these virus-like elements to cause epigenetic instability. The resulting epigenetic variability has been proposed to underlie a considerable amount of phenotypic diversity in mammals [57], [58].

Waterland and Jirtle showed that metastable alleles are epigenetically susceptible to early nutritional influences of four common nutritional supplements, namely folic acid, vitamin B12, choline and betaine through direct effects on the CpG methylation status of the affected locus [58]. These new methylation patterns were retained into adulthood. Furthermore, it was found that any changes to the epigenetic state of transposons can also affect the regulation of neighbouring genes [58], [59]. These findings support the notion that the epigenetic state of transposable elements can be metastable and is indeed susceptible to early methyl donor nutrition [58]. It is important to remember that transposable elements make up >40% of the human genome and that they are silenced mainly by CpG methylation [59]. It is this nature of metastable epialleles that is proposed to cause much of the phenotypic diversity within and among individuals.

Epigenetics and variation between individuals

MZ twin discordance is a widely observed phenomenon of multi-factorial diseases. Environmental factors have been the customary explanation for how two genetically identical individuals can display phenotypic differences. Fraga et al. examined the global- and locus-specific levels of DNA methylation and histone acetylation, the two key epigenetic modifications, in 40 MZ twin pairs [60]. They found a clear relationship between the age of twins and the amount of epigenetic differences between them [60]. Although MZ twins are epigenetically indistinguishable at birth and during the early years of life, with the progression of age twins generally exhibited remarkable variation in the overall content and distribution of epigenetic modifications. This in turn resulted in different gene expression profiles between twins [60]. Therefore, the molecular basis for MZ discordance may lie in the significant epigenetic differences accrued over time in the regulation of disease-relevant genes. For instance, in one twin epigenetic dysregulation of a susceptibility gene may reach a critical threshold and clinical symptoms may become apparent, while no effect occurs for the other twin. The need to reach an essential threshold and the discovered increase of epigenetic changes with age may also explain the later age of onset observed in complex disorders compared to simple Mendelian disorders such as cystic fibrosis, which generally arise in childhood.

POE have been found in several complex disorders and it is one of the classic epigenetic features of differential regulation of gene expression. POE refers to an association of a phenotype with either maternal or paternal alleles. For example, in one linkage study some loci on chromosome 18q21 showed POE in some bipolar (BP) families because there was a clear overrepresentation of paternal but not maternal alleles carrying a particular linkage marker (D18541) [61]. Clinical observations have noted that the risk for BP I disorder is generally higher for maternal relatives as well as for children of affected women [62]. Several other studies also suggest that POE contribute to the transmission of BP I disorder [63–65].

Gender-specific differences have been observed in the prevalence, severity and treatment response of mood disorders and schizophrenia [66–69]. Such gender-specific effects may be triggered by sex hormones that mediate changes in the epigenetic regulation of critical genes and as a result cause differential epigenetic modifications in one gender but not the other [70–72]. Several studies looking at a serine to glycine variant in exon 1 (Ser9Gly) of the dopamine D3 receptor gene (DRD3) reported a significant excess of both forms of homozygotes in schizophrenia patients compared to controls [73–75], indicating that heterozygosity for this variant may be a protective feature. Interestingly, a gender-specific effect was observed with a significant excess of homozygotes found in male but not in female subjects [73], [74], [76].

Epigenetics and variation within an individual

Continual fluctuations in the course and/or severity of a disease are also characteristic features of complex psychiatric illnesses that have puzzled clinicians and researchers. Alternating remission and relapse are commonly observed in major depressive disorder (MDD), schizophrenia and BP. These variations in disease course are consistent with the dynamic shifting that occurs in the epigenetic regulation of gene expression. Epigenetic modifications are reversible and epigenetic regulation as a whole is a dynamic process that can change over time. It is believed that epigenetic modifications represent a dynamic molecular mechanism that allows an organism to respond and adapt to its current environment via changes in gene activity.

Drug treatments influence epigenetics

Additional data that clearly implicate epigenetics in psychiatric disorders come from various drug treatments. For example, administering L-methionine to chronic schizophrenia patients results in the exacerbation of psychotic symptoms, but no adverse effects are seen in healthy controls [77], [78]. This compound is involved in the production of S-adenosyl methionine, which provides the methyl groups for DNA methylation. In contrast, valproate causes a reduction in DNA methylation levels and is currently used as an anti-manic and mood-stabilizing drug [79], [80].

There is somewhat conflicting evidence for the importance of methylation state in the aetiology and treatment of depression. For instance, folic acid and methionine may have some role in the treatment of major depression [81]. Both compounds are required for the methylation process, suggesting that hypermethylation may alleviate depressive symptoms. This theory may be supported by findings that excessive alcohol consumption leads to folate deficiency and thus presumably to hypomethylation and has been associated with depression and dementia in older adults [82], [83]. In contrast, animal models of alcohol abuse during gestation have demonstrated hypermethylation of specific genes including brain-derived neurotrophic factor (BDNF), which is required for the survival and function of neurons, resulting in reduced levels of mRNA [81], [84]. BDNF has been implicated in the regulation of stress responses as well as in the aetiology of mood disorders in humans [85], [86].

Further evidence for the relevance of epigenetic phenomena in psychiatry comes from the finding that non-selective monoamine oxidase inhibitors potently inhibit the demethylation of histone H3 lysine 4 (H3K4) [87]. This demethylation is carried out by an enzyme called BHC110/LSD1, resulting in the repression of gene expression [88]. BHC110/LSD1 has a marked homology to monoamine oxidases, which are drug targets in the treatment of depression. The most potent inhibitor of demethylation by BHC110/LSD1 was tranylcypromine [87]. Downstream effects of the use of this inhibitor were increased global levels in H3K4 methylation and the transcriptional expression of BHC110/LSD1 target genes [87]. These findings implicate histone demethylation as a potential novel mechanism of action for antidepressive drugs.

Does epigenetics impact on the regulation of other psychiatric candidate genes?

The 5-hydroxytryptamine (serotonin) receptor 2A gene (HTR2A) was found to be paternally imprinted and thus expressed only from the maternally inherited allele in human fibroblasts [89]. This POE was consistent with CpG island methylation in the promoter region of HTR2A on the paternal allele [89]. The mouse gene (Htr2a) was observed to be also paternally imprinted in several different organs including the cerebrum [90]. However, several studies on HTR2A expression in human brain tissue have led to conflicting findings. One study found evidence for polymorphic imprinting, with four out of 18 individuals showing monoallelic gene expression, whereas the remaining 14 individuals displayed biallelic expression of HTR2A [91]. Another study found an increased expression of one allele among schizophrenia patients [92]. However, a study by Bray et al. could not replicate either finding and found no inter-individual variability in relative mRNA levels of the two alleles in the post-mortem brains of 23 individuals [93]. Therefore, it is currently uncertain whether HTR2A is epigenetically regulated in the human brain and whether altered methylation patterns in this gene have any clinical relevance.

A gene that shows overall altered mRNA transcript levels with disease status is Reelin (RELN). RELN has long been a psychiatric candidate gene for schizophrenia, BP, major depression and autism [94]. Reelin functions in cell–cell interaction, synaptogenesis and neuronal migration, and the expression of this gene is tightly correlated with the methylation status of its promoter region [95–97]. Significant reduction in RELN mRNA levels has been reported in post-mortem brain of schizophrenia and BP sufferers [95–97]. But it is unclear whether the observed reduction is causal to the development of these illnesses or only a downstream consequence. In addition, it remains to be determined whether this reduction in expression may reflect changes in the epigenetic status of the RELN promoter region.

Discussion

In the last decade it has become apparent that genetic variations that affect the structure and function of proteins may not be the primary component of human phenotypic diversity and causes of disease susceptibility. Instead, factors that influence allelic expression have been shown to play a key role in modulating diversity, disease origins and progression. Differential expression of alleles appears to be a common phenomenon, with 20–50% of all human genes predicted to be affected [4]. Epigenetic modifications, like DNA methylation and histone acetylation, control the amount of mRNA made. Thus, aberrant epigenetic regulation may have significant biological consequences and contribute to diseases if incorrect quantities of a structurally and functionally normal protein are made at the wrong place or time. Indeed abnormal epigenetic regulation of gene expression has already been directly implicated in the aetiology of a number of diseases, and this review has summarized some of the evidence that indicates a potentially important role for epigenetics in the origins and characteristic traits of psychiatric illnesses.

Many of these complex phenomena, for instance MZ twin discordance and POE, and the observed phenotypic diversity of psychiatric conditions have baffled clinicians and scientists alike because they are not readily explained by classic genetic features such as DNA mutations in susceptibility genes. To date, psychiatric genetic research has mainly concentrated on linking specific sequence variation (mutations) in proposed psychiatric candidate genes to particular mental illnesses or aspects of these illnesses, such as drug response. However, it has proven difficult to find robust genetic associations. Mental illnesses are complex disorders caused by multiple genes and environmental influences. Epigenetics can be regarded as the interface at which environmental stimuli and internal factors (such as hormones) interact with gene expression. As shown by studies investigating the origins of MZ twin discordance, varying levels of gene expression of disease-relevant genes due to differences in epigenetic regulation between twins, seem to be a significant cause of disparate phenotypic outcomes [60]. In other words, even though MZ twins are genetically identical, and as such may carry the same DNA mutation that is proposed to increase the risk for a mental illness or adverse drug response, it is the level at which the involved alleles are expressed that may determine whether a particular phenotype is displayed. Epigenetic mechanisms have been clearly shown to play a crucial role in regulating allelic expression levels in the mammalian genome. The dynamic nature of epigenetic marks, as well as their reversibility and heritability, provides a cohesive explanation for many features of complex traits that are observed in mental illness.

The concept of epigenetics in the origin of diseases has been well supported and consequently begun to be incorporated in the detection and treatment of certain medical conditions. For example, aberrant DNA methylation patterns are being explored as biomarkers for cancer detection, assessment of prognosis and prediction of response to therapy. Inhibitors of histone deacetylases and DNA methyltransferases are being used to treat certain haematological cancers for which hypermethylation has been found [98], [99]. DNA methylation inhibitors were shown to re-activate aberrantly silenced tumour suppressor genes leading to growth arrest and apoptosis [99]. Demethylation inhibitors, in contrast, are proposed to be potential anti-metastatic agents by methylating and silencing metastatic genes [100]. These recent developments make epigenetic therapy an exciting new field in the fight against cancer [101].

By analogy, pharmacological therapies in mental illness are likely to benefit from an epigenetic perspective as well, as highlighted by recent findings that the effects of non-selective monoamine oxidase inhibitors, such as tranylcypromine, suggest that histone demethylation may be a novel mechanism for antidepressant drugs [87], [88], [102], [103].

The ultimate significance of these findings to human mental health remains to be determined. But it is now clear that environmental stimuli may influence disease susceptibility by changing epigenetic modifications of relevant genes. Furthermore, interventions designed to produce epigenetic modifications may offer new approaches to the treatments for mental disorders.

Footnotes

Acknowledgements

S. Stuffrein-Roberts is funded by a University of Otago Postgraduate PhD Scholarship and a Lotteries Health Research Scholarship. M. Kennedy's and P. Joyce's research is supported by the Health Research Council of New Zealand. The authors would like to thank Katrina Light for critical reading of the manuscript.

References

1. Ridley

. Nature via nurture: genes, experience, and what makes us human. HarperCollins Publishers, New York 2003.

2. Kendler

Baker

. Genetic influences on measures of the environment: a systematic review. Psychol Med 2007; 37: 615–626.

3. Caspi

Sugden

Moffitt

. Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene. Science 2003; 301: 386–389.

4. Lo

Wang

. Allelic variation in gene expression is common in the human genome. Genome Res 2003; 13: 1855–1862.

5. Bray

Buckland

Owen

O'Donovan

. Cis-acting variation in the expression of a high proportion of genes in human brain. Hum Genet 2003; 113: 149–153.

6. Yan

Yuan

Velculescu

Vogelstein

Kinzler

. Allelic variation in human gene expression. Science 2002; 297: 1143.

7. Cowles

Hirschhorn

Altshuler

Lander

. Detection of regulatory variation in mouse genes. Nat Genet 2002; 32: 432–437.

8. Felsenfeld

Groudine

. Controlling the double helix. Nature 2003; 421: 448–453.

9. Marks

Rifkind

Richon

Breslow

Miller

Kelly

. Histone deacetylases and cancer: causes and therapies. Nat Rev Cancer 2001; 1: 194–202.

10.

10. Kornberg

Lorch

. Twenty-five years of the nucleosome, fundamental particle of the eukaryote chromosome. Cell 1999; 98: 285–294.

11.

11. Luger

Rechsteiner

Flaus

Waye

Richmond

. Characterization of nucleosome core particles containing histone proteins made in bacteria. J Mol Biol 1997; 272: 301–311.

12.

12. Spencer

Davie

. Role of covalent modifications of histones in regulating gene expression. Gene 1999; 240: 1–12.

13.

13. Strahl

Allis

. The language of covalent histone modifications. Nature 2000; 403: 41–45.

14.

14. Berger

. Histone modifications in transcriptional regulation. Curr Opin Genet Dev 2002; 12: 142–148.

15.

15. Hans

Dimitrov

. Histone H3 phosphorylation and cell division. Oncogene 2001; 20: 3021–3027.

16.

16. Nowak

Corces

. Phosphorylation of histone H3: a balancing act between chromosome condensation and transcriptional activation. Trends Genet 2004; 20: 214–220.

17.

17. Loizou

Murr

Finkbeiner

Sawan

Wang

Herceg

. Epigenetic information in chromatin: the code of entry for DNA repair. Cell Cycle 2006; 5: 696–701.

18.

18. Chiang

Gordon

Tal

. S-Adenosylmethionine and methylation. FASEB J 1996; 10: 471–480.

19.

19. Jaenisch

Bird

. Epigenetic regulation of gene expression: how the genome integrates intrinsic and environmental signals. Nat Genet 2003; 33(Suppl)245–254.

20.

20. Bird

. CpG-rich islands and the function of DNA methylation. Nature 1986; 321: 209–213.

21.

21. Bird

Wolffe

. Methylation-induced repression: belts, braces, and chromatin. Cell 1999; 99: 451–454.

22.

22. Gruenbaum

Cedar

Razin

. Substrate and sequence specificity of a eukaryotic DNA methylase. Nature 1982; 295: 620–622.

23.

23. Bestor

. Cloning of a mammalian DNA methyltransferase. Gene 1988; 74: 9–12.

24.

24. Okano

Bell

Haber

. DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 1999; 99: 247–257.

25.

25. Li

Bestor

Jaenisch

. Targeted mutation of the DNA methyltransferase gene results in embryonic lethality. Cell 1992; 69: 915–926.

26.

26. Lei

Okano

. De novo DNA cytosine methyltransferase activities in mouse embryonic stem cells. Development 1996; 122: 3195–3205.

27.

27. Van den Veyver

. Genetic effects of methylation diets. Annu Rev Nutr 2002; 22: 255–282.

28.

28. Reik

Dean

Walter

. Epigenetic reprogramming in mammalian development. Science 2001; 293: 1089–1093.

29.

29. Waterland

Garza

. Potential mechanisms of metabolic imprinting that lead to chronic disease. Am J Clin Nutr 1999; 69: 179–197.

30.

30. Waterland

Lin

Smith

Jirtle

. Post-weaning diet affects genomic imprinting at the insulin-like growth factor 2 (Igf2) locus. Hum Mol Genet 2006; 15: 705–716.

31.

31. Doherty

Mann

Tremblay

Bartolomei

Schultz

. Differential effects of culture on imprinted H19 expression in the preimplantation mouse embryo. Biol Reprod 2000; 62: 1526–1535.

32.

32. Orstavik

Eiklid

van der Hagen

. Another case of imprinting defect in a girl with Angelman syndrome who was conceived by intracytoplasmic semen injection. Am J Hum Genet 2003; 72: 218–219.

33.

33. DeBaun

Niemitz

Feinberg

. Association of in vitro fertilization with Beckwith-Wiedemann syndrome and epigenetic alterations of LIT1 and H19. Am J Hum Genet 2003; 72: 156–160.

34.

34. Thompson

Konfortova

Gregory

Reik

Dean

Feil

. Environmental effects on genomic imprinting in mammals. Toxicol Lett 2001; 120: 143–150.

35.

35. Ke

Lei

James

. Uteroplacental insufficiency affects epigenetic determinants of chromatin structure in brains of neonatal and juvenile IUGR rats. Physiol Genomics 2006; 25: 16–28.

36.

36. MacLennan

James

Melnyk

. Uteroplacental insufficiency alters DNA methylation, one-carbon metabolism, and histone acetylation in IUGR rats. Physiol Genomics 2004; 18: 43–50.

37.

37. Pham

MacLennan

Chiu

Laksana

Hsu

Lane

. Uteroplacental insufficiency increases apoptosis and alters p53 gene methylation in the full-term IUGR rat kidney. Am J Physiol Regul Integr Comp Physiol 2003; 285: R962–970.

38.

38. Susser

Lin

. Schizophrenia after prenatal exposure to the Dutch Hunger Winter of 1944–1945. Arch Gen Psychiatry 1992; 49: 983–988.

39.

39. Susser

Neugebauer

Hoek

. Schizophrenia after prenatal famine. Further evidence. Arch Gen Psychiatry 1996; 53: 25–31.

40.

40. Hoek

Brown

Susser

. The Dutch famine and schizophrenia spectrum disorders. Soc Psychiatry Psychiatr Epidemiol 1998; 33: 373–379.

41.

41. Hoek

Susser

Buck

Lumey

Lin

Gorman

. Schizoid personality disorder after prenatal exposure to famine. Am J Psychiatry 1996; 153: 1637–1639.

42.

42. Susser

Hoek

Brown

. Neurodevelopmental disorders after prenatal famine: the story of the Dutch Famine Study. Am J Epidemiol 1998; 147: 213–216.

43.

43. St Clair

Wang

. Rates of adult schizophrenia following prenatal exposure to the Chinese famine of 1959–1961. JAMA 2005; 294: 557–562.

44.

44. Lackland

Egan

Ferguson

. Low birth weight as a risk factor for hypertension. J Clin Hypertens (Greenwich) 2003; 5: 133–136.

45.

45. Kyle

Pichard

. The Dutch Famine of 1944–1945: a pathophysiological model of long-term consequences of wasting disease. Curr Opin Clin Nutr Metab Care 2006; 9: 388–394.

46.

46. Barker

. The fetal origins of adult hypertension. J Hypertens Suppl 1992; 10: S39–44.

47.

47. Barker

. Fetal growth and adult disease. Br J Obstet Gynaecol 1992; 99: 275–276.

48.

48. Szyf

Weaver

Champagne

Diorio

Meaney

. Maternal programming of steroid receptor expression and phenotype through DNA methylation in the rat. Front Neuroendocrinol 2005; 26: 139–162.

49.

49. Weaver

Cervoni

Champagne

. Epigenetic programming by maternal behavior. Nat Neurosci 2004; 7: 847–854.

50.

50. Lefebvre

Viville

Barton

Ishino

Surani

. Genomic structure and parent-of-origin-specific methylation of Peg1. Hum Mol Genet 1997; 6: 1907–1915.

51.

51. Kaneko-Ishino

Kuroiwa

Miyoshi

. Peg1/Mest imprinted gene on chromosome 6 identified by cDNA subtraction hybridization. Nat Genet 1995; 11: 52–59.

52.

52. Lefebvre

Viville

Barton

Ishino

Keverne

Surani

. Abnormal maternal behaviour and growth retardation associated with loss of the imprinted gene Mest. Nat Genet 1998; 20: 163–169.

53.

53. Xu

Bestor

Bourc'his

. Chromosome instability and immunodeficiency syndrome caused by mutations in a DNA methyltransferase gene. Nature 1999; 402: 187–191.

54.

54. Ehrlich

Buchanan

Tsien

. DNA methyltransferase 3B mutations linked to the ICF syndrome cause dysregulation of lymphogenesis genes. Hum Mol Genet 2001; 10: 2917–2931.

55.

55. Bienvenu

Chelly

. Molecular genetics of Rett syndrome: when DNA methylation goes unrecognized. Nat Rev Genet 2006; 7: 415–426.

56.

56. Rakyan

Blewitt

Druker

Preis

Whitelaw

. Metastable epialleles in mammals. Trends Genet 2002; 18: 348–351.

57.

57. Whitelaw

Martin

. Retrotransposons as epigenetic mediators of phenotypic variation in mammals. Nat Genet 2001; 27: 361–365.

58.

58. Waterland

Jirtle

. Transposable elements: targets for early nutritional effects on epigenetic gene regulation. Mol Cell Biol 2003; 23: 5293–5300.

59.

59. Yoder

Walsh

Bestor

. Cytosine methylation and the ecology of intragenomic parasites. Trends Genet 1997; 13: 335–340.

60.

60. Fraga

Ballestar

Paz

. Epigenetic differences arise during the lifetime of monozygotic twins. Proc Natl Acad Sci USA 2005; 102: 10 604–10 609.

61.

61. Stine

Koskela

. Evidence for linkage of bipolar disorder to chromosome 18 with a parent-of-origin effect. Am J Hum Genet 1995; 57: 1384–1394.

62.

62. McMahon

Stine

Meyers

Simpson

DePaulo

. Patterns of maternal transmission in bipolar affective disorder. Am J Hum Genet 1995; 56: 1277–1286.

63.

63. Grigoroiu-Serbanescu

Nothen

. Clinical evidence for genomic imprinting in bipolar I disorder. Acta Psychiatr Scand 1995; 92: 365–370.

64.

64. Kato

Winokur

Coryell

Keller

Endicott

Rice

. Parent-of-origin effect in transmission of bipolar disorder. Am J Med Genet 1996; 67: 546–550.

65.

65. Gershon

Badner

Detera-Wadleigh

Ferraro

Berrettini

. Maternal inheritance and chromosome 18 allele sharing in unilineal bipolar illness pedigrees. Am J Med Genet 1996; 67: 202–207.

66.

66. Angst

Gamma

Gastpar

Lepine

Mendlewicz

Tylee

. Gender differences in depression. Epidemiological findings from the European DEPRES I and II studies. Eur Arch Psychiatry Clin Neurosci 2002; 252: 201–209.

67.

67. Hafner

Riecher-Rossler

Maurer

Fatkenheuer

Loffler

. First onset and early symptomatology of schizophrenia. A chapter of epidemiological and neurobiological research into age and sex differences. Eur Arch Psychiatry Clin Neurosci 1992; 242: 109–118.

68.

68. Osran

Reist

Chen

Lifrak

Chicz-DeMet

Parker

. Adrenal androgens and cortisol in major depression. Am J Psychiatry 1993; 150: 806–809.

69.

69. Ozcan

Banoglu

. Gonadal hormones in schizophrenia and mood disorders. Eur Arch Psychiatry Clin Neurosci 2003; 253: 193–196.

70.

70. Saluz

Jiricny

Jost

. Genomic sequencing reveals a positive correlation between the kinetics of strand-specific DNA demethylation of the overlapping estradiol/glucocorticoid-receptor binding sites and the rate of avian vitellogenin mRNA synthesis. Proc Natl Acad Sci USA 1986; 83: 7167–7171.

71.

71. Truss

Chalepakis

Pina

. Transcriptional control by steroid hormones. J Steroid Biochem Mol Biol 1992; 41: 241–248.

72.

72. Yokomori

Moore

Negishi

. Sexually dimorphic DNA demethylation in the promoter of the Slp (sex-limited protein) gene in mouse liver. Proc Natl Acad Sci USA 1995; 92: 1302–1306.

73.

73. Griffon

Crocq

Pilon

. Dopamine D3 receptor gene: organization, transcript variants, and polymorphism associated with schizophrenia. Am J Med Genet 1996; 67: 63–70.

74.

74. Asherson

Mant

Holmans

. Linkage, association and mutational analysis of the dopamine D3 receptor gene in schizophrenia. Mol Psychiatry 1996; 1: 125–132.

75.

75. Williams

Spurlock

Holmans

. A meta-analysis and transmission disequilibrium study of association between the dopamine D3 receptor gene and schizophrenia. Mol Psychiatry 1998; 3: 141–149.

76.

76. Petronis

. The genes for major psychosis: aberrant sequence or regulation?. Neuropsychopharmacology 2000; 23: 1–12.

77.

77. Nestoros

Ban

Lehmann

. Transmethylation hypothesis of schizophrenia: methionine and nicotinic acid. Int Pharmacopsychiatry 1977; 12: 215–246.

78.

78. Cohen

Lipinski

Vuckovic

Prosser

. Blood S-adenosyl-L-methionine levels in psychiatric disorders. Am J Psychiatry 1982; 139: 229–231.

79.

79. Alonso-Aperte

Ubeda

Achon

Perez-Miguelsanz

Varela-Moreiras

. Impaired methionine synthesis and hypomethylation in rats exposed to valproate during gestation. Neurology 1999; 52: 750–756.

80.

80. Detich

Bovenzi

Szyf

. Valproate induces replication-independent active DNA demethylation. J Biol Chem 2003; 278: 27 586–27 592.

81.

81. Abdolmaleky

Smith

Faraone

. Methylomics in psychiatry: modulation of gene–environment interactions may be through DNA methylation. Am J Med Genet B Neuropsychiatr Genet 2004; 127: 51–59.

82.

82. Fink

Hays

Moore

Beck

. Alcohol-related problems in older persons. Determinants, consequences, and screening. Arch Intern Med 1996; 156: 1150–1156.

83.

83. Whelan

. Alcohol-related health problems in the elderly. Med J Aust 1995; 162: 325–327.

84.

84. Maier

Cramer

West

Sohrabji

. Alcohol exposure during the first two trimesters equivalent alters granule cell number and neurotrophin expression in the developing rat olfactory bulb. J Neurobiol 1999; 41: 414–423.

85.

85. Duman

Monteggia

. A neurotrophic model for stress-related mood disorders. Biol Psychiatry 2006; 59: 1116–1127.

86.

86. Frey

Andreazza

Cereser

. Effects of mood stabilizers on hippocampus BDNF levels in an animal model of mania. Life Sci 2006; 79: 281–286.

87.

87. Lee

Wynder

Schmidt

McCafferty

Shiekhattar

. Histone H3 lysine 4 demethylation is a target of nonselective antidepressive medications. Chem Biol 2006; 13: 563–567.

88.

88. Shi

Lan

Matson

. Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell 2004; 119: 941–953.

89.

89. Kato

Shimizu

Nagayoshi

Kaneko

Sasaki

Ikawa

. Genomic imprinting of the human serotonin-receptor (HTR2) gene involved in development of retinoblastoma. Am J Hum Genet 1996; 59: 1084–1090.

90.

90. Kato

Ikawa

Hayashizaki

Shibata

. Paternal imprinting of mouse serotonin receptor 2A gene Htr2 in embryonic eye: a conserved imprinting regulation on the RB/Rb locus. Genomics 1998; 47: 146–148.

91.

91. Bunzel

Blumcke

Cichon

. Polymorphic imprinting of the serotonin-2A (5-HT2A) receptor gene in human adult brain. Brain Res Mol Brain Res 1998; 59: 90–92.

92.

92. Polesskaya

Sokolov

. Differential expression of the “C” and “T” alleles of the 5-HT2A receptor gene in the temporal cortex of normal individuals and schizophrenics. J Neurosci Res 2002; 67: 812–822.

93.

93. Bray

Buckland

Hall

Owen

O'Donovan

. The serotonin-2A receptor gene locus does not contain common polymorphism affecting mRNA levels in adult brain. Mol Psychiatry 2004; 9: 109–114.

94.

94. Fatemi

. Reelin mutations in mouse and man: from reeler mouse to schizophrenia, mood disorders, autism and lissencephaly. Mol Psychiatry 2001; 6: 129–133.

95.

95. Grayson

Jia

Chen

. Reelin promoter hypermethylation in schizophrenia. Proc Natl Acad Sci USA 2005; 102: 9341–9346.

96.

96. Abdolmaleky

Thiagalingam

Wilcox

. Genetics and epigenetics in major psychiatric disorders: dilemmas, achievements, applications, and future scope. Am J Pharmacogenomics 2005; 5: 149–160.

97.

97. Abdolmaleky

Cheng

Russo

. Hypermethylation of the reelin (RELN) promoter in the brain of schizophrenic patients: a preliminary report. Am J Med Genet B Neuropsychiatr Genet 2005; 134: 60–66.

98.

98. Sengupta

Seto

. Regulation of histone deacetylase activities. J Cell Biochem 2004; 93: 57–67.

99.

99. Baylin

. DNA methylation and gene silencing in cancer. Nat Clin Pract Oncol 2005; 2(Suppl 1)S4–11.

100.

100. Szyf

Pakneshan

Rabbani

. DNA methylation and breast cancer. Biochem Pharmacol 2004; 68: 1187–1197.

101.

101. Egger

Liang

Aparicio

Jones

. Epigenetics in human disease and prospects for epigenetic therapy. Nature 2004; 429: 457–463.

102.

102. Himmelhoch

Fuchs

Symons

. A double-blind study of tranylcypromine treatment of major anergic depression. J Nerv Ment Dis 1982; 170: 628–634.

103.

103. Himmelhoch

Thase

Mallinger

Houck

. Tranylcypromine versus imipramine in anergic bipolar depression. Am J Psychiatry 1991; 148: 910–916.