EPIGENETICS IN REPRODUCTIVE MEDICINE
I. Basic concepts of epigeneticsDespite the fact that every cell
in a human body contains the same genetic material, not every cell
looks or behaves the same. Long nerve cells stretch out the entire
length of an arm or a leg; cells in the retina of the eye can sense
light; immune cells patrol the body for invaders to destroy. How
does each cell retain its unique properties when, in its
DNA-containing nucleus, it has the same master set of genes as
every other cell? The answer is in the epigenetic regulation of the
genes [1].
All cells in the human body carry the same DNA complement, which
originates from a single cell at conception. Highly orchestrated
epigenetic mechanisms facilitate the complex patterning required to
ensure normal human development and support stable regulation of
appropriate patterns of gene expression in diverse cell types.
Epigenetic mechanisms define mitotically heritable differences in
gene expression potential without altering the primary DNA
sequence. These mechanisms are highly regulated by a large number
of proteins that establish, read, and erase specific epigenetic
modifications, thereby defining where and when the transcriptional
machinery can access the primary DNA sequences to drive normal
growth and differentiation in the developing embryo and fetus.
Several types of epigenetic marks work in concert to drive
appropriate gene expression. These include DNA methylation at CpG
dinucleotides, covalent modifications of histone proteins, ncRNAs,
and other complementary mechanisms controlling higher order
chromatin organization within the cell nucleus [2].
1. DNA methylationDNAmethylation is typically associated with
gene silencing through binding of methylation-sensitive DNA binding
proteins and/or by interacting with various modifications of
histone proteins that modulate access of gene promoters to
transcriptional machinery . In eukaryotic species, DNA methylation
involves transfer of a methyl group to the cytosine of the CpG
dinucleotide .The vast majority of mammalian DNA methylation occurs
at CpG dinucleotides. CpGs are distributed non randomly in the
genome. They are concentrated in genomic regions called CpG islands
ranging in size from 200 bp to several kilobases. These CpG
islands, often unmethylated, are typically located within gene
promoters of actively transcribed housekeeping genes and tumor
suppressor genes [2].
Specific proteins, such as DNA methyltransferases, can establish
or maintain DNA methylation patterns. Studies in mice demonstrate
that DNA methyltransferases are essential for normal embryonic
development. DNA methylation is established de novo by the DNA
methyltranferase (DNMT) enzymes DNMT3a and DNMT3band maintained
through mitosis primarily by the DNMT1 enzyme. DNMT1 is the primary
maintenance methyltransferase with a high affinity for
hemimethylated DNA. Its primary function is to copy the methylation
patterns during replication. DNMT1o, one developmental
stage-specific isoform of DNMT1, is an oocyte-derived protein that
enters nuclei at the eight-cell stage of early embryos and has an
essential role in maintenance of epigenetic marks [2].
DNA demethylation is also critical during primordial germ cell
(PGC) and early embryo development. This can occur via passive
demethylation that is associated with cell division or via active
demethylation using excision repair mechanisms. The active pathway
requires hydroxylation of the 5-methylcytosine to
5-hydroxymethylcytosine by the enzymes TET1 and TET2, followed by
deamination by AID and APOBEC1 before base or nucleotide excision
repair. All the enzymes in this pathway are expressed in mouse
PGCs, suggesting a role in gametic epigenetic reprogramming
[2].
2. Histone ModificationThe basic unit of chromatin consists of
an octamer of histone proteins, two each of H2A, H2B, H3, and H4.
DNA wraps around this core, which provides structural stability and
the capacity to regulate gene expression (Fig. 1). Each core
histone within the nucleosome contains a globular domain and a
highly dynamic N-terminal tail extending from the globular domains
(Fig. 1). Histone proteins have tails that can have a number of
post-translational modifications including acetylation,
methylation, phosphorylation, ubiquitylation, sumoylation,
ADP-ribosylation, proline isomerization, citrullination,
butyrylation, propionylation, and glycosylation. Recently published
data from the ENCODE Project Consortium analyzed 11 such histone
post-translational modifications including acetylation and
methylation, which mark active and repressive chromatin, as well as
modifications associated with transcription. Assessing various
histone modifications in a number of tissues, that data set
identified different chromatin states including inactive, bimodal,
and active, each of which has different functional properties.
Bimodal states, in which a combination of active and repressive
marks are present in the chromatin of a promoter region of a gene,
facilitate rapid changes in gene expression, as might be expected
during early development, when differentiation and specification
occur [2].
Chromatin assembly and disassembly are highly orchestrated
processes that are coordinated by histone chaperones and ATP
dependent chromatin remodeling complexes. Histone chaperones
promote chromatin assembly by preventing non-specific histone-DNA
interactions while also promoting the correct histone-DNA
interactions [3].3. Regulatory ncRNAsRegulatory ncRNAs including
small interfering RNAs (siRNAs), microRNAs (miRNAs), and long
ncRNAs (lncRNAs) play important roles in gene expression regulation
at several levels: transcription, mRNA degradation, splicing, and
translation. SiRNAs are double-stranded RNAs (dsRNA) that mediate
post-transcriptional silencing, in part by inducing heterochromatin
to recruit histone deacetylase complexes [2].MiRNAs comprise a
novel class of endogenous, small (1824 nucleotides in length);
single-stranded RNAs generated from precursor RNA cleaved by two
RNA polymerase III enzymes DROSHA and DICER to produce mature
miRNA. These miRNAs can control gene expression by targeting
specific mRNAs for degradation and/or translational repression.
They can also control gene expression by recruiting
chromatin-modifying complexes to DNA through binding to DNA
regulatory regions, thereby altering chromatin conformation (33,
34). Expression of miRNA in human blastocysts correlates with
maintenance of pluripotency in embryo development [2].
LincRNAs, a subset of lncRNA, exhibit high conservation across
different species. They have been shown to guide
chromatin-modifying complexes to specific genomic loci, thereby
participating in the establishment of cell typespecific epigenetic
states. In embryonic development, expression of lncRNAs, regulated
by the pluripotent transcription factors OCT4 and NANOG,
facilitates cell lineagespecific gene expression (38). LincRNAs
also play an important role in developmental processes such as
X-chromosome inactivation and genomic imprinting [2].
II. Epigenetics and InfertilityFactors Associated with the
Infertility Phenotype
Several studies have identified different genetic and epigenetic
factors as being involved with the onset and progression of the
infertility phenotype. However, possible perturbed mechanisms
involved in regulating gene expression in the disease state are
rather poorly understood. The classic nature versus nurture
argument on whether ones environment rather than genetic makeup
could influence the onset of infertility and other complex
disorders has been reviewed by various studies [4].
Epigenetic Factors Influencing the Infertility PhenotypeSeveral
histone deacetylases and demethylases have recently been identified
and attributed with regulation of chromatin state. However, their
functional role and association with diseased states is only just
beginning to be understood, particularly those factors affecting
male and female infertility. Post translational modifications of
histone tails by factors including histone chaperones and
methyltransferases are involved in the proper regulation of gene
expression. Due its dynamic nature and plasticity, the landscape of
chromatin can be altered, rendering a region of the genome active
or inactive. This altered state of packaging renders certain
regions of the genome more accessible to transcription machinery
(euchromatin) and are marked by DNA hypomethylation, RNA Pol II,
and covalent histone modifications such as histone 3 trimethylated
at lysine 4 (H3K4me3) and the histone variant H2A.Z.
Inactive/repressed regions are known to be associated with DNA
hypermethylation, histone 3 trimethylated at lysine 27 (H3K27me3),
and SUZ12 (part of the polycomb group complex, PcG) [4].
Due to this, the main focus of recent epigenetic research has
focussed on discovering new factors involved in altering chromatin
state and further looking at its involvement in diseased and normal
tissue. Recent studies have identified a critical role for the
JHMD2A (Jumonji C domain-containing histone demethylase 2A) histone
demthylase in male infertility, obesity, and spermatogenesis. Using
knockout mice as models, these studies identified a critical role
for JHMD2A in the regulation and expression of two genes, protamine
1 (OMIM #182880, Prm1) and transition nuclear protein 1 (OMIM
#190231, Tnp1) involved in the condensation and proper packaging of
chromatin in the male sperm. A higher degree of spermDNA compaction
has previously been attributed to the increased presence of highly
basic protamine proteins compared to histones in chromatin, a
deficiency of which has been associated with infertility in mice.
Identification of other regulatory mechanisms involved in the
recruitment of factors, in addition to JHMD2A, involved in the
deposition of histones along with others affecting transcriptional
activity of genes involved in infertility will increase our
understanding of mechanisms involved in both perturbed and normal
states [4].
a. Epigenetics and Female InfertilityInfertilityThe World Health
Organization defines the term primary infertility as the inability
to bear any children, whether this is the result of the inability
to conceive a child, or the inability to carry a child to full term
after 12 months of unprotected sexual intercourse. Primary
infertility is sometimes known as primary sterility. However, in
many medical studies, the term primary infertility is only used to
describe a situation where a couple is not able to conceive
[5].
Secondary infertility is defined as the inability to have a
second child after a first birth. Secondary infertility has shown
to have a high geographical correlation with primary infertility.
Fecundity describes the ability to conceive after several years of
exposure to risk of pregnancy. Fecundity is often evaluated as the
time necessary for a couple to achieve pregnancy. The World Health
Organization recommends defining fecundity as the ability for a
couple to conceive after two years of attempting to become pregnant
[5].
The terms infertility and infecundity are often confused.
Fertility describes the actual production of live offspring, while
fecundity describes the ability to produce live offspring.
Fecundity cannot be directly measured, though it may be assessed
clinically. Typically, fecundity may be assessed by the time span
between a couples decision to attempt to conceive and a successful
pregnancy [5].
EpigeneticsThe epigenetic mechanism of action suggests that
environmental factors alter how a gene is expressed, but without
directly changing the DNA sequence. Epigenetics is the study of
inherited changes in phenotype (factors that account for
appearance) that are not directly related to, nor explained by
changes in our DNA pattern. For this reason, this field of study is
known as "epi," the greek root for "above," indicating that a
change has occurred that is not directly related to the genetic
code, but above it somehow. In epigenetics, non-genetic causes are
considered responsible for different expressions of phenotypes. Or,
termed in a different way, epigenetics describes changes in the
expression of our genes that are not caused by alterations in the
DNA sequence. Essentially, a different factor accounts for the
change in gene expression [5].
Exogenous, or environmental components may affect gene
regulation and thus, potentially, subsequent expression in the
phenotype. Changes to gene expression induced by environmental
contaminants can be permanent or transient. Research has shown that
epigenetic changes may in fact be reversed [5].
b. Epigenetics and male infertilityAberrant epigenetic
regulation, male infertility and embryonic developmentIt is crucial
that proper regulation of epigenetic processes is maintained
throughout spermatogenesis to not only ensure proper sperm
function, but also proper embryonic development. It has been found
that the sperm epigenetic environment plays a role in establishing
epigenetic marks in the embryo, thus aberrant epigenetic regulation
in spermatogenesis has a profound effect on both male fertility and
embryonic development [6].1. DNA methylationImproper DNA
methylation of various genes has been implicated in abnormal semen
parameters, as well as several instances of male factor
infertility. This aberrant methylation can occur globally or be
limited to one specific locus. A study by Houshdaran et al.
demonstrated that poor sperm concentration, motility and morphology
were associated with broad DNA hypermethylation across a number of
loci. Four of these sequences PAX8, NTF3, SFN and HRAS were single
copy sequences unique to non-imprinted genes. Moreover, the
repetitive element Satellite 2 was also found to be
hypermethy-lated. The authors proposed that hypermethylation of
these loci results from the improper erasure of already established
methylation marks rather than aberrant de novo methylation
following epigenetic reprogramming. The data from this study
suggest that methylation defects present outside of imprinted loci
may be a key factor in some cases of infertility. Recent research
has identified a critical role for the JHMD2A (Jumonji C-terminal
containing histone demethylase 2A) histone demethylase in male
infertility, obesity and spermatogenesis. Studies using knock-out
mice models identified a critical role for JHMD2A in the regulation
and expression of protamine 1 and transition nuclear protein 1,
both of which are critical for DNA condensation during chromosomal
packaging in sperm. The lack of proper DNA packaging in sperm has
been associated with infertility in mice. It is possible that
aberrations of a number of other proteins regulating the activities
of the proteins involved in DNA compac-tion could cause infertility
[6].
Methylenetetrahydrofolate reductase (MTHFR) is a key enzyme in
the folate pathway that catalyzes the reduction of 5,10-
methylenetetrahydrofolate to 5-methylenetetrahydrofolate. The
enzyme maintains bioavailability of methionine so it can be
converted to S-adenosylmethionine, a methyl donor for a variety of
substrates including DNA. To better characterize the role of MTHFR
in spermatogenesis and male fertility, Khazamipour et al.compared
the methylation status of the MTHFR gene promoter in the testes of
men with non-obstructive azoospermia to men with obstructive
azoospermia without defects in spermatogenesis. It was found that
53% of men with non-obstructive azoospermia had hypermethylation of
the MTHFR promoter, whereas none of the men with obstructive
azoospermia exhibited hypermethylation of this region. These
statistically significant data indicate that MTHFR hypermethylation
is a specific epigenetic aberration and may strongly contribute to
certain cases of male infertility. Interestingly, a study by Kelly
et al. demonstrated that adminis-tration of betaine during
pregnancy, nursing and post-weaning can indeed improve testicular
histology and fertility .Wu et al., in a very recent study
concluded that hypermethylation of MTHFR gene promoter in sperm was
associated with idiopathic male infertility. The authors
demonstrated that the number of patients with hypermethylation was
three times to that of control individuals [6].
2. Genome imprintingAppropriate establishment of genomic
imprints is critical to the maintenance of fertility. Indeed, both
paternal and maternal imprinting defects have been identified in
several groups of men experiencing male factor infertility.
Substantial work has been directed towards confirming the presence
of these imprinting abnormalities. A study by Houshdaran et al.
demonstrated that poor semen parameters were linked with increased
DNA methylation at several differentially methylated loci, more
than likely due to a defect in methylation erasure during
epigenetic reprogramming. These loci included PLAG1, a maternally
imprinted gene, DIRAS3 and MEST. An additional study by Kobayashi
et al. examined seven imprinted genes, including the paternally
imprinted GTL2 and H19 loci, in 97 infertile men. They found that
14.4% of the patients studied exhibited abnormal paternal
imprinting and 20.6% exhibited abnormal maternal imprinting. Donors
with normal sperm count and motility were methylated at the H19
locus, whereas one patient with moderate oligospermia and three
patients with severe oligosper-mia showed no methylation of the
locus. Similarly, healthy donors displayed methylation at the GTL2
locus while two patients with moderate oligospermia and four with
severe oligospermia displayed no methylation. Further sequencing of
both loci concurrently showed that one patient with severe
oligospermia exhibited a defect in both the H19 and GTL2 loci.
Moreover, five of the ten patients with severe oligospermia and
three of the eight patients with moderate oligospermia displayed
aberrant methylation of maternally imprinted loci and
differentiallymethylated regions [6].
Poplinski et al. compared the methylation status of the H19/
IGF2 imprinting control region 1 (ICR1) and MEST loci in 33 healthy
donors and 148 men with idiopathic infertility. Normozoospermic men
displayed high methylation at the H19/IGF2 ICR1 locus and low
methylation at the MEST locus, while low methylation of the
H19/IGF2 ICR1locus and high methylation of the MEST locus was found
to be associated with low sperm concentration. The H19/IGF2
ICR1locus of men with idiopathic infertility was 89.6% methylated
with less than 40 million sperm in comparison to 95.9% methylated
in normal donors. As methylation of this locus increased, sperm
count increased linearly. Moreover, decreased methylation of
H19/IGF2 ICR1locus directly correlated with decreased sperm
motility. Data from two different studies by Marques et al. also
indicated that some oligozoospermic patients and secretory
azoospermic patients with hypospermatogenesis exhibit loss of
methylation at H19. Similarly, in 2010 Boissonnas et al. found that
many patients with teratozoospermia and
oligoasthenoteratozoospermia exhibited hypomethylation at variable
CpG islands at the H19 locus. Hypermethylation at MEST was more
strongly linked with poor sperm quality than hypomethylation at
H19/IGF2 ICR1. This hypermethylation of MEST was observed in
samples from infertile men with less than 40% sperm motility and
less than 5% normal sperm morphology. Men with idiopathic
infertility exhibited MEST methylation of 9.6% in comparison to
4.3% in controls. Furthermore, increased methylation of MEST
linearly correlated with decreased sperm motility. These data are
supported by the findings of a study by Marques et al [6].
3. Nuclear protein transitioningThe exchange of protamines for
histones is a crucial step in the process of spermatogenesis,
causing the DNA to be tightly wrapped for efficient transmission of
nuclear material to the oocyte upon fertilization. It is known that
the Prm1 to Prm2 ratio (P1/P2 ratio) is strictly maintained and
regulated. Indeed, deviation from the standard ratio of 0.81.2 has
been shown to lead to infertility. A change in either direction of
this ratio adversely affects semen quality and DNA integrity.
Patients with abnormally depressed or elevated P1/P2 ratios are
characterized by poor sperm concentration, motility and morphology
as well as decreased fertilization capabilities. Studies suggest
that the most common cause of infertility by aberrant protamine
exchange is an increase in the P1/P2 ratio caused by a decrease in
Prm2 levels, although improper regulation of Prm1 levels has also
been implicated in some cases of male infertility. Furthermore, a
study by de Yebra et al. showed that men with a higher P1/P2 ratio
were also more likely to have lower total protamine levels and
higher intermediate protein levels. Moreover, this same study
additionally demonstrated that some infertile men completely lack
Prm2 in their sperm nuclei. There have been no cases reported of
fertile men with severely altered P1/P2 ratios; indeed, it appears
to be a characteristic limited exclusively to certain infertile
males. Interestingly, a link between abnormal protamine
incorporation and aberrant genomic imprinting has recently been
discovered. Hammoud et al. found that infertile males with abnormal
protamines exhibited statistically significant hypermethylation at
the imprinted loci KCNQ1, LIT1 and SNRPN. Moreover, this study also
demonstrated that these patientsshowed hypomethylation of the H19
locus [6].
As discussed previously, hyperacetylation of histone H4 is
required in the transition from histones to protamines. This step
decreases the affinity of the interaction between the sperm
histones and DNA to allow the exchange for transition proteins to
occur. A study by Sonnack et al. showed that men exhibiting
qualitative and/or quantitative infertility have significantly
decreased levels of histone H4 acetylation associated with impaired
spermatogenesis. In the seminiferous tubules of men with round
spermatid maturation arrest, only approximately 60% of spermatids
were immunopositive for this hyperacetylation and many were
multinucleated. Moreover, infertile men withqualitatively normal
spermatogenesis exhibited approximately 90% immunopositive
spermatids and infertile men with qualitatively abnormal
spermatogenesis exhibited approximately 75% immunopositive
spermatids. This contrasts significantly with the almost all
hyperacetylated round spermatids in fertile men. Interestingly,
spermatocytes in the seminiferous tubules of men with round
spermatid maturation arrest exhibited an additional signal,
indicating that early hyperacetylation of histone H4 may lead to
premature nuclear protein transitioning and subsequent infertility
[6].
4. Chromosome structureThe global structure of chromatin is
known to affect gene expression by modulating which regions are
available to be accessed by transcription factors and other
transcriptional proteins and which are not. This feature of
epigenetic regulation becomes particularly important when
considering normally expressed genes that are crucial for proper
spermatogenesis and subsequent oocyte fertilization. It has now
been established that an increase in total heterochromatic variants
is strongly linked to some cases of male factor infertility.
Indeed, it has been demonstrated that there is an increase in the
frequency of chromosomal variants, from 32.55% to 58.68%, in
infertile men compared to controls. The large polymorphic variation
9hq+, located in centromeric heterochromatin on chromosome 9, was
shown to increase in frequency from 4.25% in controls to 14.69% in
men with severe infertility. It is thought that this rise in
heterochromatic regions, not only on chromosome 9 but amongst many
chromosomes, may down-regulate normally active genes. Indeed,
polymorphic variations on the Y chromosome have been implicated in
male infertility based on this reasoning. All the major genes/loci
known to be epigenetically different in some infertile individuals
are listed in Table 2 [6].
III. Epigenetics and PCOS1. miRNALittle is known about the roles
of miRNAs during follicular development, steroidogenesis and in
PCOS. Several studies on miRNA expression have been done on intact
ovaries (chicken, mice, pig, sheep and cattle), as well on the
different ovarian components, such as granulosa cells (mice, pig,
horses and human), theca cells, follicular fluid (humans, cattle
and mares), cumulus cells (mare), cumulus-oocyte complexes (COCs)
(cattle) and corpora lutea (cattle) [7].
The possible modes of action for miRNA within the
pathophysiology of PCOS have only been sparsely investigated, and
thus far, only a few miRNA-PCOS studies exist (see Table 1).
It might be possible that identified PCOS susceptibility genes,
such as DENND1A, which interestingly also encodes miR-601, could
result in genetic and epigenetic factors overlapping, thus
influencing miRNA target specificity. A pilot study investigating
global methylation in twenty PCOS women and 20 BMI- and age-matched
controls, using peripheral leukocyte DNA, showed no significant
differences in the median global DNA methylation percentages.
Despite the negative result, epigenetics may still play a part in
PCOS pathogenesis, since GC and ovary gene-expression could be
tissue specifically epigenetically modified in PCOS. Thus, PCOS is
genetically complex with a large degree of heterogeneity and is
considerably influenced by environmental and genetic cues, one of
these being microRNAs [7].
2. Serum/Plasma miRNA Biomarkers for PCOSPresent abundantly in
serum, miRNAs could serve as a non-invasive biomarker for PCOS, as
they have been shown to be stable in serum, are resistant to
nuclease activity and are easy to detect. It is not known
specifically how miRNAs enter serum or whether the miRNAs present
in serum are disease-specific, since serum is a result of different
components secreted by various tissues and cells, and identifying
their cellular origin can be difficult. Currently, several other
biomarkers in the serum of PCOS women are used for diagnostic
purposes, e.g., luteinizing hormone (LH) and androgen
concentrations, as well as follicle-stimulating hormone (FSH)
[7].
A recent case-control study investigating 12 PCOS patients, 12
healthy females and 12 male controls, subdivided further based on
BMI levels, revealed that obesity significantly reduced the
expression of four miRNAs selected for evaluation in whole blood:
miR-21, miR-27b, miR-103 and miR-155 in control women and men, but
tending to show an increase in expression in PCOS women. Further
analysis of their hormone profile showed a positive correlation
between serum free testosterone levels and miR-21, miR-27b and
miR-155. Perhaps, the elevated free testosterone found in the PCOS
samples could partly explain the observed increase of these miRNAs.
Further, bioinformatics analysis and target gene analysis revealed
that miR-21, miR-27b, miR-103 and miR-155 could be involved in
hormone metabolism, as well as reproductive cellular processes
[7].
Using miRNA arrays, the expression of serum miRNAs in patients
with PCOS compared to age-matched controls has been evaluated .
Following an initial miRNA profiling based on a relative two-fold
change in expression levels, nine miRNAs (miR-222, miR-16, miR-19a,
miR-106b, miR-30c, miR-146a, miR-24, miR-186 and miR-320) were
chosen for further analysis. The expression levels for eight of the
miRNAs were upregulated in serum from PCOS patients, whereas
miR-320 displayed decreased expression in the PCOS subjects.
However, following Q-PCR validation of the nine miRNAs expression
in the entire study population (n = 68 PCOS, n = 68 controls), only
miR-222, miR-146a and miR-30c remained significantly increased in
the PCOS patients. Sensitivity and specificity analysis, using
receiver operating characteristic (ROC) curves and area under the
curve (AUC), revealed that a combination of the three miRNAs was
able to distinguish between the PCOS and controls. In addition,
correlation analysis adjusted for age and BMI showed that miR-222
strongly correlated positively with serum insulin levels in PCOS
women. Interestingly, upregulated expression levels of miR-222 have
also been associated with type 2 diabetes and gestational diabetes
mellitus. Further, miR-146a correlated negatively with serum
testosterone in PCOS women. Decreased miR-146 has been linked to
inflammation and insulin resistance in T2D individuals. An
interesting observation made by Long et al. was that most of the
miRNAs present and differentially expressed in ovarian tissue from
PCOS women were not released into the blood and, therefore, were
not altered in PCOS serum [7].
In conclusion, identification of distinct miRNAs present within
the circulation would prove a useful tool for diagnosis and perhaps
treatment of PCOS. Comparing the miRNA profile of PCOS patients to
healthy controls reveals that miRNAs might contribute to the
pathogenesis. Indeed, miR-21, miR-27b and miR-103 are associated
with PCOS, as well as metabolic features, such as obesity, T2D,
low-grade inflammation and adipogenesis dysfunction. Furthermore,
insulin sensitivity and the suppression of androgens have been
associated with miR-222 and miR-146a, respectively. Profiling of
serum miRNAs does not necessarily reflect the more local changes
within the ovary, and the functional role and significance of
miRNAs in blood from PCOS patients still need to be determined
[7].
3. MicroRNAs as Biomarkers for PCOS Based on Follicular Fluid
ContentComparing the miRNAs found in follicular fluid to the miRNAs
found in the bloodreveals the common occurrence of miR-186, miR-21,
miR-155, miR-103, miR-19a and miR-16, although with different
expressions levels and significance associated with PCOS. Moreover,
the miRNA profile of follicular fluid varies between studies,
highlighting that PCOS is a complex and heterogenic syndrome.
Interestingly though, increased expression of miR-146 was found in
serum from PCOS patients and also in follicular fluid by Roth et
al. and Sang et al., respectively. Adding to this, miR-222 and
miR-24 were also found to be highly expressed in follicular fluid,
as well as identified in PCOS serum. A differential expression,
albeit in different directions, was observed for miR-320. Taken
together, this still warrants further studies [7].
4. Possible Role for miRNA in the Abnormal Follicular
Development and Function in PCOSMany different theories have been
brought forth in an attempt to explain the mechanisms responsible
for the impaired ovulation, abnormal follicular development and
excessive follicle formation commonly found in women with PCOS, but
with varying results. An altered appearance and function of
granulosa cells (GCs) with respect to FSH, LH and androgens has
been proposed. Defects in steroidogenesis by the theca cells (TCs)
and increased activation of primordial follicles, abnormal
expression of anti-Mllerian hormone, increased follicle survival
and/or a decreased apoptosis rate have also been reported. Many of
the factors involved in these processes are still unknown and the
mechanisms unestablished [7].
Taken together, altered expression of ovarian miRNAs might play
a role in the processes determining the fate of granulosa cells
(proliferation and differentiation vs. apoptosis), and this might
lead to the hyper-proliferating granulosa cells, as seen in PCOS.
The roles and mechanisms of miR-224, miR-320 and miR-383 in GCs
during folliculogenesis in general and in PCOS remain unknown
[7].
IV. Epigenetics and Endometriosis
1. DNA methylationRecently it has been shown that DNMT1, DNMT3a
and DNMT3b are over expressed in endometriotic tissue. These
findings are likely to provoke new ideas regarding the origin and
aetiology of endometriosis. For example, over-expression of these
enzymes would be expected to alter global DNA expression in
endometriotic cells. DNA microarray analysis of endometriotic
tissue supports this expectation since a substantial number of
genes display significantly altered expression patterns. Another
possible explanation for these observed irregularities in gene
expression may be due to abnormalities in the regulation and
function of the major transcription factor, NF-B, in endometriosis
(for review see Guo 2007 ). While this may be so, the introduction
of epigenetics into the fray offers new insights into the origin
and progression of disease, such as the combined effect of
disrupting traditional and epigenetic regulators of transcription.
In support of this notion Wren et al demonstrated that epigenetic
mechanisms such as histone modifications, methylation and
acetylation may play a role in the aetiology of endometriosis.
However, results from microarray studies provide an unusual
paradox. The majority of studies found almost the same number of
down regulated genes as there were up regulated in ectopic
endometrial tissue. For example, Kao et al reported 91 genes
significantly over expressed and 115 under expressed. Similarly
Eyster et al 2002 and Eyster et al 2007 reported more genes over
expressed in ectopic endometrial tissue. Yet enhancement of DNMT
function in endometriosis should lead to increased levels of DNA
methylation hence, increasing the number of silenced or down
regulated genes. This raises the question as to how a system can be
in place where global gene expression in endometriotic cells should
be down-regulated by over active methylation, and yet the evidence
clearly shows many genes are up-regulated. However, it should be
noted that Burney et al reported a higher frequency of under
expressed genes in the eutopic endometrium of women with
endometriosis vs disease free controls. A possible explanation to
this paradox will be discussed in section 3.6. Of course it must be
considered that not all genes are epigenetically regulated, genetic
mechanisms are likely to play a significant role in aberrant gene
expression in endometriosis [8].
2. Epigenetic Modification of Steroid Synthesis and Receptors in
EndometriosisThe deregulation of DNMTs is not the only evidence
supporting the hypothesis that epigenetics plays a major role in
endometriosis. Izawa et al demonstrated that the expression of the
cytochrome p450 aromatase enzyme (CYP19) is dependent on the
methylation status of its promoter by treating endometriotic cells
with the demthylating agent 5-aza-deoxycytidine and observing the
fold change in aromatase mRNA expression. Current studies have
reported either a weak or no association between polymorphisms of
the aromatase gene and endometriosis that could account for the
observed over expression of aromatase in endometriosis. Those
studies that have associated certain aromatase polymorphisms with
endometriosis have been criticised for faulty data analysis or non
reproducible results [8].Aromatase is a key enzyme involved in the
synthesis of estrogen and plays a crucial role in the pathogenesis
of endometriosis. With the exception of two studies aromatase is
reported to be highly up regulated in endometriotic cells whilst
being nearly undetectable in normal endometrium. The importance of
aromatase in the pathology of endometriosis is aptly demonstrated
by the use of aromatase suppressing drugs for the treatment of the
disease. This class of drugs, although with limited clinical data,
have shown to be effective in the symptomatic treatment of
endometriosis. Aromatase is normally expressed in a cyclic fashion
throughout the menstrual cycle in eutopic endometrium however,
expression levels are consistently elevated in endometriotic cells.
If the over-expression is initiated by hypomethylation of the
promoter and maintained by aromatase activating cytokines such as
IL-6, IL-11 and TNF, all of which have been shown to be
dysregulated in endometriosis , a consequence would be
over-expression of aromatase leading to increased synthesis of
estrone, which is converted to estradiol, a potent estrogenic
factor that initiates a number of pathways leading to the
proliferation and survival of endometriotic cells. The conversion
of estrone to estradiol is catalysed by the enzyme
17-hydroxysteroid dehydrogenase type 1 (17HSD I), which is
reportedly up regulated in endometriosis. The increased activation
of aromatase in endometriotic cells leads to a self sustaining
positive feedback loop for estradiol production, whereby
prostaglandin E2 (PGE2) activity induces the up regulation of
aromatase leading to increased estradiol levels. In turn, this
leads to the up regulation of the cycloxygenase-2 enzyme (COX-2)
resulting in the formation of more PGE2, an important factor in the
pathology of endometriosis thus the cycle becomes self perpetuating
(Figure 25) [8].
Aberrant methylation of the aromatase promoter is not the only
factor altering gene expression due to epigenetic alterations in
endometriosis. Regulation of aromatase is mediated by steroidogenic
factor-1 (SF-1) which is the aromatase enhancer, and chicken
ovalbumin upstream promoter transcription factor (COUP-TF) which is
its repressor. Indeed, SF-1 has recently been shown to be
over-expressed in endometriotic cells. An explanation for the
apparent over-expression of SF-1 has been proposed by Xue et al who
observed hypomethylation of the CpG island near to its promoter
region. The discovery of epigenetic modifications in the promoters
of aromatase and its enhancer SF-1 provide some understanding of
the establishment of the reported positive feedback loop. As with
mutations, once these epimutations are established, they are
retained throughout each cellular division, ensuring the survival
of the endometriotic cells [8].
It is not only the synthesis of estrogen that is affected by
aberrant methylation in endometriosis. In order for estrogen to
mediate its mitogenic effects within the cell it must first bind to
its receptor, of which there are two variants, estrogen receptor
(ERA) and estrogen receptor (ERB) coded for by separate genes.
Cells that over express estrogen receptors are highly sensitive to
estrogenic effects. Such cell types include breast and ovarian
cancer cells which, as with endometriotic cells, possess an
enhanced proliferative capacity. Recent studies have shown that the
mRNA of one isoforms of the estrogen receptor (estrogen receptor 2
gene, encoding estrogen receptor ) is over expressed in
endometriosis. This apparent over-expression was found to result
from hypomethylation of the CpG islands in the promoter of the ESR2
gene. ERB is important as it is known to regulate several genes
involved in signal transduction, cell cycle progression and
apoptosis, however in contrast to endometriosis several studies
have shown ERB to be down regulated in ovarian cancer and it
thought that loss of ERB expression may induce malignant
transformation. It is also important to note that several endocrine
disrupters though to be risk factors for endometriosis mediate
signalling cascades via ERB. Therefore, not only does epigenetic
modification lead to enhanced estrogen production but it also leads
to increased sensitivity towards estrogen and estrogen-like
compounds in endometriotic cells, resulting in a self sustaining
endometriotic cell population [8].
Due to abnormal estrogen synthesis and metabolism observed in
endometriosis, progestogenic agents are commonly administered to
women with the disease in order to suppress endometriotic cellular
proliferation by down regulating estrogen production and acting as
an anti-inflammatory agent. The efficacy of progestogens in
relieving persistent pain symptoms associated with endometriosis is
relatively poor, and a reported 9% of women are completely
unresponsive to progestogen treatment. The relative inefficiency or
total lack of response to progestogenic treatment has thus, led
some to conclude that endometriotic cells are somehow resistant to
the effects of progesterone. Evidence for this comes from studies
of the progesterone receptors in endometriotic cells. As with ER,
there are two progesterone receptor isoforms, PR-A and PR-B. Unlike
the ER, these encode as splice variants of the same gene and each
has distinct functions and distinct levels of expression in the
eutopic endometrium, depending on the phase of the menstrual cycle.
PR-B is a transcriptional activator for several genes containing a
PR-B dependent promoter. PR-A, on the other hand is a
transcriptional repressor for PR-B and ER. Studies have shown that
PR-B expression is absent, and only very low levels of PR-A are
expressed in endometriotic cells, offering some explanation for
progesterone resistance in endometriosis. Additionally, Wu et al
showed that the aberrant hypermethylation of the PR-B promoter,
reduces its expression to an almost silenced state. PR-A and PR-B,
although coded by the same gene, have distinct promoters, thus it
may be possible that aberrant methylation of the PR-A promoter may
be responsible for its reduced expression. Conversely Wu et al
suggested that alteration of PR-A expression may not be due to
altered methylation of its promoter, but is likely due to other as
yet, unknown mechanisms. Nevertheless, it is reasonable to conclude
that there are epigenetic mechanisms by which estrogen production
is both enhanced and unopposed in endometriosis (Figure 26)
[8].
3. HOXA10 in EndometriosisHOX are a family of genes containing
homeobox domains that act as transcription factors essential for
regulating genes associated embryonic development. Several members
of the HOX gene family play crucial roles during embryogenesis, for
example HOXA9, HOXA10, HOXA11 and HOXA13 are involved with the
development of the female reproductive tract, and unlike the
majority of HOX genes they are expressed into adulthood. The
involvement of HOXA10 in the development of the uterus affords
specific interest with regards to endometriosis since any aberrant
expression of HOXA10 may result in abnormalities, in either the
function or morphology of the uterus. HOXA10 expression is
reportedly down regulated in patients with endometriosis, perhaps
reflecting the findings that patients with endometriosis are more
likely to present with anatomical complications of the reproductive
tract. HOXA10 under-expression in endometriosis patients may also
explain the associated subfertility observed in these patients
since HOXA10 along with HOXA11 are responsible for successful
implantation of the embryo [8].
The origin of the down regulation of the HOXA10 gene in
endometriosis was investigated by Wu et al 2005 and Kim et al 2007.
Wu et als study screened eutopic endometrium of women with
endometriosis, whereas Kim et al screened eutopic endometrium from
baboons with experimentally induced endometriosis. Both studies
identified hypermethylation in the promoter region of the HOXA10
gene. However, these studies examined only methylation patterns of
HOXA10 in the eutopic endometrium of endometriosis cases and
controls, but did not examine the methylation status of HOXA10 in
ectopic endometrium. Wu et als study was also confined only to
women with stage III-IV endometriosis. Therefore, definite
conclusions regarding the aberrant expression of HOXA10 in
endometriotic tissue in humans cannot be made. However, a study by
Lee et al 2009 using a murine model of experimentally induced
endometriosis, found inducing endometriosis led to methylation
dependant changes in HOXA10 expression in eutopic endometrium. This
essentially turns current thinking on its head, as it has long been
thought eutopic endometrium dictates the fate and function of
ectopic endometrium (via polyclonal origin of ectopic endometrium
from refluxed eutopic cells) not, as Lee et al demonstrated, the
other way around. Although it may be that interplay of signalling
exists between the two cell types, the mechanism by which ectopic
endometrium can influence epigenetic alteration of eutopic
endometrium remains to be elucidated [8].
Interestingly, further study has reported that HOXA10
hypomethylation can be induced by in-utero exposure to
diethylstilbestrol (DES), a known endocrine disruptor. The
ramifications of DES exposure are discussed further in section 3.5.
The effect of altered HOXA10 expression in eutopic endometrium is
obvious in terms of uterine morphology and embryonic implantation.
What the biological significance of HOXA10 down regulation, in
endometriotic tissue may be remains speculative. Some clues may
arise from microarray study of HOXA10 knockdown cells, which
reported HOXA10 as a regulator of hundreds of genes involved in a
variety of cellular processes. Of particular interest was the
finding that HOXA10 knockdown led to a 5.78 fold increase in the
CYP19 (aromatase) gene, the significance of which is discussed in
section 3.2. It is also important to note that HOXA10 and HOXA11
are progesterone responsive genes and members of the HOX family
themselves regulate a number of other genes including IGFBP-1 and
integrins [8].
Nevertheless, the altered expression of HOXA10 in women with
endometriosis may provide an explanation for one of the most
puzzling aspects of endometriosis, which is, if retrograde
menstruation is near universal, why do only a fraction of women
develop endometriosis? It may be that HOXA10 aberrations result in
improper development of the uterus. This is supported by the
observed uterine anatomical abnormalities in women with
endometriosis such as increased frequencies of septate uterus and
the reported decreased elasticity of the reproductive organs. Both
aspects are thought to increase the volume of menstrual reflux in
some women, overwhelming the immune system which is thus unable to
remove all the refluxed endometrial cells [8].
4. Epigenetics and the EnvironmentEpigenetics provides a link
between genotype and the environment, and how exposures to
different environmental, pharmacological and dietary elements can
translate into heritable changes in gene expression. During early
mammalian development the methylome is stripped and then re-applied
in order to start development from a blank state in which
methylation errors are removed, for this reason it was thought that
alteration of epigenetic marks such as methylation patterns could
not effect subsequent generations. If epigenetic marks were
heritable then the complex phenotypic consequences they encode,
which may include disease phenotypes, must be heritable also. DNA
methylation by DNMTs is dependent on methyl donors such as
S-adenosylmethionine (SAM), the major methyl donor, which is
synthesised as part of the methionine cycle. The formation of this
cycle is, in turn, dependent on dietary factors such as folic acid,
vitamin B12, choline and betaine. Animal studies have shown that
restricting dietary methyl donors produces a reduction in DNA
methylation, and in some cases, increased risk of developing
tumours, indicating that epigenetic aberrations mediated by dietary
changes can result in complex disease phenotypes. The sparse
epidemiological studies which have reported on the influence of
diet and endometriosis suggest that a diet high in fruit and green
vegetables and low in meat and alcohol consumption is protective
against developing endometriosis, however some findings were
inconsistent. From an epigenetic perspective a diet high in fruit
and green vegetables would provide a significant source of methyl
donors, which may protect against demethylation induced genetic
instability during foetal development. Alcohol consumption has been
shown to alter histone acetylation and methylation patterns which
may account for the observed estrogenic effect of alcohol [8].
V. Epigenetics and Fibroids
1. DNA MethylationDeoxyribonucleci acid methylation that occurs
at the C5 position of cytosine, resulting in 5-methylcytosine
(5mC), mostly within CpG dinucleotides , is involved in various
developmental processes by silencing, switching, and stabilizing
genes. Deoxyribonucleci acid hypomethylation and imbalanced
expression of DNA methyltransferases (DNMT1, DNMT3A, and DNMT3B)
are found in human uterine leiomyoma compared with the adjacent
myometria. In addition, the aberrant DNA hypomethylation is found
in the distal promoter region of ER-a (21188 to 2790) in the
uterine leiomyoma compared with the myometrium. A restriction
landmark genomic scanning profile study found 29 aberrant
methylation spots(10 methylated and 19 demethylated) in leiomyoma
compared with myometrium. This study also found that DNMT1 and
DNMT3A mRNA expression levels are higher in leiomyoma compared with
myometrium. Recently, Maekaya et al identified 14 hypomethylated
genes (FAM9A, CPXCR1, CXORF45, TAF1, NXF5, VBP1, GABRE, DBX53,
FHL1, BRCC3, DMD, GJB1, AP1S2, and PCD11X) and one hypermethylated
gene (HDAC8) located on the X chromosome in uterine leiomyomas.
Most recently, Navarro et al found 55 genes with differential
promoter methylation and concominant differences in mRNA expression
in uterine leiomyoma vs. normal myometrium. Eighty percent of the
identified genes showed an inverse relationship between DNA
methylation status and mRNA expression in uterine leiomyoma, and
the majority of genes (62%) displayed hypermethylation associated
with gene silencing. Interestingly, they found three known tumor
suppressors genesKLF11, DLEC1, and KRT19withhypermethylation, mRNA
repression, and protein expression in leiomyoma. These results
indicate a possible functional role of promoter DNA
methylation-mediated gene silencing in the pathogenesis of uterine
leiomyoma [10].
2. Histone ModificationHistone modification is the second most
important epigenetic factor that has a critical role in regulation
of gene expression. Histones proteins can be modified in many ways
in their N-terminal tail, including acetylation, phosphorylation,
methylation, ubiquitylation, sumoylation, and adenosine diphosphate
(ADP) ribosylation, deamination, and proline isomerization. Histone
deacetylase 6 (HDAC6) is a regulatory factor in the endocrine
traffic network and possesses histone deacetylase activity and
represses transcription. Wei et al examined the HDAC6 expressions
and its pathogenic role in the uterine leiomyoma. They found a
regular pattern of increasing HDAC6 and ER-a expression in
leiomyoma samples.
It is well known that epigenetic modifications are acquired
during development and play a central role in cellular
differentiation and normal tissue and organ function in adulthood.
However, during crucial stages of development, environmental
exposures can alter genome state related to differentiation
programming of cells or organs, thus promoting disease
susceptibility in later life. A recent study reported that through
nongenomic effects on developing uterus during developmental
reprogramming, environmental Es recruit the epigenetic regulator
EZH2 and reduce the levels of repressive histone mark H3K27me3 in
chromatin and promote uterine tumorigenesis [10].
3. MicroRNAMicroRNAs are a novel class of small
nonprotein-coding RNAs that regulate a high number of biological
processes by targeting mRNAs for cleavage or translational
repression. Studies have shown that several miRNAs, including let7,
miR-21, miR-93, miR-106b, and miR-200 and their predicted target
genes, are significantly dysregulated in uterine leiomyoma compared
with normal myometrium. Additionally, miRNA expression seems to be
strongly associated with tumor size and race. Pan et al reported
that miR-21 is overexpressed in leiomyomas, with specific elevation
during the secretory phase of the menstrual cycle in women who
received depot-medroxyprogesterone acetate and oral contraceptives,
but decreased owing to gonadotropin releasing hormone agonists
(GnRHa) therapy. Recently Zavadil et al examined global correlation
patterns between altered miRNA expression and the predicted target
genes in uterine leiomyomas and matched myometria. They found that
numbers of dysregulated miRNAs are inversely correlated with their
targets at the protein level. Patterns of inverse association of
miRNA with mRNA expression in uterine leiomyomas revealed an
involvement of multiple candidate pathways, including extensive
transcriptional reprogramming, cell proliferation control,
mitogen-activated protein kinase (MAPK), transforming growth factor
(TGF)-b, WNT, Janus kinase/signal transducers and activators of
transcription signaling, remodeling of cell adhesion, and cellcell
and cellmatrix contacts. More recently, Fitzgerald et al reported
that elevated leiomyoma miR-21 levels are predicted to decrease
programmed cell death 4 (PDCD-4) levels, thus leiomyomas differ
from other tumors in which loss of PDCD-4 is associated with tumor
progression [10].
Reference
1. Sarah C. P. Williams. Epigenetics. PNAS. 2013 February;110
(9): 32092. Michal Inbar-Feigenberg, M.D.,Sanaa Choufani, Ph.D,
Darci T. Butcher, Ph.D., Maian Roifman, M.D., and Rosanna Weksberg,
M.D, Ph.D. Basic concepts of epigenetics. Fertility and Sterility.
2013 March;99: 6073. Varija N Budhavarapu, Myrriah Chavez and
Jessica K Tyler. How is epigenetic information maintained through
DNA replication?. Epigenetics & Chromatin 2013, 6:324.
SheroyMinocherhomji,1 Prochi F. Madon,2 and Firuza R. Parikh2.
Review Article: Epigenetic Regulatory Mechanisms Associated with
Infertility. Hindawi Publishing Corporation ,Obstetrics and
Gynecology International Volume 2010, Article ID 198709, 7 pages
doi:10.1155/2010/1987095. FEMALE REPRODUCTIVE HEALTH AND THE
ENVIRONMENT. Training Module 2 Children's Environmental Health..
Public Health and the Environment.World Health Organization.
www.who.int/ceh6. Singh Rajender , Kelsey Avery , Ashok Agarwal ,
Review :Epigenetics, spermatogenesis and male infertility.
Elseveir: Mutation Research. 2011;727: 6271.7. Anja Elaine Srensen
, Marie Louise Wissing , Sofia Sal , Anne Lis Mikkelsen Englund and
Louise Torp Dalgaard. Review :MicroRNAs Related to Polycystic Ovary
Syndrome (PCOS). Genes 2014, 5, 684-708; doi:10.3390/genes50306848.
Matthew David Rosser. The Emerging Role of Epigenetics in the
Aetiology of Endometriosis. De Monfort University Leiceister.9.
Serdar E. Bulun, M.D. Review Article: Mechanism Disease Uterine
Fibroids. NEnglJMed2013;369:1344-55. DOI:10.1056/NEJMra120999310.
Md Soriful Islam, Ph.D., Olga Protic, M.Sc., Piergiorgio Stortoni,
M.D., Gianluca Grechi, M.D. ,Pasquale Lamanna, M.D., Felice
Petraglia, M.D., Mario Castellucci, M.D., Ph.D., and Pasquapina
Ciarmela, Ph.D. Complex networks ofmultiple factors in the
pathogenesis of uterine leiomyoma. Original Articles : Gynecology
and Menopause.