DNA methylation, a well-characterised epigenetic modification, is a heritable covalent modification and is binary in nature. Most methylation occurs at the number five carbon of the cytosine pyrimidine ring (5-methyl-cytosine or 5 mC) and represents less than 5% of all cytosines in our genomes. Genomic methylation patterns are propagated during cell division by DNA methyl-transferases (DNMT1, DNMT3A/B), which catalyse the transfer of a methyl group to DNA [2]. Five DNMT enzymes—DNMT1, DNMT2, DNMT3A, DNMT3B, and DNMT3L—actively regulate two different processes, that is, maintenance and de novo methylation activities [3]. One of the important sites for DNA methylation is CpG-enriched regions associated with promoters called "CpG islands [4,5]." The majority of these sites are located in the promoter region and first exon of genes. DNMTs, along with other enzymes, can orchestrate gene silencing and maintain a repressive chromatin state. Once methylated, proteins such as MeCP2, MBD1, MBD2, and MBD4, which have a methyl-binding domain (MBD), bind to the DNA, which further impedes recruitment of transcription factors to DNA, leading to abrogated gene expression [6]. Hypermethylation of DNA can also recruit histone deacetylase (HDAC), leading to inevitable alterations in gene expression [7,8]. DNA demethylation, a related event that is equally important, converts methyl-cytosine into cytosine either actively or passively. The replication-independent active mechanism uses the ten-eleven translocation (TET) enzyme family (TET1, TET2, and TET3) to catalyse hydroxylation of 5mC followed by activation-induced cytidine deamination [9,10]. On the other hand, replication-dependent passive demethylation occurs during inadequate availability of global methyl donor S-adenosyl-L-methionine (SAM) or DNMTs-mediated demethylation in the presence of Ca²⁺ ions and reducing surroundings [11]. Thus, promoter hypermethylation causes gene silencing whereas promoter demethylation results in gene expression.

Histones are the globular proteins that undergo posttranslational modification and alter regulation of gene expression, and DNA replication, recombination, and repair. While nucleosomes represent the primary step in the construction of higher-order chromatin structures [12], histones have a protruding charged 15-38 amino acid N-terminus ("histone tail") that influences nucleosome assembly into higher order chromatin structures. In its condensed state, chromatin remains in a folded configuration so that the nucleosomes are stacked, and hence, not readily accessible to gene activation. However, covalent modifications such as acetylation, methylation, phosphorylation, poly-ADP ribosylation, and ubiquitination on the long tails of histones alter histones-DNA interaction and higher order chromatin folding. These post-translational covalent modifications regulate the contact between the octamer core and DNA, and determine DNA accessibility to transcription factor complexes. The capability to accumulate information appears to dwell in the amino-terminal tails of the four core histones, which are exposed on the nucleosome surface and are subject to enzyme-catalysed post-translational modifications of select amino acids, including lysine acetylation, lysine and arginine methylation, serine or threonine phosphorylation, lysine ubiquitination, lysine sumoylation, or glutamine ADP ribosylation (Table 1). Epigenetic modification of the histone tail plays a key role in transcriptional regulation, DNA repair, DNA replication, alternative splicing, and chromosome condensation. With reference to its transcriptional state, the human genome can be approximately compartmentalised into actively transcribed euchromatin and transcriptionally inactive heterochromatin [13]. Euchromatin is characterised by high levels of acetylation and trimethylated H3K4, H3K36, and H3K79. On the other hand, heterochromatin is categorized by low levels of acetylation and elevated levels of H3K9, H3K27, and H4K20 methylation [14]. Interestingly, such dynamic modifications are actively added or removed by different histone-modifying enzymes (writers and erasers, respectively) that catalyse modification of specific amino acids with definite modifying groups in a site-specific manner to manage transcriptional control (Table 1). With specific modification, histones regulate the structural organization of chromatin by altering the electrostatic charge provided by the substituted group and facilitating recognition sites for different adaptor proteins (readers) such as proteins containing a bromodomain, which binds to the acetylated lysine inscriptions.

MicroRNAs (miRNA) are single-stranded RNAs approximately 21–23 nucleotides in length that are transcribed from DNA but not translated into proteins. These non-coding RNAs (ncRNAs) regulate several molecular pathways at the transcriptional or post-transcriptional level. miRNA genes primarily reside between genes (intergenic) or within introns (intronic) of genes and are transcribed to a primary miRNA (pri-miRNA) mediated by polymerase II or III [15]. The pri-miRNA is processed within the nuclear compartment to a precursor miRNA (pre-miRNA) by Drosha, a class 2 RNase III enzyme. Subsequently, the transport of pre-miRNAs to the cytoplasm is arbitrated by exportin-5. In the cytoplasmic region, they are processed further to develop into mature miRNAs by Dicer, an RNase III type protein, into 21–25 nucleotide double-stranded RNAs (dsRNAs) and loaded onto the Argonaute (Argo) protein to generate the effector RNA-induced silencing complex (RISC) [16]. RISC binds to mRNA, forming a RISC-mRNA complex on which miRNAs mediate sequence-specific recognition and binding, which further degrades or silences the target mRNA depending on the sequence complementarity. While the majority of researchers believe that miRNAs restrain translation, evidence that miRNAs can actually augment translation through alterations in the Argo component of the RISC has also been reported. Thus, while miRNAs appear to police translation in an inhibitory fashion, they may also boost translation in defined biological settings.

DNA packaging in nucleosomes might affect all stages of transcription, thereby regulating gene expression. The precise position of nucleosomes around the transcription start sites has an essential degree of control over the initiation of transcription [17]. Nucleosome positioning not only decides the accessibility of the transcription factors to their target DNA sequence but has also been reported to take part in shaping the methylation landscape. Besides transcription regulation, nucleosome occupancy also participates in directing meiotic recombination events. The precise function of nucleosomes is influenced by the incorporation of different histone variants that are incorporated into chromatin independently, outside the S-phase. Often linked with specific histone modifications, nucleosome remodelling machinery is also influenced by DNA methylation. Thus, the interaction among diverse epigenetic partners is often evident.

Male germ cells have compact nuclear DNA which are required for the transmission of the paternal genome to the oocyte. The highly compacted paternal DNA residing in the sperm head goes through extensive remodelling [20]. The process of spermatogenesis involves condensation of chromatin in the spermatid head prior to conversion of spermatids to spermatozoa [21]. The spermatids have a less compact genome in the early stages of spermatogenesis that is further compacted in the sperm genome by the substitution of histones by non-histone proteins. In the process of histone replacement, they are first replaced by transition proteins (TNP1 and TNP2) and eventually by protamines (P1 and P2). This allows the protamines-bound sperm genome structure to compact 6 to 20 times more than the histone-bound nuclear structure that makes it transcriptionally and translationally inert. Human spermatozoa are acknowledged to enclose a huge range of RNA molecules, including over 100 types of miRNA [22]. On the other hand, the mature sperm cells enclose only the mRNA and small RNAs that were present in the spermatids at the early phases of spermatogenesis. It was previously assumed that there is no participation of sperm transcriptome in embryo development but several recent studies have described the involvement of sperm genome organization as well as paternal transcriptome in early embryonic development. Moreover, chromatin of mature spermatozoa retains small amounts of histones, which is crucial in the differentiation and early development of the zygote. Surprisingly, the male pronucleus also displays elevated levels of histone acetylation, which supports higher transcription from the S phase in the zygote stage and later. Demethylation in the PGCs results in activation of genes: deleted in azoospermia-like (Dazl) and synaptonemal complex protein 3 (Sycp3), that are essential for gametogenesis [23]. This also regulates expression of testis-expressed (Tex19.1, Tex19.2) and Mili (also called Piwi-like 2) genes required to suppress any unmasked retrotransposon action and for a stable genome integrity [24]. miRNAs also regulate the transcriptional silencing in compact chromatin-containing elongating spermatids [25]. Dicer activity has been found to be crucial during male germ cell differentiation, as reported in Dicer knockout studies, which shows that early Dicer deletion results in damage accumulation and compromised spermatogenesis [26,27]. Increasing evidence supports the vital role played by small RNA-mediated RNA regulation in normal spermatogenesis and can affect male fertility if it fluctuates from the normal germline.

The maternal DNA contained in the oocyte is bound by histones acquired during oocyte growth comprising post-translational modifications associated with stalled metaphase II. The major difference between the chromatin of oocyte and somatic nuclei is the deficiency of H1 linker histones in oocytes, which are replaced with a specific H1 variant whose role in early embryogenesis remains to be understood. Histone methylations play important roles in the regulation of chromatin structure and gene expression during follicle maturation, especially for oocyte meiotic maturation. Euchromatin histone-lysine N methyl-transferase 2 (EHMT2) methylate H3K9me1 and H3K9me2 are crucial for early meiotic progression. It has been established that H3K9me3 appears in growing oocytes from early antral follicles, increases subsequently during the growth phase, and is retained during oocyte meiotic maturation and activation [28,29]. However, at different developmental stages, sensitivity toward methylation of a diversity of histones including H3K4me, H3K4me2, and H3K4me3 has been found in granulosa cells of follicles as well as in oocytes from primary to antral-stage follicles [30]. In addition, histone ubiquitination may be another regulatory factor during follicle development. Ubiquitinated histone H2A is coupled with transcriptional silencing of large chromatin areas during meiosis in females. The production of gametes might require orderly and proper epigenetic reprogramming in premigratory and migratory germ cells for an appropriate epigenotype to support subsequent normal development. Genomic imprinting in the oocytes of females occurs after birth and is arrested at the diplotene stage of meiotic prophase I, and is then completed by the de novo methylation process in the fully-grown oocyte stage.

Results of recent work have significantly improved our understanding of the sperm epigenome and its probable role in embryonic development. These new outcomes have facilitated a broad definition of a normal epigenetic state in the male gamete while also providing insight into the potential aetiologies of various idiopathic male infertility cases [57]. Emerging studies have shown notable associations of aberrant DNA methylation in spermatozoa with idiopathic male infertility as well as increased frequency of spontaneous abortions and imprinting disorders [58]. In addition, the epigenetic errors in the process of spermatogenesis in humans have been acknowledged as causes for reduced sperm competency and reduced fertility in males. The fundamental causes of infertility in males may be seminal oxidative stress, double- and single-stranded DNA damage of both the nuclear and mitochondrial genome of sperm, sudden telomere attrition and aberrant epigenetic alterations that can not only affect a single person but also their descendants [59]. The relationships between DNA methylation levels and male fertility in humans have been investigated in several studies. The methylation load and defects of the sperm DNA have been confirmed as a cause of infertility. Emerging evidence suggest that miRNAs play an important role in male fertility. Specifically, miRNA knockout of Dicer in the sertoli cells has been shown to result in testicular dysgenesis and infertility due to alterations in sertoli cell architecture [60]. A concerted effort to identify the roles of particular miRNAs during spermatogenesis should help determine whether miRNAs can serve as important predictive or diagnostic indicators and/or a system for treating male infertility. Epimutations could lead to male infertility, as suggested in the literature. Often hypermethylation has been found to be associated with poor semen parameters or male infertility. Evaluating the implications of lifestyle habits, food intake and environmental factors (e.g., drugs and toxins) on the sperm epigenome and the subsequent effect on fertility outcomes is needed. This should be of help to better understand epigenome regulation in the male germline, and may also suggest new strategies for this line of research.

In females, epigenetic changes in the oocytes can be affected by factors such as maternal nutrition and exposure to certain toxins that have been linked with neonatal developmental and gestational defects [61]. Perturbations in epigenetic mechanisms may further elucidate the intrinsic causes of infertility or state of fertility or ovarian tumorigenesis [62]. Epigenetics in women with infertility is more complex and less explored; the factors associated with epigenetic abnormalities are diverse. These factors include ovulation problems (more common due to polycystic ovary syndrome [PCOS]), endometriosis, inflammatory disorders, age-related factors, and nutritional status [63]. PCOS is a complex multi-factorial, chronic disease state that is one of the leading causes of anovulatory infertility and subfertility. PCOS is distinguished by a range of ovarian, hormonal, and metabolic disturbances and exhibits chronic menstrual irregularities, follicular cysts on ultrasonography, hyperinsulinemia, obesity, and hyperandrogenemia. In utero hyperandrogenism exposure may perturb the epigenetic reprogramming in foetal reproductive tissue and could be a cause of post-natal PCOS. Endometriosis has not been proven to have a genetic cause, but it is worth considering that abnormal expression of candidate genes could be provoked by different epigenetic modifications including DNA methylation, heterochromatization or aberrant regulation of miRNA expression [64].

It is well established that the decline in female reproductive responses is associated with advanced maternal age and postovulatory aging of oocytes [65], which leads to compromised quality of oocytes and a lower pregnancy rate. Understanding the underlying causes of these circumstances, which may include mitochondrial dysfunctions, aneuploidy, or epigenetic changes, has recently drawn increased attention. This is an important factor in reproduction and reproductive health in order to ensure that the correct epigenetic modifications occur during oogenesis and early embryo development [66].