Answer these questions (given attachments-study)

Epigenetics: molecular mechanisms and implications for disease Adam E. Handel1,2, George C. Ebers1,2 and Sreeram V. Ramagopalan1,2

1 Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Oxford, OX3 7BN, United Kingdom

2 Department of Clinical Neurology, Level 3, West Wing, University of Oxford, John Radcliffe Hospital, Oxford, OX3 9DU,

United Kingdom

Opinion

Glossary

Chorionicity: a term denoting whether a pair of twins share one placenta

(monochorionic) or have one placenta each (dichorionic).

Epigenetics: heritable changes to gene expression without alterations to the

underlying DNA sequence, mostly mediated by DNA methylation and histone

modifications.

Heritability: a concept designed to tease out the relative contributions of genes

and environment by comparing the concordance rate for a phenotype in

monozygotic twins with that in dizygotic twins.

MicroRNA (miRNA): small lengths of RNA that are not translated into proteins.

Zygosity: a term denoting whether a pair of twins is derived from the same pair

of ova and spermatozoa (monozygotic and thus genetically identical) or from

Epigenetics is rising to prominence in biology as a mechanism by which environmental factors have inter- mediate-term effects on gene expression without chan- ging the underlying genetic sequence. This can occur through the selective methylation of DNA bases and modification of histones. There are wide-ranging implications for the gene–environment debate and epi- genetic mechanisms are causing a reevaluation of many traditional concepts such as heritability. The reversible nature of epigenetics also provides plausible treatment or prevention prospects for diseases previously thought hard-coded into the genome. Here, we consider how growing knowledge of epigenetics is altering our un- derstanding of biology and medicine, and its implica- tions for future research.

The emerging importance of epigenetics in biology and medicine The field of epigenetics is revolutionising our understand- ing of biology, medicine and evolution. However, as a concept, epigenetics itself has undergone an evolutionary process over the past century. Originally, around the 1940 s, the term referred to the processes by which genes bring about specific phenotypes. Since then, it has changed in meaning, firstly to describe those effects on gene expres- sion resulting from the methylation of DNA and then finally to a term encompassing every mechanism that results in heritable changes to gene expression without affecting the DNA sequence [1]. It is in this latter sense that we use the term in this article.

Epigenetic alterations occur through DNA methylation of cytosine–guanine base (CpG) dinucleotides and a wide range of histone modifications, including methylation, acetylation, phosphorylation, sumoylation and ubiquitina- tion (Figure 1) [2], collectively known as the epigenome. Nearly all cells in the human body have the same genotype, but cells have hugely different phenotypes and this to some extent arises from differences in the epigenome.

Some, but by no means all, of these modifications can be inherited [1,3]. A critical distinction must be made between those changes to the epigenome passed on to further generations of cells through mitosis [1] and those passed on to further generations of offspring [3]. The mitotic inheritance of epigenetic marks is well accepted and is a critical component of cell differentiation. The mechanism

Corresponding authors: Ebers, G.C. ([email protected]); Ramagopalan, S.V. ([email protected]).

1471-4914/$ – see front matter � 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.molmed.2

by which the epigenome is transmitted via mitosis is becoming increasingly well understood as result of several recent studies into the role of DNA methyl transferases (DNMTs) and histone deacetylases (HDACs) [4–6]. Its transmission between different generations of organisms is a subject of considerable controversy as it was tradition- ally thought that epigenetic marks were cleared through the process of meiosis [7]. More recent studies suggest that microRNA (miRNA) molecules transmitted through meio- sis can restore the state of the epigenome in zygotes [8,9].

These molecular aspects of epigenetic inheritance have profound implications for medicine [10]. As the epigenome can be modified in response to environmental stimuli over a timescale far too short for any hard-coded genetic change to arise it presents a potential mechanism by which exposure to an environmental factor can have long-lasting effects over the life of the organism and perhaps into subsequent generations [11]. Current knowledge of the epigenetic basis of disease and treatment is very limited but recent years have seen critical advances in this field.

Appreciation of the molecular mechanisms underlying these effects and the epidemiological methods for studying them in populations are now advanced to the stage where important hypotheses can be raised about the implication of epigenetics for biology and medicine.

Epigenetic mechanisms The epigenetic status of the genome varies dynamically from tissue to tissue compared to the largely static DNA sequence [12]. Alterations to the epigenome involve a large number of enzymes acting in concert to methylate DNA and apply multiple post-transcriptional modifications to histones (Figure 1).

two separate pairs of ova and spermatozoa (dizygotic and thus as genetically

similar as siblings).

009.11.003 Available online 21 December 2009 7

Figure 1. Molecular mechanisms of epigenetic changes. (a) DNA prior to epigenetic silencing. The histone complex (H) is acetylated in the resting state (red groups). A

stretch of CpG is unmethylated. (b) LSD1 and HP1 bind to different regions of the histone complex. (c) LSD1 and HP1 recruit DNA methylases (DNMT3a/DNMT3b) and the

CpG island is methylated. This decreases transcription factor binding and gene expression. (d) MBP binds to the methylated CpG island and recruits HDAC. (e) Acetylated

histones condense and the DNA is epigenetically inactivated.

Opinion Trends in Molecular Medicine Vol.16 No.1

Mechanisms of DNA methylation

DNA methylation is a potent way of transcriptionally silencing genes [13]. It is a process that occurs at cytosine bases across the entire genome but is particularly concen- trated at CpG islands, regions of DNA highly enriched in sequences of a cytosine base followed directly by a guanine base. The enzymatic pathway leading to DNA methylation is relatively well understood now, with the involvement of multiple DNMT proteins – DNMT1, which is primarily responsible for maintaining previously extant DNA meth- ylation, and the DNMT3a/DNMT3b complex which estab- lishes de novo DNA methylation [14].

Mechanisms of histone modification

The sheer diversity of post-transcriptional modifications that are found on components of the histone octamer is impressive [15]. An equally wide array of proteins has been shown to bring about these changes, key among which are the histone acetyl transferases (HATs) and HDACs. These modifications have an effect on gene expression by altering the structural configuration of the genome through electrostatic interactions between variably charged histones and negatively charged DNA. This renders particular regions more or less accessible to transcription factors [16].

Interplay between DNA methylation and histone

modification

Neither mechanism operates in isolation. Methyl–CpG- binding proteins (MBPs) bind to regions of methylated

8

DNA and form complexes with HDACs which then result in histone deacetylation downstream of the meth- ylated CpG sequence, with transcriptional silencing of that region of the genome [14]. Similarly, proteins that either bind modified histones (heterochromatin protein 1, HP1) or directly modify histones (lysine- specific demethylase 1, LSD1 and protein arginine meth- yltransferase 5, PRMT1) recruit DNA methyltransfer- ases to induce DNA methylation in the silenced gene [17–19].

Regulation of epigenetic alterations

Unfortunately that still leaves the ‘‘chicken-and-egg’’ problem of how the establishment of these epigenetic mechanisms is regulated at the molecular level. The rules governing the establishment of epigenetic marks are not yet fully understood but there is compelling evidence that miRNA molecules have a central role in this process. This makes intuitive sense as miRNA can interact with DNA in a sequence-specific manner. EZH2 is a protein responsible for histone methylation [20]. A recent study in cultured breast cancer cells showed that both the expression of EZH2 and histone methylation of its target genes was decreased, depending on the level of a particular miRNA called miR-101 [21]. It is likely that miRNA control of epigenetic alterations has universal relevance in the regulation of cellular processes [22]. However, this cannot be the whole story because miR- NAs themselves are frequently under epigenetic control [23,24].

Table 1. Potential epigenetic diseasesa

Disease Potential mechanisms Evidence

Cancer Silencing of tumour suppressor genes or

activation of oncogenes through DNA

methylation and histone modifications

[31] mediated by miRNA [21]

Global alterations in DNA methylation and

histones in promoter regions of tumour

suppressor genes [30–32]

DNA hypermethylation of ATM increases risk of

breast cancer [87]

Epigenetic silencing of tumour suppression

miRNA in ovarian cancer and many types of

metastatic tumour [88,89]

miRNA-mediated histone modification in

cancerous cells [21]

Inherited epigenetic changes in a pedigree

with colon cancer [33]

Major psychosis Differential expression of stress-related

and neurodevelopmental genes through

DNA methylation [28]

Parent-of-origin effect in bipolar disorder [90]

Altered patterns of DNA methylation in

stress-related and neurodevelopmental

genes [28]

Stress/depression Reduced expression of glucocorticoid receptor

through DNA methylation [63]

Increased DNA methylation of glucocorticoid

receptor with child abuse in suicide victims [63]

Autism Differential expression of genes on 7q through

epigenetic imprinting [39]

Maternal parent-of-origin effects and altered

patterns of epigenetic imprinting in candidate

genes [39]

Beckwith–Wiedemann syndrome Failure of DNA methylation of KCNQ1OT1

encoding a non-translated RNA [26]

Differential DNA methylation of KCNQ1OT1 in

discordant MZ twins [26]

Immunodeficiency, centromere

instability, facial anomalies

syndrome

Failure of DNA methylation of multiple

immunological and developmental genes

by DNMT3b [29]

Most cases are associated with mutations in

DNMT3b and this results in aberrant DNA

methylation of multiple immunological and

developmental genes [29]

Asthma Reduced HDAC and increased HAT activity

causing decreased inhibition of NF-kB through

increased histone acetylation [38]

Maternal parent-of-origin effects [91]

Altered histone modifications in airways [38]

Chronic obstructive

pulmonary disease

Reduced HDAC causing decreased inhibition

of NF-kB through increased histone acetylation [38]

Altered histone modifications in airways [38]

Multiple sclerosis Increased silencing of FOXP3, a protein involved

in generating regulatory T cells through DNA

methylation [35]

Maternal parent-of-origin effects mediated by

HLA-DRB1*15 allele transmission [92–95]

Differential expression of HLA class II risk factors

through histone modifications [37]

Sex-specific transgenerational distortion of

transmission of HLA-DRB1*15 alleles [96]

Altered myelin processing as a result of increased

PAD2 expression through reduced DNA

methylation [97]

Few large pedigrees [98]

Month of birth effects (might be mediated

by folate or vitamin D) [99]

Small study showed differences in DNA

methylation in PAD2, a protein involved in

myelin processing [97]

Juvenile idiopathic arthritis Increased silencing of FOXP3, a protein involved

in generating regulatory T cells through DNA

methylation [35]

Maternal parent-of-origin effects [100]

Psoriatic arthritis Increased silencing of FOXP3, a protein involved

in generating regulatory T cells through DNA

methylation [35]

Paternal parent-of-origin effects [101]

Obesity Altered silencing of unidentified genes

by FTO, a DNA methyltransferase [50]

Parent-of-origin transgenerational

effects [46]

A susceptibility gene, FTO, encodes a

DNA methyltransferase [50] a This table illustrates some diseases in which epigenetics is likely to play a role in aetiology or pathogenesis.

Opinion Trends in Molecular Medicine Vol.16 No.1

Epigenetics and disease Is epigenetics relevant to medicine? The answer is an emphatic yes. The classical genetic conundrums of incom- plete penetrance and variable expressivity can, in part, be explained by differences in epigenetics [25]. Indeed, for the rare epigenetic disorder, Beckwith–Wiedemann syndrome, monozygotic (MZ) twins discordant for disease have been shown to be discordant for methylation [26]. DNA meth- ylation might also be important in aetiology for many common complex diseases, including cancer, psychosis and many others (Table 1) [10,27,28].

Epigenetic mechanisms and disease

The mechanisms underlying several diseases resulting from epigenetic factors are now known. For example, twins concordant for Beckwith–Wiedemann syndrome are also concordant for a failure of DNA methylation at KCNQ1OT1, encoding a non-translated RNA [26]. A mech- anism is also known for a rare condition known as immu- nodeficiency, centromere instability, facial anomalies syndrome. The majority of cases are caused by a mutation in DNMT3b that results in defective suppression of multiple immunological and developmental genes [29].

9

Opinion Trends in Molecular Medicine Vol.16 No.1

Epigenetics is a relatively new field, thus this degree of evidence is not yet available for many other conditions which are plausible, if not compelling, candidates for hav- ing an epigenetic basis. In neoplasia, there are clear epi- genetic abnormalities in cancerous cells, with aberrant global DNA methylation and histone acetylation [30–32]. The clearest inherited epigenetic mechanism has been seen in colon cancer, where the DNA mismatch repair enzyme encoded by MLH1 is inactivated by DNA methyl- ation in affected individuals across generations [33]. A recent study performed a genome-wide methylation screen in haematological malignancies and found several genes that showed differential DNA methylation in neoplasia [34]. In this way, even though the mechanisms by which these genes are regulated or their modes of action are not yet understood, epigenetic techniques have identified genes that might be important in cancer pathogenesis.

This is now true for several other disorders. For example, autoimmune or inflammatory diseases are often associated with dysregulation of T cell responses. FOXP3, a key gene in the development of regulatory T cells, is regulated by DNA methylation, suggesting that epigenetic mechanisms operating in T cells are likely to play a key role in these disorders [35]. Another potential epigenetic mech- anism of autoimmune disease susceptibility is the regula- tion of class II Human Leukocyte Antigens (HLA), a key risk factor in autoimmunity [36], by histone methylation [37]. Asthma and chronic obstructive pulmonary disease appear to share a common histone-related susceptibility pathway, in which deactivation of HDACs, and in asthma, activation of HATs, results in defective silencing of the proinflammatory gene NF-kB with subsequent upregula-

Figure 2. Molecular mechanisms of epigenetic inheritance. (a) Inheritance of epigenetic

methylation pattern is initially present only on the original strands of DNA. NP95 binds m

the complementary daughter strand of DNA [4]. When the strands associate with histon

histone modifications [1]. (b) Transgenerational inheritance of epigenetic modifications.

and histone modifications to a large extent [7,41]. Retained miRNA recruits DNMT3a/3

10

tion of multiple inflammatory mediators [38]. Recent data has implicated alterations in DNA methylation of genes involved in multiple pathways including stress response and neurodevelopment in the pathogenesis of major psy- chosis [28]. In autism, certain aspects of disease might result from dysregulated parental imprinting of genes on chromosome 7q and elsewhere [39].

Still, more diseases are likely to have epigenetic aspects but the mechanisms of these have not yet been identified. Therefore, epidemiological data will be useful to provide indirect evidence of epigenetic effects.

Epidemiological evidence of epigenetics

Epidemiological studies can undercover evidence that is highly suggestive of epigenetic aetiology. Many epigenetic changes are expressed differentially, dependent on the parent-of-origin and the sex of the offspring [40]. These effects result in sex-specific and parent-of-origin differ- ences in disease risk. Epigenetic alterations are unlikely to be stable over time or between generations, and thus epigenetic diseases are likely to have pedigrees of only limited size and will potentially manifest in an age-de- pendent and relapsing-remitting manner [13]. The degree to which epigenetic marks are transmitted between gener- ations of cells and generations of organisms is critical to both biology and medicine.

Epigenetic inheritance There are two distinct pathways by which epigenetic marks could be inherited (Figure 2). One is from one generation of cells to another and the other is from one generation of organisms to another. The evidence for

modifications through mitosis. In semi-conservative replication, the original DNA

ethylated DNA (shown in red) and recruits DNMT1 which methylates DNA bases on

e complexes, further enzymes such as HDACs are recruited to restore the original

The process of embryogenesis clears the epigenome of DNA methylation (MeDNA)

b complexes to restore the original epigenetic marks [8,9].

Opinion Trends in Molecular Medicine Vol.16 No.1

inheritance of the epigenome through mitosis is strong, but it is still unclear to what extent this is true through different generations of organisms.

Mitotic inheritance

A recent study demonstrated one mechanism by which mitotic inheritance might occur. NP95, a protein binding methylated DNA, recruits DNMT1 [4]. During semi-con- servative DNA replication this results in methylation of the daughter strand, with subsequent recruitment of histone-modifying proteins [5], thus maintaining the epi- genome in the subsequent generation of cells. Further evidence for the importance of this mechanism is provided by an NP95 knockout model in embryonic stem cells, which results in a failure of epigenetic inheritance through mitosis.

Transgenerational inheritance

Is the epigenome inherited between generations? It had generally been thought that the epigenetic state of the genome is cleared between generations [7,41], as what was known about DNA replication did not make allowance for methylation status. However there is now increasing evi- dence that transgenerational epigenetic inheritance occurs, and there is an expectation that this would be tightly regulated [42]. One potential mechanism, shown in a mouse model, by which both clearance of DNA meth- ylation after implantation of the embryo and inheritance of the epigenome can coexist, is through encoding epigenetic changes in RNA molecules which are themselves trans- mitted between generations [8]. The injection of miRNA molecules targeting CDK9, a gene involved in cardiac de- velopment, into mouse ova resulted in offspring suffering from epigenetically mediated cardiomyopathy [9]. Invert- ebrate experiments have shown that RNA-mediated phe- notypic effects can be transmitted through many generations [43,44]. It appears that the pathophysiological effects of epigenetic changes can be transmitted through several generations in mammals as well. Exposure of an initial generation of rats to endocrine disruptors resulted in deficient spermatogenesis mediated by changes to DNA methylation, and this was transmitted to three subsequent generations [45]. Thus, at least in animal models, envir- onmentally induced epigenetic changes are able to be inherited and presumably involve reestablishment of epigenetic marks after replication perhaps by RNA mech- anisms similar to what happens to CDK9.

The most convincing example of epigenetic inheritance in humans is the transmission of a cancer-associated epimutation in the MLH gene from a mother to one of three sons [33]. There is further indirect epidemiological evidence of epigenetic inheritance in humans. Looking at three generational families, Pembrey et al. showed that a paternal grandfather’s food supply was linked to the mortality risk of his grandsons, while a paternal grand- mother’s food supply was linked to the mortality risk of her granddaughters [46] – epigenetics provides a possible if not probable mechanism for this effect [45]. A recent study found that individuals whose parents were exposed to famine around the time of their conception showed altera- tion in DNA methylation at several candidate loci impli-

cated in metabolic pathways [47]. This does not prove epigenetic inheritance, however, as the zygotes respon- sible for generating the individual studies were also directly exposed to the same environmental circum- stances. In the same way, it is not surprising that a grandmother can alter the epigenetic state of sensitive alleles in the resulting next two generations as the cells of all three generations can be directly exposed to any environmental influence [48]. Pembrey et al. also showed that early paternal smoking was associated with increased body mass index in sons but not daughters [46]. One gene consistently linked to obesity, FTO [49], has been shown to transcribe a protein with DNA demethylase activity [50]. This provides a potential epi- genetic mechanism to explain the link between paternal smoking and filial obesity. The persistence of an altered phenotype in the fourth generation would allow us to test transmission of epigenetic marks over grandparental environmental–epigenome influences. Given the wealth of evidence that has shown that multigenerational epige- netic inheritance occurs in rodent models, it would be surprising if a similar mechanism did not exist in humans.

Inheritance of the epigenome is a powerful way by which adaption of one generation of organisms to a changed environment can be carried to many subsequent gener- ations. As we have shown, this can result in either adaptive or deleterious effects in these later generations. The implications of epigenetic inheritance for epidemiological studies have only just begun to be appreciated.

Epidemiological implications of epigenetic inheritance

Much effort in genetic epidemiology is concentrated on separating the relative contributions of genes and environ- ment to a given phenotype. The classical twin study was first proposed by Galton to do precisely this [51]. There have always been nagging reservations about stochastic factors and differing environment in these comparisons. A higher concordance rate between MZ twins compared with dizygotic (DZ) twins provides evidence for genes influen- cing phenotypes. MZ twins discordant for a trait are inevi- tably put forth in favour of environmental factors in complex disease aetiology. However, even this comfortable concept is under attack. Several studies implicate stochas- tic or random epigenetic differences between identical twins in the phenomenon of discordance [12,52,53].

One pivotal twin study showed that the distribution of DNA methylation across the genome in MZ twins diverged over time [52]. Very recent studies have revealed that DNA methylation in MZ twins is more similar than DZ twins [12,53], but unexpectedly that almost all of the similarity between MZ twins came from dichorionic rather than monochorionic MZ twins [12]. Different tissues also showed significantly different DNA methylation patterns [12].

So what are the implications of these results? Specu- lation that in addition to identical DNA, epigenetic sim- ilarity might also contribute to phenotypic similarity among MZ twin pairs compared to DZ twins has now been experimentally confirmed. There is evidence that mono- chorionicity in MZ twins is responsible for discordance in several traits [54], and thus the importance of chorionicity

11

Opinion Trends in Molecular Medicine Vol.16 No.1

(usually unknown) in twin studies cannot be underesti- mated, particularly now as we enter the era of epigenetics. Population registers which include chorionicity data on twins, such as the East Flanders Prospective Twin Survey [55], will be essential if effective genetic epidemiological research is to be conducted into the role of epigenetic factors in disease.

Measures of ‘heritability’ need to be reconsidered, as concordance is probably a reflection of genetic (including gene load), epigenetic, stochastic and environmental com- ponents [25,56]. Interpretation of heritability from twin studies is rendered even more problematic in diseases with a relatively late age-of-onset, as the epigenetic profiles of MZ twins have been shown to diverge over time [52,53].

Epigenetics and the environment Environmental factors can influence the epigenome (Table 2) [11]. The dynamic yet long-lasting nature of epigenetic changes is a critical way in which organisms can adapt rapidly to environmental changes. As surmised by Lamarck [57], if nature could develop a mechanism to translate environmental exposure and threat to the benefit of future offspring it would – and it has.

It is clear that environmental factors can have an effect on the phenotype manifested: in mice, the administration of methyl donor substances (folate, vitamin B12, choline or betadine) to mothers causes DNA methylation-induced suppression of the agouti gene and a shift towards brown coloration in their offspring [58–60].

In humans, perinatal exposure to diethylstilbestrol in female infants results in a subsequent increase in the risk of gynaecological malignancy, an effect that is thought to be mediated by altered DNA methylation [61]. It is likely that many other environmental factors act via epigenetic alterations as well [11].

Psychological aspects of the early environment also appear to exert an effect on the epigenome. Rats raised by mothers displaying low levels of maternal care demon- strated higher levels of DNA methylation in the glucocor- ticoid receptor (GR) promoter [62]. This was a specific effect of exposure to maternal nurturing: when the offspring of caring mothers were raised by animals showing low levels of maternal care, the effect on the epigenome was the same as for the true biological offspring. A similar effect on the DNA methylation of the GR promoter is seen in suicide victims who were abused as children [63], suggesting that

Table 2. Environmental factors affecting the epigenomea

Environmental factor Evidence for epigenetics

Methyl donor (folate, vitamin B12,

choline, betadine)

Administration to mice causes ch

in region near Agouti gene [59,6

Sex hormones Endocrine toxin causes inheritab

Perinatal exposure of human infa

gynaecological malignancy [61]

Smoking Paternal parent-of-origin transge

changes [46]

Nutrition Parent-of-origin transgenerationa

of epigenetic changes [46]

Vitamin D Vitamin D receptor binding comp

Stress Stressful upbringing causes epig

Childhood abuse associated with

human suicide victims [63] a This table shows environmental factors that could result in epigenetic changes.

12

stressful aspects of the environment might have a role in regulating the epigenome in humans. Epigenetics is thus plausibly involved in many of the diseases associated with stress [64].

Epigenetic marks at a tissue level can therefore be a reflection of an individual’s environmental exposures, and as such epigenetic marks can change during the lifetime of a cell/tissue [52]. The implications of changes in long-term, heritable gene expression are clearly of potentially huge importance to evolutionary theory.

Epigenetics and evolution As Dobzhansky [65] observed ‘‘nothing in biology makes sense except in the context of evolution’’. It might be speculated that if nature could find a way to transmit the mother’s environmental experience or stress to the benefit of the offspring she would put it into play [66].

In microorganisms there is good evidence that heritable epigenetic alterations underlie many rapid phenotypic changes, for example those influencing virulence proteins in bacteria and malaria [67,68]. A wealth of evidence suggests that a similar, heritable interaction between environmental exposure occurs in mammals [45–47]. This adaptation to environmental conditions thus occurs over just a few generations, orders of magnitude faster than that mediated by most DNA mutational mechanisms [66]. As adaptation will impact on reproductive fitness it follows that epigenomics will directly affect natural selection. Furthermore, as methylated cytosine undergoes base mutation far more rapidly than unmethylated cytosine (10–7 per generation vs. 10�8 per generation) epigenetics can also directly impact on genomic mutagenesis [69].

The concept of directed mutagenesis raises the long- neglected spectre of Lamarckian evolution from the scien- tific graveyard. It is likely that Darwinian evolution and Lamarckian theory (which Darwin accepted) have coex- isted peacefully through the medium of epigenetics.

Epigenetics and medicine The effect of the environment on epigenomics has intriguing and profound public health implications: the fact that an individual’s behaviours can affect the next two generations is rarely considered. The implications for the search for environmental factors, particularly in the ubiquitous case- control study, are enormous. Similarly, if environmental factors can impact on disease susceptibility then treatments

anges in coat colour mediated by altered DNA methylation

0]

le defects in DNA methylation-mediated spermatogenesis in rats [45]

nts to diethylstilbesterol causes epigenetic increase in

nerational effects on obesity in sons is highly suggestive of epigenetic

l effect on mortality in sex-matched offspring is highly suggestive

lex can cause epigenetic changes [102]

enetic changes in glucocorticoid receptor promoter region in rats [62]

epigenetic changes in glucocorticoid receptor promoter region in

Opinion Trends in Molecular Medicine Vol.16 No.1

(which are, after all, environmental factors themselves) might also be able to alter the course of disease.

Epidemiology

The long-term changes in gene expression and the ability of these changes to be transmitted to later generations characteristic of epigenetic marks imply that exposure to the causative factor might precede detection of any altera- tion in disease incidence or prevalence by many years [46].

Using multiple sclerosis (MS) as an example, there has been a progressive increase in incidence over the past century resulting from an increase in MS among females [70]. The change has occurred too rapidly to be mediated by anything other than currently unidentified environmental factors and the transgenerational aspects to this trend suggest epigenetics. The potential involvement of epige- netics renders a precise timing of when the putative environmental factor could have acted difficult to detect by the ubiquitous case-control studies now in use.

Epigenetics also has wide-ranging methodological implications within both epidemiology and genetic studies. Epigenetic factors operate at the interface between genetics and environment and have the potential to violate the assumption of the independence of genotype from environmental factors inherent in some epidemiological techniques [71].

Epigenetics might also affect our interpretation of pub- lished studies of genome-wide association. If epigenetic changes are sufficiently permanent, these epimutations will be in linkage disequilibrium with the single nucleotide polymorphisms (SNPs) commonly used to investigate potentially causative regions of the genome [72]. Thus, genome-wide association studies might report both signifi- cant genetic and epigenetic loci.

Public health

The most exciting part of epigenetics in modern medicine is the possibility of intervening where the genome and environment intersect, as unlike the DNA sequence, epi- genetic changes are reversible.

There is no direct evidence that any current public health measures address underlying epigenetic causation of disease. It is tempting to hypothesise that some of the efficacy of folate supplements in the prevention of neural tube defects might be secondary to epigenetic changes. In support of this, a mouse model of neural tube defects was found to be associated with SNPs in a CpG island within the candidate gene promoter, a region preserved between humans and mice [73]. A critical area for public health research will be evaluating what use can be made of the

Table 3. Epigenetic mechanisms of anticancer drugsa

Agent Cancers

5-aza-20-deoxycytidine Myelodysplastic syn

acute myeloid leuka

Depsipeptide Chronic lymphocyti

acute myeloid leuka

and cutaneous T ce

Valproate Myelodysplastic syn

acute myeloid leuka

Suberoylanilide hydroxamic acid Cutaneous T cell lym a This table summarises what is currently known about the mechanisms by which som

malleable new epigenetic aspects of aetiology to render common ‘‘genetic’’ diseases, such as cancer, diabetes and MS, preventable or curable.

Epigenetics has implications for population exposure to environmental factors beyond the confusion it can cause epidemiologists. A key aspect of the transgenerational effects seen with DNA methylation or histone modifi- cations is that these alterations might endure long after the original environmental stimulus has been removed. Unfortunately, intra- and transgenerational decay of epi- genetic changes is not yet well understood. However, Slat- kin has recently proposed a model of epigenetic inheritance which might allow estimates of the expected population- wide effects of epigenetics when rates of transgeneration decay are elucidated [72]. The implications of this model are that, dependent on the permanency of epigenetic changes, even if public health officials act as quickly as possible to limit exposure, the effects of the exposure might still endure for several generations [74].

Epigenetic treatment

The best public health efforts are unlikely to eradicate disease of epigenetic origin entirely and thus treatments based on epigenomic alterations must be considered. Thus far, most research into epigenetic treatment modalities has focused on cancer, largely because this is the disease in which the role of epigenetics is currently best understood.

Several different classes of drugs have been used to intervene in the epigenetic processes contributing to the development of cancer (Table 3) [13]. Inhibition of DNA methylation and histone deacetylation has shown promise in clinical trials in myelodysplastic syndrome [75–77], acute myeloid leukaemia [78,79] and T cell lymphoma [80,81] with still more promising treatment candidates on the horizon [13].

Given the unexpected finding that valproic acid, a drug commonly used to treat epilepsy, has epigenetic effects [82], it is highly probable that many other drugs are epigenetically active. Certainly, the tricyclic antidepress- ant, imipramine, and steroids appear capable of altering the epigenome [38,83]. These previously unconsidered aspects of drug activity have the potential to impact hugely on our therapeutic armamentarium.

There are several potential limitations of current epi- genetic treatments. The most serious problem is a lack of specificity: although increased epigenetic silencing of tumour suppressor genes undoubtedly contributes to car- cinogenesis, simply inducing global DNA demethylation can itself result in chromosomal instability [84]. One method to circumvent these drawbacks would be to use

Mechanism

drome [75,76,78],

emia [78]

Inhibits DNA methylase

c leukaemia [79],

emia [79], peripheral

ll lymphoma [81]

Inhibits histone deacetylase

drome [77],

emia [77]

Inhibits histone deacetylase

phoma [80] Inhibits histone deacetylase

e anticancer drugs currently in use have an epigenetic action.

13

Box 1. Outstanding questions

� How are epigenetic marks established at specific regions within the genome? Further studies into sequence-specific candidate

proteins and miRNA might provide some answers.

� Are epigenetic changes transmitted between generations of humans? Only long-term, molecular studies of defined cohorts

will be able to answer this question.

� How does the epigenome change over the lifetime of an organism? Longitudinal twin studies should provide definitive

answers in the future.

� Which diseases have an epigenetic basis? Our understanding has been greatly improved over the past few years but further work,

particularly in autoimmune disease where there is strong

epidemiological evidence suggesting a role for epigenetics, is

required. Genome-wide, tissue-specific DNA methylation and

chromatin immunoprecipitation (ChIP) studies might yield an

answer for many diseases in the near future.

� How can we translate our knowledge of epigenetics into treatments? There are many valiant efforts underway to bring

epigenetically active treatments to clinical trials. However, where

epigenetics could make the biggest difference to health is in the

prevention of diseases, and only through increased understand-

ing of which epigenetic susceptibility loci would be amenable to

modification will that become a reality.

Opinion Trends in Molecular Medicine Vol.16 No.1

morespecifictreatments,such asRNA moleculesthatwould interfere with aberrant epigenetic changes [21]. Epigenetic effects of treatment might also have the potential to be transmitted between generations which could limit the use but might also have wide-ranging implications for cur- rent drug treatments that alter the epigenome with or without the knowledge of physicians. Current knowledge of the action of these agents in cancer suggests that their effect is reversed soon after termination of therapy [13], but there is no guarantee that this would occur in conditions in which there is no abnormal turnover of cells.

Epigenetic reprogramming

A potentially powerful treatment aspect that epigenetic medicine makes a real prospect is the generation of plur- ipotent stem cells from somatic cells. Differentiation, as discussed earlier, is the result of epigenetic changes result- ing in the subsequent loss of pluripotential. Resetting the epigenomeofsuchcellscouldgenerateatheoreticallyunlim- ited number of stem cells for therapeutic applications [85]. A recent study has illustrated how this might be done by using oestrogen-related receptor b (Esrrb) to induce a pluripoten- tial phenotype in murine fibroblasts with resetting of the epigenome to one similar to stem cells [86].

Epigenetic changes are likely to have effects on disease risk across entire populations but the very fact that these effects exist opens up an entirely new vista of treatment opportunities.

Concluding remarks Epigenetics holds substantial promise in helping to explain many previously intractable conundrums in human genetics. Changes are brought about by the concerted action of a huge multitude of proteins, including DNA methyltransferases, histone deacetylases and histone acetyl transferases. Consensus opinion suggests that epi- genetic marks are transmitted through mitosis to future generations of cells. However, despite strong epi- demiological and some molecular evidence for the trans- mission of these modifications to subsequent generations of organisms, the details still remain controversial. Transge- nerational inheritance of the epigenome has immense implications for previous comfortable epidemiological con- cepts such as measures of ‘heritability’ derived from twin studies and the ability of case-control studies to pin down the precise moment of exposure to environmental risk factors. There is rapidly accumulating research implicat- ing the involvement of epigenetic mechanisms in many common conditions. Treatments aimed at manipulating the epigenome are currently under trial for many haema- tological malignancies and far more will no doubt begin in the near future. Further efforts should be exerted towards manipulating the epigenome to interfere with conditions previously attributed to immutable genetic risk. Direct manipulation of the epigenome, now emerging as a real possibility, could conceivably generate an unlimited quantity of stem cells for the treatment of vast numbers of degenerative conditions. Although there is much work still to do (Box 1), the epigenome can no longer be ignored and promises to completely revolutionise our understand- ing of cellular, evolutionary and medical biology.

14

Acknowledgements We would like to thank all members of the Ebers group for hugely helpful support and advice. In particular, we would like to mention Lahiru Handunnetthi, Julia Morahan and Katie Morrison, discussions with whom have been invaluable in the preparation of this manuscript.

References 1 Probst, A.V. et al. (2009) Epigenetic inheritance during the cell cycle.

Nat. Rev. Mol. Cell Biol. 10, 192–206 2 Bernstein, B.E. et al. (2007) The mammalian epigenome. Cell 128,

669–681 3 Youngson, N.A. and Whitelaw, E. (2008) Transgenerational

epigenetic effects. Annu. Rev. Genomics Hum. Genet. 9, 233–257 4 Sharif, J. et al. (2007) The SRA protein Np95 mediates epigenetic

inheritance by recruiting Dnmt1 to methylated DNA. Nature 450, 908–912

5 Rountree, M.R. et al. (2000) DNMT1 binds HDAC2 and a new co- repressor, DMAP1, to form a complex at replication foci. Nat. Genet. 25, 269–277

6 Fuks, F. et al. (2000) DNA methyltransferase Dnmt1 associates with histone deacetylase activity. Nat. Genet. 24, 88–91

7 Mayer, W. et al. (2000) Demethylation of the zygotic paternal genome. Nature 403, 501–502

8 Rassoulzadegan, M. et al. (2006) RNA-mediated non-mendelian inheritance of an epigenetic change in the mouse. Nature 441, 469– 474

9 Wagner, K.D. et al. (2008) RNA induction and inheritance of epigenetic cardiac hypertrophy in the mouse. Dev. Cell 14, 962– 969

10 Feinberg, A.P. (2008) Epigenetics at the epicenter of modern medicine. J. Am. Med. Assoc. 299, 1345–1350

11 Jirtle, R.L. and Skinner, M.K. (2007) Environmental epigenomics and disease susceptibility. Nat. Rev. Genet. 8, 253–262

12 Kaminsky, Z.A. et al. (2009) DNA methylation profiles in monozygotic and dizygotic twins. Nat. Genet. 41, 240–245

13 Ptak, C. and Petronis, A. (2008) Epigenetics and complex disease: from etiology to new therapeutics. Annu. Rev. Pharmacol. Toxicol. 48, 257–276

14 Rottach, A. et al. (2009) DNA methylation-mediated epigenetic control. J. Cell. Biochem. 108, 43–51

15 Kouzarides, T. (2007) Chromatin modifications and their function. Cell 128, 693–705

16 Laskowski, R.A. and Thornton, J.M. (2008) Understanding the molecular machinery of genetics through 3D structures. Nat. Rev. Genet. 9, 141–151

Opinion Trends in Molecular Medicine Vol.16 No.1

17 Smallwood, A. et al. (2007) Functional cooperation between HP1 and DNMT1 mediates gene silencing. Genes Dev. 21, 1169–1178

18 Wang, J. et al. (2009) The lysine demethylase LSD1 (KDM1) is required for maintenance of global DNA methylation. Nat. Genet. 41, 125–129

19 Zhao, Q. et al. (2009) PRMT5-mediated methylation of histone H4R3 recruits DNMT3A, coupling histone and DNA methylation in gene silencing. Nat. Struct. Mol. Biol. 16, 304–311

20 Bracken, A.P. et al. (2003) EZH2 is downstream of the pRB-E2F pathway, essential for proliferation and amplified in cancer. EMBO J. 22, 5323–5335

21 Varambally, S. et al. (2008) Genomic loss of microRNA-101 leads to overexpression of histone methyltransferase EZH2 in cancer. Science 322, 1695–1699

22 Chuang, J.C. and Jones, P.A. (2007) Epigenetics and microRNAs. Pediatr. Res. 61, 24R–29R

23 Saito, Y. et al. (2006) Specific activation of microRNA-127 with downregulation of the proto-oncogene BCL6 by chromatin- modifying drugs in human cancer cells. Cancer Cell 9, 435–443

24 Lujambio, A. et al. (2007) Genetic unmasking of an epigenetically silenced microRNA in human cancer cells. Cancer Res. 67, 1424–1429

25 Petronis, A. (2006) Epigenetics and twins: three variations on the theme. Trends Genet. 22, 347–350

26 Weksberg, R. et al. (2002) Discordant KCNQ1OT1 imprinting in sets of monozygotic twins discordant for Beckwith–Wiedemann syndrome. Hum. Mol. Genet. 11, 1317–1325

27 Feinberg, A.P. and Tycko, B. (2004) The history of cancer epigenetics. Nat. Rev. Cancer 4, 143–153

28 Mill, J. et al. (2008) Epigenomic profiling reveals DNA-methylation changes associated with major psychosis. Am. J. Hum. Genet. 82, 696– 711

29 Jin, B. et al. (2008) DNA methyltransferase 3B (DNMT3B) mutations in ICF syndrome lead to altered epigenetic modifications and aberrant expression of genes regulating development, neurogenesis and immune function. Hum. Mol. Genet. 17, 690–709

30 Fraga, M.F. et al. (2005) Loss of acetylation at Lys16 and trimethylation at Lys20 of histone H4 is a common hallmark of human cancer. Nat. Genet. 37, 391–400

31 Esteller, M. (2007) Cancer epigenomics: DNA methylomes and histone-modification maps. Nat. Rev. Genet. 8, 286–298

32 Seligson, D.B. et al. (2005) Global histone modification patterns predict risk of prostate cancer recurrence. Nature 435, 1262–1266

33 Hitchins, M.P. et al. (2007) Inheritance of a cancer-associated MLH1 germ-line epimutation. N. Engl. J. Med. 356, 697–705

34 Martin-Subero, J.I. et al. (2009) A comprehensive microarray-based DNA methylation study of 367 hematological neoplasms. PLoS ONE 4, e6986

35 Huehn, J. et al. (2009) Epigenetic control of FOXP3 expression: the key to a stable regulatory T-cell lineage? Nat. Rev. Immunol. 9, 83–89

36 Fernando, M.M. et al. (2008) Defining the role of the MHC in autoimmunity: a review and pooled analysis. PLoS Genet. 4, e1000024

37 Chou, S.D. and Tomasi, T.B. (2008) Spatial distribution of histone methylation during MHC class II expression. Mol. Immunol. 45, 971– 980

38 Adcock, I.M. et al. (2005) Histone deacetylation: an important mechanism in inflammatory lung diseases. COPD 2, 445–455

39 Schanen, N.C. (2006) Epigenetics of autism spectrum disorders. Hum. Mol. Genet. 15 (Spec. No. 2), R138–R150

40 Reik, W. and Walter, J. (2001) Genomic imprinting: parental influence on the genome. Nat. Rev. Genet. 2, 21–32

41 Rougier, N. et al. (1998) Chromosome methylation patterns during mammalian preimplantation development. Genes Dev. 12, 2108– 2113

42 Morgan, D.K. and Whitelaw, E. (2008) The case for transgenerational epigenetic inheritance in humans. Mamm. Genome 19, 394–397

43 Fire, A. et al. (1998) Potent and specific genetic interference by double- stranded RNA in Caenorhabditis elegans. Nature 391, 806–811

44 Vastenhouw, N.L. et al. (2006) Gene expression: long-term gene silencing by RNAi. Nature 442, 882

45 Anway, M.D. et al. (2005) Epigenetic transgenerational actions of endocrine disruptors and male fertility. Science 308, 1466–1469

46 Pembrey, M.E. et al. (2006) Sex-specific, male-line transgenerational responses in humans. Eur. J. Hum. Genet. 14, 159–166

47 Tobi, E.W. et al. (2009) DNA methylation differences after exposure to prenatal famine are common and timing- and sex-specific. Hum. Mol. Genet. 18, 4046–4053

48 Whitelaw, N.C. and Whitelaw, E. (2006) How lifetimes shape epigenotype within and across generations. Hum. Mol. Genet. 15 (Spec. No. 2), R131–R137

49 Qi, L. et al. (2008) Fat mass-and obesity-associated (FTO) gene variant is associated with obesity: longitudinal analyses in two cohort studies and functional test. Diabetes 57, 3145–3151

50 Gerken, T. et al. (2007) The obesity-associated FTO gene encodes a 2- oxoglutarate-dependent nucleic acid demethylase. Science 318, 1469– 1472

51 Galton, F. (1883) ‘‘History of Twins’’ in Human Faculty and its Development, McMillan and Co

52 Fraga, M.F. et al. (2005) Epigenetic differences arise during the lifetime of monozygotic twins. Proc. Natl. Acad. Sci. U. S. A. 102, 10604–10609

53 Boks, M.P. et al. (2009) The relationship of DNA methylation with age, gender and genotype in twins and healthy controls. PLoS ONE 4, e6767

54 Race, J.P. et al. (2006) Chorion type, birthweight discordance and tooth-size variability in Australian monozygotic twins. Twin Res. Hum. Genet. 9, 285–291

55 Derom, C.A. et al. (2006) The East Flanders Prospective Twin Survey (EFPTS). Twin Res. Hum. Genet. 9, 733–738

56 Ramagopalan, S.V. et al. (2008) Genetic epidemiology: the use of old and new tools for multiple sclerosis. Trends Neurosci. 31, 645–652

57 Lamarck JBPAdCdM (1809) Philosophie zoologique: ou Exposition des considâerations relative áa l’histoire naturelle des animaux. Paris: Dentu et L’Auteur

58 Waterland, R.A. and Jirtle, R.L. (2003) Transposable elements: targets for early nutritional effects on epigenetic gene regulation. Mol. Cell. Biol. 23, 5293–5300

59 Wolff, G.L. et al. (1998) Maternal epigenetics and methyl supplements affect agouti gene expression in Avy/a mice. FASEB J. 12, 949–957

60 Cooney, C.A. et al. (2002) Maternal methyl supplements in mice affect epigenetic variation and DNA methylation of offspring. J. Nutr. 132, 2393S–2400S

61 Li, S. et al. (2003) Environmental exposure, DNA methylation, and gene regulation: lessons from diethylstilbestrol-induced cancers. Ann. N. Y. Acad. Sci. 983, 161–169

62 Weaver, I.C.G. et al. (2004) Epigenetic programming by maternal behavior. Nat. Neurosci. 7, 847–854

63 McGowan, P.O. et al. (2009) Epigenetic regulation of the glucocorticoid receptor in human brain associates with childhood abuse. Nat. Neurosci. 12, 342–348

64 Rao, K. (2009) Recent research in stress, coping and women’s health. Curr. Opin. Psychiatry 22, 188–193

65 Dobzhansky, T. (1964) Biology, molecular and organismic. Am. Zool. 4, 443–452

66 Rando, O.J. and Verstrepen, K.J. (2007) Timescales of genetic and epigenetic inheritance. Cell 128, 655–668

67 Hernday, A. et al. (2002) Self-perpetuating epigenetic pili switches in bacteria. Proc. Natl. Acad. Sci. U. S. A. 99 (Suppl. 4), 16470– 16476

68 Roberts, D.J. et al. (1992) Rapid switching to multiple antigenic and adhesive phenotypes in malaria. Nature 357, 689–692

69 Garrick, D. et al. (1998) Repeat-induced gene silencing in mammals. Nat. Genet. 18, 56–59

70 Orton, S-M. et al. (2006) Sex ratio of multiple sclerosis in Canada: a longitudinal study. Lancet Neurol. 5, 932–936

71 Ogbuanu, I.U. et al. (2009) Can we apply the Mendelian randomization methodology without considering epigenetic effects? Emerg. Themes Epidemiol. 6, 3

72 Slatkin, M. (2009) Epigenetic inheritance and the missing heritability problem. Genetics 182, 845–850

73 Rogner, U.C. et al. (2002) SNPs in the CpG island of NAP1L2: a possible link between DNA methylation and neural tube defects? Am. J. Med. Genet. 110, 208–214

74 Handel, A.E. and Ramagopalan, S.V. (2009) Public health implications of epigenetics. Genetics 182, 1397–1398

75 Wijermans, P. et al. (2000) Low-dose 5-aza-20-deoxycytidine, a DNA hypomethylating agent, for the treatment of high-risk myelodysplastic

15

Opinion Trends in Molecular Medicine Vol.16 No.1

syndrome: a multicenter phase II study in elderly patients. J. Clin. Oncol. 18, 956–962

76 Kantarjian, H. et al. (2006) Decitabine improves patient outcomes in myelodysplastic syndromes: results of a phase III randomized study. Cancer 106, 1794–1803

77 Kuendgen, A. et al. (2005) Results of a phase 2 study of valproic acid alone or in combination with all-trans retinoic acid in 75 patients with myelodysplastic syndrome and relapsed or refractory acute myeloid leukemia. Ann. Hematol. 84 (Suppl. 1), 61–66

78 Garcia-Manero, G. et al. (2006) Phase 1/2 study of the combination of 5-aza-20-deoxycytidine with valproic acid in patients with leukemia. Blood 108, 3271–3279

79 Byrd, J.C. et al. (2005) A phase 1 and pharmacodynamic study of depsipeptide (FK228) in chronic lymphocytic leukemia and acute myeloid leukemia. Blood 105, 959–967

80 Duvic, M. et al. (2007) Phase 2 trial of oral vorinostat (suberoylanilide hydroxamic acid, SAHA) for refractory cutaneous T-cell lymphoma (CTCL). Blood 109, 31–39

81 Piekarz, R.L. et al. (2001) Inhibitor of histone deacetylation, depsipeptide (FR901228), in the treatment of peripheral and cutaneous T-cell lymphoma: a case report. Blood 98, 2865–2868

82 Milutinovic, S. et al. (2007) Valproate induces widespread epigenetic reprogramming which involves demethylation of specific genes. Carcinogenesis 28, 560–571

83 Tsankova, N.M. et al. (2006) Sustained hippocampal chromatin regulation in a mouse model of depression and antidepressant action. Nat. Neurosci. 9, 519–525

84 Eden, A. et al. (2003) Chromosomal instability and tumors promoted by DNA hypomethylation. Science 300, 455

85 Hayashi, K. and Surani, M.A. (2009) Resetting the epigenome beyond pluripotency in the germline. Cell Stem Cell 4, 493–498

86 Feng, B. et al. (2009) Reprogramming of fibroblasts into induced pluripotent stem cells with orphan nuclear receptor Esrrb. Nat. Cell Biol. 11, 197–203

87 Flanagan, J.M. et al. (2009) Gene-body hypermethylation of ATM in peripheral blood DNA of bilateral breast cancer patients. Hum. Mol. Genet. 18, 1332–1342

16

88 Zhang, L. et al. (2008) Genomic and epigenetic alterations deregulate microRNA expression in human epithelial ovarian cancer. Proc. Natl. Acad. Sci. U. S. A. 105, 7004–7009

89 Lujambio, A. et al. (2008) A microRNA DNA methylation signature for human cancer metastasis. Proc. Natl. Acad. Sci. U. S. A. 105, 13556– 13561

90 McInnis, M.G. et al. (2003) Genome-wide scan of bipolar disorder in 65 pedigrees: supportive evidence for linkage at 8q24, 18q22, 4q32, 2p12, and 13q12. Mol. Psychiatry 8, 288–298

91 Moffatt, M.F. and Cookson, W.O. (1998) The genetics of asthma. Maternal effects in atopic disease. Clin. Exp. Allergy 28 (Suppl. 1), 56–61

92 Ebers, G.C. et al. (2004) Parent-of-origin effect in multiple sclerosis: observations in half-siblings. Lancet 363, 1773–1774

93 Herrera, B.M. et al. (2008) Parent-of-origin effects in MS: observations from avuncular pairs. Neurology 71, 799–803

94 Hoppenbrouwers, I.A. et al. (2008) Maternal transmission of multiple sclerosis in a Dutch population. Arch. Neurol. 65, 345–348

95 Ramagopalan, S.V. et al. (2008) Parental transmission of HLA- DRB1*15 in multiple sclerosis. Hum. Genet. 122, 661–663

96 Chao, M.J. et al. (2009) Epigenetics in multiple sclerosis susceptibility: difference in transgenerational risk localizes to the major histocompatibility complex. Hum. Mol. Genet. 18, 261–266

97 Mastronardi, F.G. et al. (2007) Peptidyl argininedeiminase 2 CpG island in multiple sclerosis white matter is hypomethylated. J. Neurosci. Res. 85, 2006–2016

98 Willer, C.J. et al. (2007) A genome-wide scan in forty large pedigrees with multiple sclerosis. J. Hum. Genet. 52, 955–962

99 Willer, C.J. et al. (2005) Timing of birth and risk of multiple sclerosis: population based study. BMJ 330, 120

100 Zeft, A. et al. (2008) Familial autoimmunity: maternal parent-of-origin effect in juvenile idiopathic arthritis. Clin. Rheumatol. 27, 241–244

101 Karason, A. et al. (2003) A susceptibility gene for psoriatic arthritis maps to chromosome 16q: evidence for imprinting. Am. J. Hum. Genet. 72, 125–131

102 Issa, L.L. et al. (1998) Molecular mechanism of vitamin D receptor action. Inflamm. Res. 47, 451–475