Infertility is an often devastating diagnosis encountered by around one in six couples who are trying to conceive. Moving away from the long-held belief that infertility is primarily a female issue, it is now recognised that half, if not more, of these cases may be due to male factors. Recent evidence has suggested that epigenetic abnormalities in chromatin dynamics, DNA methylation or sperm-borne RNAs may contribute to male infertility. In light of advances in deep sequencing technologies, researchers have been able to increase the coverage and depth of sequencing results, which in turn has allowed more comprehensive analyses of spermatozoa chromatin dynamics and methylomes and enabled the discovery of new subsets of sperm RNAs. This review examines the most current literature related to epigenetic processes in the male germline and the associations of aberrant modifications with fertility and development.
Male infertility is a complex disease that, despite being extensively researched, still remains poorly understood. Although infertility can be attributed to decreased semen quality, testicular dysfunction, infections of the genital tract and known genetic disorders (such as single-gene alterations and chromosomal abnormalities) (Krausz et al. 2015), a number of male patients suffer from a condition known as idiopathic or unexplained male infertility (UMI), in which the aetiology is poorly understood (Urdinguio et al. 2015). Given the major genome reprogramming events that occur during gametogenesis and early development, in recent years, a number of studies have pointed towards aberrant epigenetic reprogramming of the genome as a potential contributory factor to male infertility (O’Doherty & McGettigan 2014, Laurentino et al. 2015, Stuppia et al. 2015, Urdinguio et al. 2015, Schagdarsurengin & Steger 2016, Kropp et al. 2017, Laqqan et al. 2017a, Nasri et al. 2017). It should be stressed that, at present, direct associations between epigenetic changes and infertility are unclear – it is indeed possible that they are simply correlated with changes in infertility and not causal.
There are currently a variety of definitions and opinions on the term ‘Epigenetics’, which can often stir debate. A widely used and accepted definition of epigenetic processes is one that defines them as heritable changes influencing gene expression that are not caused by changes to DNA sequence (Holliday 1987). These processes lead to the establishment of specialised chromatin states that are permissive or repressive to gene expression, can be influenced by the maternal or paternal environment due to their plasticity and are involved with determining cell identity (Schagdarsurengin & Steger 2016, Stewart et al. 2016). To date, DNA methylation has been the most comprehensively studied epigenetic mark (Smith & Meissner 2013), however, post-translational modifications to histone tails and the role of non-coding RNA molecules (both widely regarded as epigenetic processes/modifications) are gaining considerable attention.
Determining the underlying causes of unknown male factor infertility still remains a major challenge in reproductive medicine (Salas-Huetos et al. 2016). In the clinical setting, widely used semen analysis methods for diagnosing male factor infertility, such as microscopic examination and DNA fragmentation analyses, do not tell the whole story. These approaches are mostly ineffective for diagnosing UMI, where spermiograms of normozoospermic infertile patients are indistinguishable from those of normozoospermic fertile patients (Hamada et al. 2012). However, recent advancements in sequencing technologies have allowed researchers to evaluate the sperm epigenome and build a bigger picture of their underlying molecular profiles, helping to determine why some couples experience idiopathic infertility.
In this review, we will focus on the potential role of aberrant epigenetic reprogramming of the sperm genome on male fertility and early development, with particular emphasis on DNA methylation, histone tail modifications and non-coding RNA molecules (Fig. 1).
Epigenetic reprogramming during sperm development
The genome is extensively remodelled during mammalian germ cell development in both male and female germlines, and this epigenetic reprogramming is critical for imprinting and reprogramming in early embryos (Reik et al. 2001). Chromatin organisation in sperm and oocytes is markedly different: sperm DNA is tightly packed, histones are largely replaced with protamines and there is a large degree of DNA methylation while oocytes have less DNA methylated and prevalently open chromatin (O’Doherty & McGettigan 2014, Bao & Bedford 2016, Hanna & Kelsey 2017). Given that the emphasis of the current review is on epigenetic mechanisms during spermatogenesis and male fertility, we will be focusing solely on the paternal germline. For a comprehensive review on genome reprogramming in the female germline, see Messerschmidt et al. 2014, Stewart et al. 2016.
There are two major reprogramming events that involve genome-wide erasure and re-establishment of DNA methylation patterns during mammalian development; the first occurs in primordial germ cells (PGCs) and the second following fertilisation in the pre-implantation embryo (Prokopuk et al. 2015). These events have been mostly characterised using the murine model and, although epigenetic profiles have been assessed in spermatozoa and embryos of other species (Dean et al. 2001, Beaujean et al. 2004, Deshmukh et al. 2011, O’Doherty et al. 2012, 2015, Guo et al. 2014b), detailed comparative studies are still lacking. Firstly, during proliferation and migration of PGCs from the epiblast to colonisation of the genital ridge (~E6.5–E13.5), the PGC genome is demethylated in a biphasic manner that results in their genomes displaying very low levels of DNA methylation (Messerschmidt et al. 2014). From here, PGCs are divergently remethylated in males and females in a germ cell-specific fashion (Sasaki & Matsui 2008, Seisenberger et al. 2012). Male germ cells gain as much as 50% global methylation by E16.5 and are almost completely methylated at birth. In contrast, remethylation in the female germline occurs after birth, during the oocyte growth phase (O’Doherty et al. 2012, Tomizawa et al. 2012). Secondly, following fertilisation, male pronuclei are actively and rapidly demethylated in the zygote while the maternal genome is passively demethylated in a replication-dependent manner during cleavage (Lees-Murdock & Walsh 2008). Until recently, it was widely accepted that the maternal genome and paternal genome were not actively demethylated or passively demethylated, respectively. It should be noted, however, that active and passive demethylation do not occur exclusively on the paternal and maternal genomes, as it has been demonstrated that both genomes undergo widespread active and passive demethylation in the zygote prior to the first mitotic division (Gkountela & Clark 2014, Guo et al. 2014a).
These periods of reprogramming occur during crucial developmental time points and represent windows of susceptibility for epigenetic errors to be introduced, that may impact fertility and embryonic competence. Over four decades ago, it was reported that a Fathers occupation, and potential exposure to (spermatogenesis impairing) environmental toxins, was associated with carcinogenic defects being transmitted to their children (Fabia & Thuy 1974). Since then, a number of other studies have shown links between paternal exposure to environmental toxins (e.g. paints, hydrocarbons, pesticides) and offspring health; reviewed in (Soubry et al. 2014). Given the continuous cycles of mitosis and meiosis that occur during spermatogenesis in adult males, the chances of accumulating environmentally induced epigenetic errors during this protracted period of replication and cell division are much greater in males than in females (Messerschmidt et al. 2014).
As mentioned earlier, DNA methylation is the most widely investigated epigenetic mark. This epigenetic modification is essential for male and female gametogenesis and mammalian development (Okano et al. 1999, Bourc’his et al. 2001, Kaneda et al. 2004). In the mammalian genome, DNA methylation mostly occurs on the fifth position of cytosine bases in the context of cytosine followed by guanine (CpG dinucleotides): this is referred to as 5-methylcytosine (5mC) (Yoder et al. 1997). It can also occur to a much lesser extent at cytosine bases in a non-CpG context (Jang et al. 2017). Additionally, 5mC can be oxidised by members of the ten-eleven translocation family of proteins to form 5-hydroxymethylcytosine (5hmC), 5-formylcytosine and 5-carboxylcytosine (Ito et al. 2011, Wu & Zhang 2014, Neri et al. 2016). These intermediary modifications are much less studied compared to 5mC or 5hmC and are yet to be characterised during male gametogenesis.
Aberrations in sperm DNA methylation have also been shown to be associated with abnormalities in semen parameters (low sperm count, decreased volume of semen, reduced progressive motility, high percentage of immotile sperm lowered sperm vitality) of subfertile males that were trying to conceive for at least 10 years (Laqqan et al. 2017a). The authors of this study identified that DNA methylation was consistently different at a small number of CpGs between proven fertile males and subfertile case subjects, using Infinium Human Methylation 450 K BeadChip arrays, validating previous findings using local deep bisulfite sequencing (Gries et al. 2013). This study builds on earlier investigations, and recent studies by Nasri et al. (2017) and Kobayashi et al. (2017), showing that aberrant methylation, particularly at imprinted loci, is associated with poor-quality sperm (Marques et al. 2004, 2008, Kobayashi et al. 2007, Poplinski et al. 2010, Pacheco et al. 2011, Laurentino et al. 2015). However, the methylation differences in Laqqan et al. study were very small and restricted to a small number of CpGs; therefore, whether these differences are related to the subfertile phenotype remains to be fully determined. In a previous study using the same 450 K DNA methylation array, Urdinguio et al. revealed that DNA methylation patterns in the male partners of couples with UMI were measurably different at specific loci and Alu Yb8 repetitive elements (Urdinguio et al. 2015). Given that no abnormalities in semen parameters are distinguishable between UMI patients and fertile control patients, and that female factor infertility was ruled out, this study provides encouraging evidence to show that aberrant sperm DNA methylation may be associated with fertility impairment in couples with UMI. This is further supported by a study by Aston and colleagues showing that, in similar cohorts of patients, sperm DNA methylation may be used to predict fertility and potentially predictive of embryo quality during IVF (Aston et al. 2015).
It has been reported that smoking has a strong correlation with CpG methylation and sperm count, morphology and motility (Laqqan et al. 2017b). The authors of this study also identified that smoking is associated with changes in sperm DNA methylation patterns, specifically at CpGs located in regions related to MAPK8IP and TKR genes. A link between smoking, male fertility and methylation at imprinted loci has been highlighted in a study by Dong et al. (2016). In this study, they revealed that hypomethylation of the H19 imprint control region (ICR) and hypermethylation of the SNRPN-ICR were associated with infertility and that the risk was potentiated by smoking. In addition to smoking, alcohol consumption has been shown to be associated with DNA methylation at regulatory regions of the imprinted gene, H19, in spermatozoa of humans and mice; however, the impact of alcohol-induced alteration to DNA methylation on fertility remains to be determined (Ouko et al. 2009, Stouder et al. 2011, Lim & Song 2012).
The effect of paternal diet on sperm quality, offspring health and epigenetics has been discussed previously (Schagdarsurengin & Steger 2016). In an early study, it was demonstrated that offspring from male mice fed a low-protein diet exhibited differential expression of key lipid and cholesterol biosynthesis-related genes in the liver that may be related to altered DNA methylation patterns inherited through the male germline (Carone et al. 2010). In a similar study, differential methylation was observed at genes implicated in development, chronic diseases, diabetes, autism and schizophrenia in sperm isolated from mice fed a folate-deficient diet, providing evidence for a possible link between diet, the sperm methylome and offspring health (Lambrot et al. 2013). However, it was reported in a subsequent study from Rando and colleagues that, although paternal diet has an effect on offspring phenotype, no consistent effects of diet on the sperm methylome are observable and that variation in offspring DNA methylation is due to genetic and epigenetic variations (Rando & Simmons 2015, Shea et al. 2015, Whitelaw 2015). Although the precise epigenetic mechanism/s of how the paternal diet can affect fertility and offspring health remain to be fully unravelled, the importance of the paternal (and maternal) diet on fertility should not be disregarded and needs to be investigated further.
Chromatin structure and modifications
During the finally stages of male gametogenesis (spermiogenesis), motile spermatozoa develop from haploid round spermatids (Bao & Bedford 2016). The chromatin undergoes a dramatic reconfiguration during this time, when the vast majority (90–95%) of nucleosomal core histones (H2A, H2B, H3 and H4) are first replaced by transitionary proteins and then by protamines (Balhorn 1982, Meistrich et al. 2003, Eirin-Lopez et al. 2006, Rathke et al. 2014). This process, known as histone-to-protamine transition, facilitates a tight packaging of sperm DNA enabling condensation of sperm heads and protection of DNA from damage and mutagenesis (Ward & Coffey 1991, Rathke et al. 2014). The ratio of remaining nuclear histones to protamines has been shown to play a role in male fertility (Zhang et al. 2006). Findings from this study suggested that infertile men possess a higher proportion of spermatozoa with an increased histone-to-protamine ratio, compared to fertile controls. Histone-to-protamine transition and links with fertility have also been reported elsewhere (Carrell & Liu 2001, Aoki et al. 2006, Carrell et al. 2007). These studies showed that altered expression of protamine, protamine 1 and protamine 2 are related to diminished spermatogenesis and fertilisation capacity. In addition, cryopreservation, a routinely used process for sperm storage in fertility clinics, has been shown to impact chromatin integrity (Fortunato et al. 2013). The authors of this study reported that cryopreservation influenced chromatin decondensation and may lead to sperm chromatin cryoinjury.
The small proportion of histones (5–10%) that are retained during histone-to-protamine transition and the post-translational modification of these remaining histones and their tails is an expanding area of biological research (Carrell & Hammoud 2010, Krejci et al. 2015). Retained histones are modifiable and thought to occur at specific genomic loci that are involved with regulating transcription following fertilisation (Hammoud et al. 2009, Miller et al. 2010). However, the number of studies investigating the relationship between chromatin structure in sperm and fertility remains substantially lower, relative to studies involving DNA methylation or abundance of RNA molecules (discussed later). One potential hurdle may be that pinpointing histone modifications that are directly associated with fertility may be difficult due to the plethora of post-translational modifications (e.g. acetylation, methylation, SUMOylation, phosphorylation) that can be present on the various histone isoforms (Luense et al. 2016). However, studies have shown links between fertility and chromatin structure and are discussed below.
Several recent studies investigating the relationship between chromatin structure and male fertility have been carried out using the bovine model. An early study evaluated the differences in bull fertility, histone retention and expression of a histone variant (H3.3) and two core histones (H2B and H4) using immunoblotting, immunocytochemistry and staining techniques (de Oliveira et al. 2013). Although no differences were detected for H3.3, H2B or H4 between sperm from the low-fertility (LF) group and the high-fertility (HF) group, there were differences in chromatin condensation between the groups that were associated with in vivo bull fertility. Interestingly, using transgenic mice carrying null mutations at two genes encoding the histone variant H3.3, Tang and colleagues revealed that H3.3 is heavily involved in viability and male fertility (Tang et al. 2015). Using immunocytochemistry and flow cytometry techniques, it has been shown that acetylation and methylation of sperm histone 3 lysine 27 (H3K27ac and H3K27me3) are associated with bull fertility (Kutchy et al. 2018). Furthermore, flow cytometry, immunofluorescence and western blotting have been used to demonstrate that expression of the testis-specific histone variant 2B (TH2B) in sperm isolated from bulls with contrasting fertility scores can be potentially used to evaluate semen quality and male fertility (Kutchy et al. 2017). In another in vitro study comparing HF and LF bulls, Castro and colleagues assessed the potential contribution of chromatin integrity to fertility using three separate approaches to evaluate chromatin deficiency (CMA3), chromatin stability (SCSA; AO+) and DNA fragmentation (COMET assay) (Castro et al. 2018). Although differences in chromatin deficiency were observed between the HF and LF groups, the authors concluded that protamine deficiency in bovine spermatozoa may not have a strong biological impact on in vitro fertility.
Using a more targeted approach with a custom ChIP-on-chip array, Verma et al. investigated di-methylated H3K4 (H3K4me2) and tri-methylated H3K27 (H3K27me3) in spermatozoa of water buffalo bulls (Bubalus bubalis) identified as having wide differences in fertility (Verma et al. 2015). For H3K4me2 and H3K24me3, the authors identified 84 and 80 genes, respectively, that were differentially enriched between mature sperm from high and sub-fertile bulls. Gene ontology analysis of these differentially marked genes revealed enrichment for processes such as germ cell development, spermatogenesis and embryonic development – demonstrating that appropriate chromatin configuration at functionally relevant genes is related to bull fertility. Similarly, it has recently been shown that epigenetic marks in the sperm of Xenopus laevis, a model organism for developmental biology research, regulate embryonic gene expression (Teperek et al. 2016). Specifically, this work showed important regulatory roles for loss of methylation of histone H3 on lysine 4 (H3K4) and retention of methylation of histone H3 on lysine 27 (H3K27) and demonstrated that epigenetic cues delivered by the sperm are required for correct embryonic gene expression. Work by Öst and colleagues, using the model organism Drosophila, provided a link between paternal sugar intake, H3K9/K27me3-dependent reprogramming of metabolic genes and offspring adiposity (Ost et al. 2014). In humans, acetylation of histone H4 at lysine 12 (H4K12ac) has been implicated at developmentally important gene promoters in subfertile men (Vieweg et al. 2015).
Work by Dumasia and colleagues, looking at the roles of oestrogen and the oestrogen receptors (ESRs) during rat spermatogenesis, has shown that ESR are involved with spermatogenesis and fertility (Dumasia et al. 2015, 2016, Kumar et al. 2017); oestrogen signalling regulates DNA methylation (Dumasia et al. 2017a) and that treating adult rats with selective ESR agonists resulted in altered chromatin modifications in the testis (Dumasia et al. 2017b). More specifically, they found that ESR agonist treatment caused increased histone retention, protamine deficiencies, and altered levels of histone modifications that correspond with active and repressed chromatin states, and altered heterochromatin marks (Dumasia et al. 2017b). In another study investigating the impact of oestrogens in the environment, Filby et al. analysed Pimephales promelas, a model fish species, after exposure to differing concentrations of oestrogen as a single chemical and as part of an oestrogenic effluent from waste water treatment work (WwTW) (Filby et al. 2007). They revealed an association between oestrogens in the environment and some adverse and sex-related health effects observed in the fish including genotoxic damage, modulated immune function, and altered metabolism, in addition to endocrine disruption (Filby et al. 2007). Future studies investigating the impact of environmental oestrogens will allow us to determine how they affect chromatin remodelling during spermatogenesis and if they negatively impact male fertility.
Sperm-borne coding and non-coding RNA molecules
Until recently the presence of RNA in sperm was largely thought to be due to carryover or contamination with non-sperm cells (discussed in O’Doherty & McGettigan 2014). The presence and characterisation of RNA molecules in sperm has been confirmed and greatly facilitated in recent years through technological advances in sequencing platforms and the focus has now shifted to elucidating their function. Although transcriptionally quiescent, sperm have been shown to contain a wide variety of RNA molecules, both coding and non-coding, reported to be required for spermatogenesis, early development and epigenetic inheritance (Schuster et al. 2016). Transcriptomic analysis of sperm isolated from men of known fertility, idiopathic infertility (normozoospermic patients), and asthenozoospermia (reduced motility) has identified different RNA profiles between the patient cohorts, highlighting the potential importance of these molecules in male fertility (Bansal et al. 2015).
It has been reported in humans that as little as 2% of transcriptional output is translated into protein, thus the remaining 98% is non-coding (Mattick 2001). The inclusion of ncRNA as epigenetic regulators of gene function remains somewhat divisive and evokes discussions on whether it is appropriate to define them as such. The focus of this review is not to determine whether or not ncRNAs should be included under the epigenetic umbrella. Therefore, we will take the wider view that they are involved in epigenetic regulation and will report the most recent findings linking them with male infertility. Isolation of sperm RNAs remains methodologically challenging due to their very low abundancies and affinity of certain RNAs to chromatin (Goodrich et al. 2007, 2013). Nevertheless, many investigations, summarised hereafter, have reported associations between sperm-borne RNA molecules and fertility and shown that they have key roles in subsequent embryonic development.
A 2015 study suggested that analysing sperm RNA elements (SREs) could be an efficient predictor of effective fertility treatments on an individual level (Jodar et al. 2015). This study revealed that idiopathic infertile couples who had an incomplete set of necessary SREs had a significantly reduced probability of achieving a live birth by intrauterine insemination (IUI) or timed intercourse (TIC) from 73 to 27%. When using assisted reproductive technologies (ART), such as IVF or ICSI, the absence of SREs did not seem to be critical. Furthermore, this study showed that the infertility of idiopathic infertile couples without a complete set of SREs is most likely due to a male component. While the infertility of couples that failed to achieve a live birth, but have a complete set of SREs, could be due to female factor infertility.
Since the discovery of epigenetic reprogramming during gametogenesis and embryogenesis (Reik et al. 2001), and recent studies showing that environmentally induced DNA methylation alterations in the F1 generation are not maintained in the F2 generation (Radford et al. 2014) and that sperm methylation errors are corrected by reprogramming (de Waal et al. 2012, Iqbal et al. 2015), epigenetic researchers have shifted their focus to other molecules and mechanisms that have been proposed to play a role in maintaining transgenerational transfer of acquired traits, such as sperm RNAs and sperm RNA modifications (Chen et al. 2016). The first major piece of evidence supporting a role for sperm RNAs in the transgenerational inheritance of acquired traits came from a 2014 study which demonstrated that by injecting total sperm RNAs from mice that were exposed to mentally stressful conditions into normal zygotes, behavioural patterns and alterations in metabolism that were observed in the father can be recapitulated in the offspring, and maintained through subsequent generations (Chen et al. 2016). Subsequent studies showed that mice who were fed a high-fat diet (HFD) or high-fat-high-sugar diet also contained sperm RNAs that, when injected into normal zygotes, would produce offspring with similar paternal metabolic disorders (Chen et al. 2016). These studies have led many researchers to question which RNAs are potentially responsible for transmitting the paternal phenotypes. Recent evidence has turned researchers’ attentions to sperm tsRNAs and miRNAs, which are up regulated in mature sperm after HFD, low-protein diets or after other environmental exposures (Chen et al. 2016).
The ncRNA transcriptome can be broadly divided into two main groups based on their length. The first group is comprised of short ncRNAs ranging between 20 and 30 nucleotides and the second group is made up of lncRNA transcripts that are >200 nucleotides in length. The role of these two groups of RNAs in male factor infertility is discussed below.
Short non-coding RNAs
Short ncRNAs have been investigated extensively during the past decade (Huang et al. 2013). Two short ncRNA classes, sperm piRNA and microRNA (miRNA), which play an essential role in spermatogenesis, have been implicated in male fertility disorders (Carmell et al. 2007, Houwing et al. 2007, Das et al. 2008). Capra et al. recently investigated piRNAs and miRNAs derived from high motile and low motile cryopreserved bovine spermatozoa populations, through small RNA sequencing, and found that a number of miRNAs were differentially expressed in the two populations (Capra et al. 2017). Moreover, they found that these miRNAs targeted genes involved in apoptosis, mitochondrial membrane integrity and spermatogenesis alteration, indicating a possible role in bull fertility. In a study by Pantano et al., small RNA sequencing revealed an abundance of piRNAs derived from pseudogenes (Pantano et al. 2015). The authors proposed that these pseudogene-derived piRNAs may regulate their parent gene expression during sperm development and therefore may be required for fertility. Analysis of spermatozoa from normozoospermic fertile and normozoospermic infertile patients, using the TaqMan Array Human MicroRNA, has revealed that normozoospermic fertile and infertile individuals convey distinct miRNA cargos that may have implications in male reproductive performance (Salas-Huetos et al. 2016).
Recent studies have shown that the injection of specific subsets of miRNAs into normal zygotes can reproduce paternal phenotypes which can be followed into subsequent generations, suggesting that these miRNA can induce changes that are maintained in the germline. tsRNAs, which are derived from the 5′ end of tRNAs, have also been shown to be involved in epigenetic inheritance of acquired traits. By injecting sperm tsRNA from mice that were on a HFD into normal zygotes, researches have shown that metabolic disorders observed in the father can be recapitulated in the offspring (Chen et al. 2016).
In non-mammalian species, the recently discovered phase small-interfering RNAs (phasiRNA), that bear similarities to mammalian piRNAs (Zhai et al. 2015), have been shown to be important for male meiosis and fertility in plants (Dukowic-Schulze et al. 2016, Fan et al. 2016). A study using photoperiod-sensitive male sterility (PSMS) rice (Oryza sativa L.) demonstrated the potential biological importance of (phasiRNA), which are an emerging class of 21- or 24-nt small RNAs generated from precursor RNA. This study showed that accumulation of a 21 nt PMS1T-phasiRNA eventually caused male sterility in the rice, although further studies are needed to elucidate the mechanistic pathway involved (Fan et al. 2016).
Long non-coding RNAs
lncRNAs have gained considerable attention in recent years. They are, generally, polyadenylated non-coding RNA molecules that are RNA Polymerase II transcribed. Although classified as non-coding, it should be noted that a number of lncRNAs have been shown to harbour short open reading frames (sORFs) that can be translated into short peptides (Rohrig et al. 2002, Galindo et al. 2007, Kondo et al. 2007, 2010, Magny et al. 2013, Anderson et al. 2015). lncRNAs have been shown to possess a variety of functions such as transportation of mRNA from the nucleus (Chen & Carmichael 2009), acting as competing endogenous RNAs for miRNAs (Cheng & Lin 2013) and decoys for DNA-binding proteins (Kino et al. 2010). Some well characterised lncRNAs are involved in regulating imprinted gene expression and X chromosome inactivation and include Xist (Brockdorff et al. 1992), H19 (Brannan et al. 1990) and AIRN (Sleutels et al. 2002). For an in-depth review on the evolution, genomic contexts, biological functions, and mechanisms of action of lncRNAs (Kung et al. 2013). lncRNAs have pivotal roles during animal sperm development, for a comprehensive reviews on lncRNA during spermatogenesis (Luk et al. 2014, Zhang et al. 2016).
It has been reported that type 1 and type 2 diabetes mellitus can negatively impact fertility (Singh et al. 2014, Wiebe et al. 2014) and there is a great amount of data indicating that lncRNAs are involved in diabetes mellitus, reviewed in (He et al. 2017). A large number of lncRNAs (and mRNAs) are differentially expressed in sperm isolated from diabetic and non-diabetic mice models (Jiang et al. 2016), therefore it is possible that some of these lncRNAs are related to diabetes-related male infertility – however larger studies are needed to confirm or rule this possibility out. lncRNAs have also been shown to play a role in sperm motility (Liu et al. 2017). Using Beijing-you cocks with divergent sperm motility scores, Liu and colleagues identified several lncRNAs that were differentially expressed in the testis between animals with high and low sperm motility. Although this study provides some novel data linking lncRNAs to sperm motility (and potentially fertility), analysis of lncRNA in the high and low motility sperm, and the fertility of said sperm, would aid in the identification of motility-associated lncRNAs.
Although having important roles during spermatogenesis and potentially being important for male fertility, some lncRNAs have been shown to be indispensable for fertility. Despite being enriched in the sperm nucleus, transgenic mice carrying homozygous null mutations for the lncRNA, Malat1, have no identifiable defeats in fertility (Zhang et al. 2012, Johnson et al. 2015). Similarly, knockout of the testis-specific lncRNA, Tsx, results in viable fertile offspring (Anguera et al. 2011). These studies show that although lncRNAs can be important for germ cell development they may not be required for fertility; and thus highlight the requirement for further studies aimed at identifying the role of lncRNAs in male fertility.
Modification of RNA
In addition to the involvement of a variety of RNA molecules in male fertility, modifications to RNA itself have also been implicated in fertility. Although there are in excess of 140 modified RNA nucleotide variants, one RNA modification that is getting a lot of attention recently is N6-methyladenine (m6A), in which an adenine base is methylated at the 6′ position (Fu et al. 2014). m6A is the most abundant mRNA/lncRNA modification discovered thus far, and is the only modification that has exhibited characteristics of epigenetic regulation including readers, writers and erasers (Dominissini et al. 2012, Meyer & Jaffrey 2014, Liu & Pan 2015). m6A has been implicated at virtually every level of mRNA regulation including pre-mRNA splicing, mRNA export, and stability, and translation of mRNA (Lin et al. 2017). Due to m6A’s extensive involvement in many biological processes, it has been associated with cancer progression (Batista et al. 2014, Geula et al. 2015) and many diseases including obesity and diabetes (Zeggini et al. 2007, Jia et al. 2011).
Work by Yang and colleagues has identified an association between levels of m6A in human sperm and fertility (Yang et al. 2016). The authors revealed that individuals with asthenozoospermia have increased levels of m6A compared to controls and suggested that increased levels of spermatozoal m6A is a potential risk factor for decreased sperm motility (Yang et al. 2016). The m6A modification has also been shown to be functionally vital for the survival of cells in culture in a study demonstrating that siRNA knockdown of METTL3, the methyltransferase known to be involved in m6A methylation, results in apoptosis (Lin et al. 2016). Another study showed that inactivation of either Mettl3 or Mettl14 in a male germ-cell specific manner resulted in a lack of m6A, which results in translational dysregulation and subsequently, a depletion of spermatagonial stem cells (SSC), although normal spermatogenesis was observed (Lin et al. 2017). However, this study also showed that double deletion of Mettl3 and Mettl14 results in impaired spermiogenesis due to translational dysregulation of factors that are key to spermatogenesis (Lin et al. 2017). m6A has been shown to be dynamically regulated at different developmental stages during spermatogenesis, with m6A enrichment occurring in pachytene/diplotene spermatocytes and round spermatids (Lin et al. 2017). These studies suggest that mRNA m6A modifications are critical to the temporal regulation of translation during spermatogenesis by marking genes that are key factors during spermatogenesis, stabilising those transcripts, increasing their translation, and preventing translation of genes that would be deleterious by erasing the m6A modification.
Mutation of a gene that is involved in the regulation of RNA modifications has also been shown to potentially influence male fertility (Zheng et al. 2013). A mutation in the gene that encodes ALKBH5, an m6A demethylase, causes increased levels of m6A, which also affects the export and metabolism of mRNA, and results in aberrant spermatogenesis, low sperm count, poor quality sperm and impaired fertility (Zheng et al. 2013). ALKBH5-deficient mice have been shown to have decreased testes size (Tang et al. 2018) and a reduced breeding success rate (Zheng et al. 2013), which has steered researchers focus to ALKBH5’s role in spermatogenesis. Subsequent studies by an independent group have shown that ALKBH5 is required for late meiotic and haploid phases of spermatogenesis (Tang et al. 2018). To study m6A modification in mRNA and lncRNA researchers use a method referred to as m6A-seq or MeRIP-seq in which an m6A antibody is used to immunoprecipitate RNAs with the m6A modification, followed by high-throughput sequencing (Tang et al. 2018). Due to the advancement in deep sequencing technologies, researchers have been able to map m6A modification using this method at a single nucleotide resolution and have found these modification to be enriched around stop codons and internal exons (Liebers et al. 2014, Linder et al. 2015). These advancements will help researchers to discovery novel roles of m6A modification and other RNA modifications alike.
Deciphering the underlying cause of male infertility, and determining the key molecular players, still remains a major challenge in reproductive medicine. This review highlights that there are many epigenetic factors involved with multiple layers of gene regulation at multiple genomic loci, related to sperm function and potentially important for fertility. As with any field of biological research, technical and biological limitations need to be carefully considered, the utility of sperm epigenetics and potential hurdles have been discussed in-depth recently by Jenkins and colleagues (Jenkins et al. 2017). However, carefully designed future studies investigating epigenetic modifications of the sperm genome and the vast suite of sperm-borne RNA molecules, and their modifications, are warranted and will help to shed more light on their involvement (cause or correlation) with male infertility and help with the development of new diagnostics and possibly therapeutics. Male infertility research will greatly benefit from studies using larger sample sizes that integrate multiple epigenomic analyses (integromics). This will help in the identification of robust epigenetic markers of fertility/infertility.
The majority of the studies examining the contribution of epigenetic processes to infertility discussed in this review have used samples consisting of large numbers of sperm. Given that sperm are one of the most diverse cell types (Ramon et al. 2014) and that sperm have been reported to be epigenetically heterogeneous (Laurentino et al. 2016), future investigations using high-throughput single cell epigenomics will be important to help determine whether aberrant infertility-related epigenetic marks are present in all spermatozoa of the ejaculate. Finally, advising couples suffering from UMI, and males in general, on known risk factors such as smoking (Dong et al. 2016, Laqqan et al. 2017b), obesity (Katib 2015, Craig et al. 2017) and alcohol consumption (Ouko et al. 2009, Lim & Song 2012) may help to reduce infertility-related epigenetic aberrancies in sperm and go some way on the long road to understanding and potentially treating male factor infertility.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this review.
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
AndersonDMAndersonKMChangCLMakarewichCANelsonBRMcAnallyJRKasaragodPSheltonJMLiouJBassel-DubyR et al. 2015 A micropeptide encoded by a putative long noncoding RNA regulates muscle performance. Cell 160 595–606. (https://doi.org/10.1016/j.cell.2015.01.009)
AokiVWLiuLJonesKPHatasakaHHGibsonMPetersonCMCarrellDT 2006 Sperm protamine 1/protamine 2 ratios are related to in vitro fertilization pregnancy rates and predictive of fertilization ability. Fertility and Sterility 86 1408–1415. (https://doi.org/10.1016/j.fertnstert.2006.04.024)
AweSRenkawitz-PohlR 2010 Histone H4 acetylation is essential to proceed from a histone- to a protamine-based chromatin structure in spermatid nuclei of Drosophila melanogaster. Systems Biology in Reproductive Medicine 56 44–61. (https://doi.org/10.3109/19396360903490790)
BrockdorffNAshworthAKayGFMcCabeVMNorrisDPCooperPJSwiftSRastanS 1992 The product of the mouse Xist gene is a 15 kb inactive X-specific transcript containing no conserved ORF and located in the nucleus. Cell 71 515–526. (https://doi.org/10.1016/0092-8674(92)90519-I)
CapraETurriFLazzariBCremonesiPGliozziTMFojadelliIStellaAPizziF 2017 Small RNA sequencing of cryopreserved semen from single bull revealed altered miRNAs and piRNAs expression between high- and low-motile sperm populations. BMC Genomics 18 14. (https://doi.org/10.1186/s12864-016-3394-7)
CarrellDTLiuL 2001 Altered protamine 2 expression is uncommon in donors of known fertility, but common among men with poor fertilizing capacity, and may reflect other abnormalities of spermiogenesis. Journal of Andrology 22 604–610. (https://doi.org/10.1002/j.1939-4640.2001.tb02220.x)
CastroLSSiqueiraAFPHamiltonTRSMendesCMVisintinJAAssumpcaoM 2018 Effect of bovine sperm chromatin integrity evaluated using three different methods on in vitro fertility. Theriogenology 107 142–148. (https://doi.org/10.1016/j.theriogenology.2017.11.006)
DasPPBagijnMPGoldsteinLDWoolfordJRLehrbachNJSapetschnigABuhechaHRGilchristMJHoweKLStarkR et al. 2008 Piwi and piRNAs act upstream of an endogenous siRNA pathway to suppress Tc3 transposon mobility in the Caenorhabditis elegans germline. Molecular Cell 31 79–90. (https://doi.org/10.1016/j.molcel.2008.06.003)
de WaalEYamazakiYIngalePBartolomeiMYanagimachiRMcCarreyJR 2012 Primary epimutations introduced during intracytoplasmic sperm injection (ICSI) are corrected by germline-specific epigenetic reprogramming. PNAS 109 4163–4168. (https://doi.org/10.1073/pnas.1201990109)
DeshmukhRSOstrupOOstrupEVejlstedMNiemannHLucas-HahnAPetersenBLiJCallesenHHyttelP 2011 DNA methylation in porcine preimplantation embryos developed in vivo and produced by in vitro fertilization, parthenogenetic activation and somatic cell nuclear transfer. Epigenetics 6 177–187. (https://doi.org/10.4161/epi.6.2.13519)
DominissiniDMoshitch-MoshkovitzSSchwartzSSalmon-DivonMUngarLOsenbergSCesarkasKJacob-HirschJAmariglioNKupiecM et al. 2012 Topology of the human and mouse m6A RNA methylomes revealed by m6A-seq. Nature 485 201–206. (https://doi.org/10.1038/nature11112)
DongHWangYZouZChenLShenCXuSZhangJZhaoFGeSGaoQ et al. 2016 Abnormal methylation of imprinted genes and cigarette smoking: assessment of their association with the risk of male infertility. Reproductive Sciences 24 114–123. (https://doi.org/10.1177/1933719116650755)
GeulaSMoshitch-MoshkovitzSDominissiniDMansourAAKolNSalmon-DivonMHershkovitzVPeerEMorNManorYS et al. 2015 Stem cells. m6A mRNA methylation facilitates resolution of naive pluripotency toward differentiation. Science 347 1002–1006. (https://doi.org/10.1126/science.1261417)
HouwingSKammingaLMBerezikovECronemboldDGirardAvan den ElstHFilippovDVBlaserHRazEMoensCB et al. 2007 A role for Piwi and piRNAs in germ cell maintenance and transposon silencing in Zebrafish. Cell 129 69–82. (https://doi.org/10.1016/j.cell.2007.03.026)
KroppJCarrilloJANamousHDanielsASalihSMSongJKhatibH 2017 Male fertility status is associated with DNA methylation signatures in sperm and transcriptomic profiles of bovine preimplantation embryos. BMC Genomics 18 280. (https://doi.org/10.1186/s12864-017-3673-y)
KumarADumasiaKDeshpandeSBalasinorNH 2017 Direct regulation of genes involved in sperm release by estrogen and androgen through their receptors and coregulators. Journal of Steroid Biochemistry and Molecular Biology 171 66–74. (https://doi.org/10.1016/j.jsbmb.2017.02.017)
KutchyNAVelhoAMenezesESBJacobsenMThibaudeauGWillsRWMouraAKayaAPerkinsAMemiliE 2017 Testis specific histone 2B is associated with sperm chromatin dynamics and bull fertility-a pilot study. Reproductive Biology and Endocrinology 15 59. (https://doi.org/10.1186/s12958-017-0274-1)
LambrotRXuCSaint-PharSChountalosGCohenTPaquetMSudermanMHallettMKimminsS 2013 Low paternal dietary folate alters the mouse sperm epigenome and is associated with negative pregnancy outcomes. Nature Communications 4 2889. (https://doi.org/10.1038/ncomms3889)
LuenseLJWangXSchonSBWellerAHLinShiaoEBryantJMBartolomeiMSCoutifarisCGarciaBA et al. 2016 Comprehensive analysis of histone post-translational modifications in mouse and human male germ cells. Epigenetics and Chromatin 9 24. (https://doi.org/10.1186/s13072-016-0072-6)
NeriFIncarnatoDKrepelovaAParlatoCOlivieroS 2016 Methylation-assisted bisulfite sequencing to simultaneously map 5fC and 5caC on a genome-wide scale for DNA demethylation analysis. Nature Protocols 11 1191–1205. (https://doi.org/10.1038/nprot.2016.063)
OukoLAShantikumarKKnezovichJHaycockPSchnughDJRamsayM 2009 Effect of alcohol consumption on CpG methylation in the differentially methylated regions of H19 and IG-DMR in male gametes: implications for fetal alcohol spectrum disorders. Alcoholism: Clinical and Experimental Research 33 1615–1627. (https://doi.org/10.1111/j.1530-0277.2009.00993.x)
PachecoSEHousemanEAChristensenBCMarsitCJKelseyKTSigmanMBoekelheideK 2011 Integrative DNA methylation and gene expression analyses identify DNA packaging and epigenetic regulatory genes associated with low motility sperm. PLoS ONE 6 e20280. (https://doi.org/10.1371/journal.pone.0020280)
RamonMJimenez-RabadanPGarcia-AlvarezOMaroto-MoralesASolerAJFernandez-SantosMRPerez-GuzmanMDGardeJJ 2014 Understanding sperm heterogeneity: biological and practical implications. Reproduction in Domestic Animals 49 (Supplement 4) 30–36. (https://doi.org/10.1111/rda.12404)
StuppiaLFranzagoMBalleriniPGattaVAntonucciI 2015 Epigenetics and male reproduction: the consequences of paternal lifestyle on fertility, embryo development, and children lifetime health. Clinical Epigenetics 7 120. (https://doi.org/10.1186/s13148-015-0155-4)
TangMCJacobsSAMattiskeDMSohYMGrahamANTranALimSLHudsonDFKalitsisPO’BryanMK et al. 2015 Contribution of the two genes encoding histone variant h3.3 to viability and fertility in mice. PLoS Genetics 11 e1004964. (https://doi.org/10.1371/journal.pgen.1004964)
ViewegMDvorakova-HortovaKDudkovaBWaliszewskiPOtteMOelsBHajimohammadATurleyHSchorschMSchuppeHC et al. 2015 Methylation analysis of histone H4K12ac-associated promoters in sperm of healthy donors and subfertile patients. Clinical Epigenetics 7 31. (https://doi.org/10.1186/s13148-015-0058-4)
WiebeJCSantanaAMedina-RodriguezNHernandezMNovoaJMauricioDWagnerAM & T1DGC 2014 Fertility is reduced in women and in men with type 1 diabetes: results from the Type 1 Diabetes Genetics Consortium (T1DGC). Diabetologia 57 2501–2504. (https://doi.org/10.1007/s00125-014-3376-8)
ZegginiEWeedonMNLindgrenCMFraylingTMElliottKSLangoHTimpsonNJPerryJRRaynerNWFreathyRM 2007 Replication of genome-wide association signals in UK samples reveals risk loci for type 2 diabetes. Science 316 1336–1341. (https://doi.org/10.1126/science.1142364)
ZhangBArunGMaoYSLazarZHungGBhattacharjeeGXiaoXBoothCJWuJZhangC 2012 The lncRNA Malat1 is dispensable for mouse development but its transcription plays a cis-regulatory role in the adult. Cell Reports 2 111–123. (https://doi.org/10.1016/j.celrep.2012.06.003)