DNA methylation is a chemical modification that occurs on the DNA molecule, specifically on cytosine residues (Fig. 1). This modification involves the addition of a methyl group to the cytosine ring, resulting in the formation of 5-methylcytosine (5mC). DNA methylation is a crucial epigenetic mechanism that regulates gene expression and is involved in various cellular processes, such as development, differentiation, and genomic stability (Moore et al. 2012).

The process of DNA methylation is carried out by a family of enzymes called DNA methyltransferases (DNMTs). There are three types of DNMTs: DNMT1, DNMT2, and DNMT3. DNMT1 is responsible for maintaining the pattern of DNA methylation during DNA replication, while DNMT3a and DNMT3b are responsible for de novo methylation, meaning they add methyl groups to previously unmethylated cytosines In contrast, DNMT2, originally considered to be DNA methyltransferase is actually catalyzing the methylation of tRNA, (Jeltsch et al. 2017).

The process of DNA methylation occurs when the DNMT enzymes transfer a methyl group from the methyl donor molecule S-adenosyl methionine (SAM) to the carbon in position 5 of the cytosine ring. In animals, DNMTs predominantly recognize CpG dinucleotides (where cytosine and guanine are adjacent to each other in the DNA sequence) and add a methyl group to the cytosine residue.

The process of DNA methylation is reversible, and there are enzymes called demethylases that can remove the methyl group from 5mC to convert it back to cytosine. Ten-eleven translocation (TET) enzymes, so named due to their involvement in the translocation between chromosome 10 and chromosome 11, are responsible for the process of active DNA demethylation. This process involves the conversion of 5mC to 5-hydroxymethylcytosine (5hmC) and subsequent oxidation steps that eventually lead to the removal of the methyl group. In addition, demethylation may occur passively due to a lack of re-methylation after cell replication.

The parasitic wasp Nasonia vitripennis, commonly found in temperate zones exhibits a unique reproductive behaviour in response to short photoperiods (SP). Under autumnal SP, N. vitripennis females play a crucial role as mothers by laying eggs that will later undergo developmental arrest in the larval stage, resulting in diapause. The diapause is averted under long photoperiods (LP). This adaptation enables the species to survive the winter. The transgenerational transfer of the photoperiodic information from mothers to their offspring alluded to an epigenetic mechanism, which was shown to involve DNA methylation (Pegoraro et al. 2016). Nasonia has a complete mammalian-like kit of DNA methylation machinery (Werren et al. 2010) including all three DNA methyltransferases (Dnmt1-3), suggesting that methylation has a role to play in the life-history of this species, like in other hymenopteran insects (Lyko et al. 2010). Indeed, sequencing of bisulphite-treated DNA reveals that DNA methylation is prevalent and occurs at CpG sites as in mammals (Park et al. 2011). This is in sharp contrast to Drosophila, where only a single DNMT2 is present and methylation is rare, occurring at CpT or CpA sites (Raddatz et al. 2013).

DNA methylation analysis using reduced representation bisulphite sequencing (RRBS) revealed 31 genes whose methylation differed between SP and LP (Pegoraro et al. 2016). Given the limited CpG coverage by RRBS, it is estimated that the actual number of differentially methylated genes is 20-fold higher. Importantly, experiments knocking down Dnmt1a and Dnmt3 by dsRNAi injections, or blocking DNA methylation pharmacologically, disrupted the photoperiodic response of the wasps. Specifically, following the RNAi knockdown, the diapause incidence increased under LP, and after the pharmacological treatment the diapause increased in LP and decreased in SP. Both these experiments indicated the causal role of DNA methylation.

DNA methylation was also implicated in the mammalian photoperiodic response. The thyroid hormone (T4) plays an important signalling role in the hypothalamus in transmitting photoperiodic information to the reproductive neuroendocrine system. Changes in photoperiod regulate the expression of the enzyme deiodinase type II (DIO2) that convert T4 to either the active T3, or the enzyme deiodinase type III (DIO3) that renders the receptor inactive, creating a seasonal gating mechanism for thyroid hormone receptor signalling (Dardente et al. 2014).

The hormone melatonin, produced by the pineal gland, also plays a crucial role in the regulation of the circadian system and photoperiodic responses. Melatonin is a neurohormone that is synthesized and released in a circadian rhythm, primarily during the dark phase of the day. Its production is under the direct control of the SCN. Pineal melatonin is not only essential for the control of circadian rhythms but also plays a vital role in mediating the effects of photoperiod on the reproductive neuroendocrine system. It is necessary for the induction of dio3 mRNA expression by photoperiod (Ono et al. 2008), which acts as a molecular switch for the seasonal control of reproduction. Melatonin levels, in conjunction with the photoperiodic information received by the SCN, modulate the expression of dio3, influencing the activity of the thyroid hormone receptor signalling pathway.

A study on Siberian hamsters (Phodopus sungorus) with seasonal breeding patterns found that DNA methylation of the dio3 promoter was critical in regulating dio3 expression (Stevenson and Prendergast 2013). The study found that under LP, dnmt3b was expressed at higher levels, resulting in a reduction in dio3 mRNA levels. Specifically, methylation was detected in 85% of 17 CpG sites in the dio3 promoter region of hamsters under LP, compared to 42% under SP. The reduced dio3 promoter methylation in SP was associated with up-regulation of dio3 expression, which in turn resulted in gonadal regression (Stevenson and Prendergast 2013).

As previous studies have shown evidence for circadian changes in DNA methylation (Azzi et al. 2014), a follow-study was carried out to investigate whether the circadian patterns of hypothalamic dnmts transcripts are influenced by the seasonal time (Stevenson 2017). The results showed that expression of dnmt3a (and to a lesser extent dnmt1, dnmt3b) was significantly greater in LP compared to SP at multiple points along the circadian period.

A study by Lindner et al. (2021) demonstrated the involvement of DNA methylation in seasonal timing in the great tit (Parus major). Several loci were identified where the level of promoter methylation was associated with the reproductive timing of the females. The most interesting gene was NR5A1, which codes for a transcription factor that regulates the expression of many key genes within the reproductive axis. Three CpG sites in the promoter region of NR5A1 showed a substantial increase in DNA methylation with the reproductive timing of the females.

The Melatonin receptor 1A (MTNR1A) is a pivotal gene involved in regulating the estrus cycle and seasonal reproduction in sheep. Recent investigations conducted in ewes by He et al. (2023) revealed that changes in photoperiod trigger DNA methylation at a specific cytosine site within the core promoter region of the MTNR1A gene. Moreover, an elevation in DNA methylation levels was observed following the long photoperiod (LP), demonstrating a significant negative correlation with the expression of MTNR1A.

In contrast to the circadian pacemaker, which is relatively resistant to temperature changes, the photoperiodic timer is highly sensitive to thermal modulation (Tyukmaeva et al. 2020). DNA methylation may play a crucial role in the integration of the photoperiodic timer with temperature changes. A recent study on redheaded buntings Emberiza bruniceps (Trivedi et al. 2019) proposed that DNA methylation mediates this integration. In long-day breeder songbirds, including the male redheaded buntings, exposure to LP triggers testicular enlargement, which is amplified at higher temperatures. The hypothalamus-pituitary–gonadal (HPG) pathways are involved in the regulation of gonadal growth and development, which is photoinduced under LP.

Histones are proteins that help package DNA into a compact structure in the nucleus. Modifications to histones, such as acetylation and methylation (Fig. 1), can affect how tightly the DNA is packaged and therefore its accessibility to the transcriptional machinery (Millán-Zambrano et al. 2022). Acetylation is the addition of an acetyl group to the lysine residues in histones. This modification typically leads to a more open chromatin structure and increased gene expression. Acetylation can neutralize the positive charge on histones, reducing their ability to interact with negatively charged DNA and making the DNA more accessible to transcription factors (Rothbart and Strahl 2014).

Methylation is the addition of a methyl group to lysine or arginine residues in histones. This modification can either activate or repress gene expression, depending on the specific residue that is methylated and the number of methyl groups that are added. For example, methylation of lysine 4 on histone H3 (H3K4me) is generally associated with active gene expression, while methylation of lysine 9 on histone H3 (H3K9me) is associated with gene silencing (Vinci 2012).

A study in the flesh fly Sarcophaga bullata suggested that histone modification might be a type of epigenetic process that contributes to the regulation of pupal diapause (Reynolds et al. 2016). The total acetylation of histone H3 was significantly reduced in early diapause compared to same-stage pupae that did not undergo diapause. Furthermore, a reduction in transcription of gcn5, a gene encoding histone acetyltransferases (HAT), which is responsible for the decrease in histone H3 acetylation. Additionally, first-instar larvae in diapause show changes in genes and enzymes associated with histone acetylation/deacetylation, including a 1.8-fold increase in transcription of tip60 and reptin, genes encoding parts of the Tip60 HAT complex, and a substantial downregulation of histone deacetylases (HDAC) hdac3 and sirt1 in diapausing pupae. Overall, these results suggest that changes in histone H3 acetylation, as well as shifts in HAT and HDAC abundance and activity, are associated with diapause in S. bullata and have a significant impact on the programming, maintenance, and termination of diapause.

A similar global repression of transcription, driven by histone deacetylation, was also found in hibernating mammals. In thirteen-lined ground squirrels (Ictidomys tridecemlineatus), torpor was accompanied by attenuation of transcription, which, in turn, is associated with histone H3 acetylation (Biggar and Storey 2014). Studies in the Siberian hamster indicated that expression of hdac1, hdac2, and hdac3 was regulated differently by the photoperiod and in a tissue-dependent manner (Lynch et al. 2017). For example, hdac3 expression in the uterus increased in LP, while hdac2 was reduced. In the testes, hdac1 is downregulated under SP, while hdac3 is upregulated.

Histone methylation also serves as a regulator of photoperiodic diapause. In the cotton bollworm Helicoverpa armigera, under SP, there is a low level of tri-methylation of lysine 27 on histone H3 protein (H3K27me3). This represses the prothoracicotropic hormone (PTTH) gene expression, causing pupal diapause and lifespan extension (Lu et al. 2013). Similarly, differences in levels of H3K27me2 and H3K4me3 were associated with embryonic diapause in annual killifish (Toni and Padilla 2016).

In Drosophila melanogaster, active chromatin marks H3K4me3 and H3K36me1 were found to be reduced in diapausing ovaries, promoting diapause plasticity (Evans et al. 2022). The study also revealed that chromatin determinants of diapause plasticity may be genotype-dependent, indicating genetic variation for epigenetic regulation of reproductive diapause.

Non-coding RNA molecules are RNA molecules that do not encode for proteins, but instead have regulatory roles in gene expression. A significant class of non-coding RNAs, is microRNA (miRNA), which regulates gene expression at the post-transcriptional level (Bartel 2018). miRNA are small molecules, typically consisting of 18 to 25 nucleotides, which are involved in the RNA interference machinery, binding to the untranslated regions (UTRs) of mRNA molecules to suppress or modulate protein translation or induce mRNA decay (Fig. 1). A single miRNA has the potential to target numerous mRNAs, suggesting a global impact on the transcriptomes and proteomes of eukaryotes. Additionally, miRNAs can function as epigenetic modulators by targeting key enzymes that catalyse epigenetic reactions, such as DNMTs and HDACs (Yao et al. 2019). Furthermore, miRNA expression itself is subject to regulation by epigenetic machinery, including DNA methylation, RNA modification, and histone modification (Yao et al. 2019).

A recent study in Drosophila demonstrated the important role of miRNA in photoperiodic timing (Pegoraro et al. 2022). The comparison of miRNA expression in the head under SP and LP revealed seven miRNAs that were differentially expressed. Furthermore, overexpression of three of these miRNAs (dme-mir-2b, dme-mir-184, and dme-mir-274) in clock neurons in the brain, using the binary UAS/Gal4 system, led to a similar diapause response in SP and LP, suggesting that these miRNAs have functional roles in photoperiodic timing. Additionally, by utilizing both computational prediction and immunoprecipitation of RNA samples with the Argonaute-1 antibody from flies in LP and SP, the targets of photoperiodic miRNAs have been identified.

In a recent study conducted in sheep, the regulation of chromogranin A (CHGA) by non-coding RNA was identified (Di et al. 2023). CHGA was found to be upregulated in the pituitary pars tuberalis (PT), which plays a crucial role in encoding the photoperiodic response. Specifically, miR-25 was observed to bind to the 3' UTR region, resulting in the inhibition of CHGA expression. Additionally, miR-25 also targets a long non-coding RNA (lnc107153).

During the LP, there was a high expression of miR-25 in sheep, leading to the suppression of CHGA protein expression. However, in SP, the expression of Lnc107153 was found to be elevated, facilitating its binding to miR-25. This interaction weakened the inhibitory effect of miR-25 on CHGA expression. Consequently, there was a significant increase in the expression level of CHGA protein during SP.

Another class of non-coding RNAs is lncRNAs. These molecules are longer than 200 nucleotides and are not translated into proteins. While they do not encode proteins, they are involved in various cellular processes, including gene expression regulation (Reviewed by Statello et al. 2020). lncRNAs exhibit various interactions with DNA, other noncoding RNAs, and proteins, contributing to a wide array of modifications at transcriptional, post-transcriptional, and post-translational levels. They perform diverse roles such as serving as guides, decoys, and scaffolds, and participating in splicing, messenger RNA (mRNA) decay, and subcellular localization. Additionally, lncRNAs can impact chromatin architecture by engaging with chromatin-modulating proteins, thereby controlling transcriptional activity. This control can involve either promoting or preventing the recruitment and association of these proteins with chromatin. Through this mechanism, lncRNAs can either activate or repress transcription by recruiting regulatory factors to specific loci and modulating their functions (Reviewed by Sun et al. 2018).

A transcriptomic study (Nakayama et al. 2019) comparing SP and LP in the Japanese medaka fish (Oryzias latipes) revealed a photoperiodic lncRNA that was named a long-day-induced anti-sense intronic RNA (LDAIR). In transgenic fish where LDAIR has been knocked out, the expression of several neighboring genes has been altered. This neighborhood effect (common for lncRNA) has included the corticotropin-releasing hormone receptor 2 (CRHR2), known to be involved in the stress response. In line with this, under LP conditions (when LDAIR is expressed), wild-type medaka fish exhibited higher 'protective' diving behavior and increased light avoidance compared to knockout fish (Nakayama et al. 2019).

A recent study in sheep (Wang et al. 2022) identified 664 lncRNAs in the thyroid gland whose expression differed between short photoperiod (SP) and long photoperiod (LP). Co-expression network analysis suggested that some of these lncRNAs target genes such as CCNB3 and DMXL2, both of which are involved in reproduction.