Abstract
Several lines of evidence suggest that the presence of the Y chromosome influences DNA methylation of autosomal loci. To better understand the impact of the Y chromosome on autosomal DNA methylation patterns and its contribution to sex bias in methylation, we identified Y chromosome dependent differentially methylated regions (yDMRs) using whole-genome bisulfite sequencing methylation data from livers of mice with different combinations of sex-chromosome complement and gonadal sex. Nearly 90% of the autosomal yDMRs mapped to transposable elements (TEs) and most of them had lower methylation in XY compared to XX or XO mice. Follow-up analyses of four reporter autosomal yDMRs showed that Y-dependent methylation levels were consistent across most somatic tissues but varied in strains with different origins of the Y chromosome, suggesting that genetic variation in the Y chromosome influenced methylation levels of autosomal regions. Mice lacking the q-arm of the Y chromosome (B6.NPYq-2) as well as mice with a loss-of-function mutation in Kdm5d showed no differences in methylation levels compared to wild type mice. In conclusion, the Y-linked modifier of TE methylation is likely to reside on the short arm of Y chromosome and further studies are required to identify this gene.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Mammalian males and females differ in many aspects of their phenotypes from anatomy to behavior. At the molecular level, sexual dimorphism is largely defined by sex differences in gene regulatory networks, and the sex-chromosome complement and gonadal sex are the major factors that shape sex differences in gene regulation (Burgoyne and Arnold 2016; Xu et al. 2002). To date, multiple studies have provided in-depth analyses and demonstrated the wide-ranging impacts of gonadal sex, gonadal sex-steroid hormones, and the pituitary growth hormone on gene expression and epigenetic modifications in the mouse liver (Conforto and Waxman 2012; Lau-Corona et al. 2017; Reizel et al. 2015). In contrast, the sex-chromosome complement appears to be critical for sex-biased gene expression in mouse preimplantation embryos, cultured cells, and the immune system with the X chromosome playing a major role (AlSiraj et al. 2019; Deegan et al. 2019; Engel 2018; Werner et al. 2017; Wijchers et al. 2010). However, the specific contribution of the Y chromosome to sex bias in gene regulation is still not well-understood (Deschepper 2020).
The mammalian Y chromosome is pivotal for sex determination, gonadal differentiation, and spermatogenesis. Several lines of evidence suggest that the Y chromosome also has a role beyond reproductive tissues [(Bellott et al. 2014), reviewed in (Hughes and Page 2015)] and influences regulation of autosomal genes as well as autosomal DNA methylation in somatic cells of non-reproductive organs (Deschepper 2020; Gatev et al. 2021; Ho et al. 2018; Wijchers et al. 2010; Zhuang et al. 2020). Indeed, several Y-linked protein-coding genes with X-linked paralogs that escape X-inactivation have been implicated in gene regulation and are expressed in somatic cells of non-reproductive organs in mice and humans. Y-linked lysine (K)-specific demethylase 5D (KDM5D/Kdm5d) is of particular interest as it encodes a lysine demethylase that demethylates histone H3K4me2/3 at gene promoter regions (Iwase et al. 2007). KDM5D-dependent histone H3K4 demethylation is associated with sex-biased gene expression in mice and sex-biased DNA methylation in humans (Grafodatskaya et al. 2013; Mizukami et al. 2019). Mutations in Kdm5d and its X-linked paralog Kdm5c affect heart development in mice (Kosugi et al. 2020). Another gene of interest is the ubiquitously transcribed tetratricopeptide repeat containing, Y-linked (UTY/Uty). UTY is critical during mouse preimplantation embryo development (Shi et al. 2021; Shpargel et al. 2012).
DEAD box helicase 3, Y-linked (DDX3Y) belongs to the family of ATP-dependent RNA-helicases (Sekiguchi et al. 2004) that are involved in the silencing of foreign DNA sequences, such as transgenes or endogenous retroviruses, in Drosophila (ElMaghraby et al. 2019; Lo et al. 2016) and may have conserved this function in other species. Eukaryotic translation initiation factor 2, subunit 3, structural gene Y-linked (Eif2s3y) is indispensable for spermatogenesis (Matsubara et al. 2015; Mazeyrat et al. 2001; Yamauchi et al. 2014) but also expressed in somatic cells (Werner et al. 2017). Moreover, four mouse Y-linked genes, zinc finger protein 2, Y-linked (Zfy2), Uty, Ddx3y, ubiquitin-activating enzyme, Chr Y (Uba1y) are expressed in embryonic stem cells and cannot be excluded as potential modifiers of DNA methylation or gene regulation at the very early stages of embryonic development (Werner et al. 2017).
To date, several studies have examined the influence of sex, sex-chromosome complement, gonadal sex hormone- and pituitary hormone-signaling pathways on DNA methylation in the mouse (Hao and Waxman 2021; McCormick et al. 2017; Reizel et al. 2015; Zhuang et al. 2020). However, most of these works did not analyze the impacts of X-dosage and Y chromosome separately. The objective of the present study was to evaluate the effects of the Y chromosome on autosomal DNA methylation. Our recent works have demonstrated that the presence of the Y chromosome may influence DNA methylation levels at autosomal loci in human fibroblasts and mouse livers (Ho et al. 2018; Zhuang et al. 2020). To explore the contribution of the Y chromosome to sex-biased DNA methylation, we analyzed DNA methylation in mice with different combinations of phenotypic sex and sex-chromosome complement using whole-genome bisulfite sequencing (WGBS) methylation analysis and identified Y-dependent differentially methylated regions (yDMRs) on autosomes. We find that most mouse yDMRs are located within transposable elements (TEs) and genetic variation in the Y chromosome modifies their DNA methylation levels.
Materials and methods
Mice and crosses
C57BL/6J mice were purchased from the Jackson Laboratory (Bar Harbor, Maine, USA).
B6.YTIR consomic mice were maintained in our colony (TT) by breeding of B6.YTIR males to C57BL/6J females. Sex-reversed females with two ovaries (XY.FT), true hermaphrodites with unilateral testis and contralateral ovary (XY.HT), and males with two testes (XY.MT), as well as XX females (XX.FT) were produced by crosses between C57BL/6J females and consomic males. “T” denotes mice originating from the crosses with B6.YTIR males. Liver, lung, heart, spleen, brain, tail, and testes were collected from adult mice at eight and 16 weeks of age and used for DNA or RNA extraction. Sex phenotypes were determined based on genital and gonadal phenotypes and were concordant with secondary sex characteristics. Gonadal sex of adult mice was confirmed at the time of organ collection.
B6.C3H/HeSn-Paf mice (referred to as Paf from this point on) were generated by backcrossing C3H/HeSn-Paf/J carriers of the patchy fur (Paf) mutation purchased from the Jackson Laboratory to C57BL/6J mice for several generations. In each generation, males that carried the Paf mutation were identified based on their hair loss phenotype and crossed to C57BL/6J females. Female offspring from these crosses were genotyped using RT-PCR for the Xist gene, which is expressed in XX females but not in XO females as described in (Alton et al. 2008). Liver samples from 8-week old XO females (XO.F) and their XXPaf female (XXPaf.F) littermates from N6 and N7 generations were collected and used for DNA extraction.
Kdm5d knock-out mice carry a 2-nucleotide deletion in exon 1, which results in a frameshift mutation and premature termination of protein translation (Kosugi et al. 2020).
B6.NPYq-2 (XY*XSxra) males have a single X chromosome and Tp(Y)1CtSxr−a (Cattanach et al. 1971) attached distal to the Y*X PAR. These mice were first produced by intracytoplasmic sperm injection (ICSI) with sperm from XSxraY*X males on MF1 [National Institute for Medical Research (NIMR) colony] genetic background (Yamauchi et al. 2009), and then backcrossed to C57BL/6J for more than ten generations and maintained by ICSI. The sex reversal factor Sxra originates from an RIII strain Y chromosome.
DNA from different laboratory mouse strains was used for genotyping and pyrosequencing methylation analyses. F1 (C57BL/6J female x FVB/NJ male) mice (referred to as B6FVBF1) were generated in our mouse colony. The list of all mouse strains/crosses used in the study is provided in Table S1. The genotypes of the Y chromosome were inferred from the Mouse genomes project (MGP) data, when possible.
For most experiments, organ collection was performed on the same time of the day at Zeitgeber time (ZT) 7, i.e., seven hours after lights were switched on. All procedures were conducted in accordance with the guidelines set by the Canadian Council on Animal Care and were approved by the Animal Care Committee of the McGill University Health Center (Montreal, Quebec, Canada).
DNA extraction from mouse organs and genotyping
DNA from mouse organs was extracted using a standard proteinase K phenol/chloroform procedure or by QIAamp Fast DNA Tissue Kit (Qiagen, NL).
Genetic sex was confirmed using PCR for Y-linked genes Sry (Table S2) or Zfy (as described in Amleh et al. 2000). Other genotypes were determined by PCR followed by Sanger sequencing to detect single nucleotide polymorphisms in selected Y chromosomal genes (Tables S1, S2). Sanger sequencing was done at the CES Genome Quebec (Montreal, QC, Canada).
Detection of differentially methylated regions (DMRs) using WGBS data
Liver WGBS data from our previous study were used to identify differentially methylated regions (DMRs) (Zhuang et al. 2020). Differential methylation was detected using DSS (v2.32.0) (Park and Wu 2016) and methylKit (v1.10.0) (Akalin et al. 2012). CpG sites with low coverage (sequencing depth < 10) and that overlapped with common polymorphisms (SNP database 142) were excluded. Also, CpG sites with coverage > 500X were removed to account for PCR bias. DSS applied a 500-bp (default) smoothing window to estimate methylation levels of CpG sites and required a minimum length of 50 bp and containing at least three CpG sites to call a DMR. We also estimated DMRs with methylKit, with a tiling setting of 300 bp in length and 300 bp step-size windows. Differentially methylated tiles were detected with q-value < 0.05. For both DSS and methylKit, we only kept DMRs with methylation difference > 20%. We applied default parameters in both tools unless specified. We used the union of DMR results from both tools as our final list and DMRs from DSS were kept in the case of overlaps.
Primer design and pyrosequencing methylation analysis
Primers for pyrosequencing methylation assays of DMRs were designed using the PyroMark Assay Design 2.0 Software (Qiagen, NL). For repetitive regions, we could not anchor primers to non-repetitive flanking sequences because the closest non-repetitive regions were farther than two kb from the differentially methylated CG sites. Hence, both PCR primers were located within the repeat and, in principle, could amplify more than one copy of the transposon. The list of primers for pyrosequencing methylation assays is provided in Table S3.
250 to 1000 ng of DNA per sample were treated with sodium bisulfite using EpiTect Bisulfite Kit (Qiagen, NL). Pyrosequencing was carried out using the PyroMark Q24 Advanced platform and PyroMark Q24 Advanced CpG Reagents (Qiagen, NL). Results were analyzed using the PyroMark Q24 Advanced software (Qiagen, NL).
RNA extraction and gene expression analysis
RNA was extracted using TRIzol Reagent (Thermo Fisher Scientific, MA, US) according to the manufacturer’s instructions and followed by purification using the RNeasy MinElute Cleanup Kit (Qiagen, NL). CDNA was synthesized using one μg of RNA, oligo dT 12–18 primers and Moloney Murine Leukemia Virus reverse transcriptase (M-MuLV RT) (Thermo Fisher Scientific, MA, US). Quantitative RT-PCR (qPCR) was performed using Power SYBR Green PCR master mix (Thermo Fisher Scientific, MA, USA) and Eco Real-Time PCR System (Illumina, CA, USA). Gene expression levels were normalized to the housekeeping gene ribosomal protein L19 (Rpl19). Primers for expression analysis were designed using Primer3 software and checked with Bisearch (http://bisearch.enzim.hu) and IDT OligoAnalyzer (https://www.idtdna.com/pages/tools/oligoanalyzer). List of primers is provided in Table S2.
Uty, Ddx3y, Eif2s3y, and Kdm5d expression data were retrieved from our previously published RNA-seq results from the livers of C57BL/6N males and B6.YTIR mice (AlOgayil et al. 2021; Zhuang et al. 2020).
Statistical analysis
Non-paired t-test, one- or two-way ANOVA tests followed by multiple comparisons (GraphPad Prizm 9.0.1 software) were used to determine the statistical significance of pyrosequencing methylation and RT-qPCR data.
Results
Y-dependent autosomal DNA methylation in the mouse
To identify DMRs that were associated with the presence of the Y chromosome (yDMRs) in the mouse liver, we analyzed WGBS data from females with different sex-chromosome complements, i.e., XX females (XX.FT, n = 3), sex-reversed XY females (XY.FT, n = 3), and females with monosomy X (XO.F, n = 3), as well as males (XY.MT, n = 3) (as described in Zhuang et al. 2020, data available from GSE217666). The WGBS data were filtered to exclude regions of low coverage and common polymorphisms (see Methods and Zhuang et al. 2020). After filtering, an average of about one third (15 million) of the 43822587 CGs sites of the reference mouse genome were available for analysis.
Autosomal DNA methylation levels from the WGBS data were contrasted in three comparisons: XY.FT vs XO.F, XX.FT vs XY.FT, and XX.FT vs XY.MT using two different analysis tools, DSS and methylKit (Zhuang et al. 2020). These tools detect regions that may contain from one to multiple differentially methylated cytosines (DMCs). The XX.FT vs XY.MT comparison identifies both sex-chromosome complement dependent as well as gonadal-sex dependent DMCs and DMRs. The XX.FT vs XY.FT comparison identifies DMCs and DMRs that are sensitive to the X chromosome dosage and/or the presence of the Y chromosome, whereas the XY.FT vs XO.F comparison detects Y chromosome dependent DMRs. The XO.F mice are not congenic and carry C3H alleles on the C57BL/6J genetic background. Hence, the XY.FT vs XO.F comparison would detect regions that are sensitive to variation in the genetic background too. To circumvent the impact of genetic variation and reduce the chance of false-positive results, Y-dependent DMRs (yDMRs) were identified as those that overlapped between three comparisons, the XY.FT vs XO.F, XX.FT vs XY.FT, and XX.FT vs XY.MT. In total, 785 autosomal DMRs were found in the XY.FT vs XO.F comparison, 802 DMRs—in the XX.FT vs XY.FT, and 2854—in the XX.FT vs XY.MT comparison. Nineteen autosomal DMRs (two with higher and 17 with lower methylation in XY mice) were common between the three comparisons and were defined as yDMRs. The mean length of autosomal yDMRs was 346.5 bp, with maximum length of 892 and minimum length of 137 bp. The average GC content of yDMRs was 42%. Seventeen of these (89%) overlapped with transposable elements (TEs), eight LINEs, one SINE, and eight LTRs and only two resided in non-repetitive regions (Table 1). All yDMR-associated LINE1 belong to the families of young and transcriptionally active L1_Md transposons, and three of these yDMRs (on chromosomes 12, 14, and 17) map to the promoter region of the open reading frame 1 (ORF1) of L1_Md.
For five yDMRs, including two yDMRs located within the ERVB4_1I_MM-int (referred to as ERVB4-proximal and ERVB4-distal from this point on) located in intron 1 of the agouti gene on chromosome 2, the yDMR located within the intracisternal A particle IAPz-int repeat on chromosome 16, and two in the non-repetitive regions of chromosomes 2 and 6, we were able to design pyrosequencing primers (Tables 1, S3; Fig. 1A–G). Pyrosequencing methylation analysis in independently collected samples of XX.FT, XY.FT, XY.HT, XY.MT, XXPaf.F, and XO.F livers showed that methylation of the ERVB4, Ch6qA1, and Caprin1 yDMRs was Y chromosome dependent, as methylation levels were different in mice that carried the Y chromosome, independent of their gonadal sex (Fig. 1H–K). The IAPz-int region on chromosome 16 showed no significant differences in methylation levels between XX.FT and XY.FT females and was not used in further experiments.
Y-dependent DNA methylation is present across different mouse organs
To determine whether the effect of the Y chromosome on methylation was limited to liver, we tested yDMR methylation in other organs from the same mice: lung, heart, spleen, brain, tail, and testis. These organs represent the three germ layers of the mouse embryo with brain originating from the ectoderm, heart and spleen from the mesoderm, and liver and lung from the endoderm. The tail consists of skin (ectoderm) and bone (mesoderm), whereas adult testis consists of somatic (mesoderm) and germ cells. The ERVB4-proximal yDMR had higher methylation in most organs of XY compared to XX animals, whereas no differences between males and females were found in the spleen (Fig. 1L). In the testis and tail, methylation levels were lower than in other organs. The ERVB4-distal, Ch6qA1, and Caprin1 yDMRs had lower methylation in all organs of XY compared to XX mice (Fig. 1M–O).
Genetic variation in the Y chromosome influences yDMR methylation levels
Genetic variation may influence DNA methylation levels in cis, by affecting transcription factor binding sites, or trans through genetic variants in transcription factors or epigenetic modifiers that alter their ability to bind DNA or their binding specificity. To evaluate the influence of genetic variation on Y-dependent methylation, we first examined yDMR methylation levels in females and males from seven laboratory mouse strains, C57BL/6J, C57BL/6N, C3H (C3H/HeJ and C3H/HeH), DBA1/J, MOLF/EiJ, CAST/EiJ, FVB/NCrl, as well as F1(C57BL/6J female x FVB/NJ male) mice that carry the FVB/NJ Y chromosome. The subspecific origin of the Y chromosome in each of these strains was inferred from the Mouse genomes project (MGP) data and data from (Chang et al. 2017; Hughes and Page 2015; Morgan et al. 2015) (Table 2). To confirm or establish the genetic origin of the Y chromosome for strains not represented in the above sets of data, we genotyped male mice for four Y-chromosomal SNPs (Table S1).
Sex differences in methylation of the ERVB4-proximal yDMR were found in C3H, B6FVBF1, and MOLF/EiJ mice, but not C57BL/6J, C57BL/6N, and DBA/1J mice. Sex differences in methylation of the ERVB4-distal yDMR were found in most strains (Fig. 2). Variation in methylation among XX samples from different strains reflected the effects of the genetic background in the absence of the Y chromosome, whereas variation among XY samples reflected the influence of genetic background including the Y chromosome (Fig. 2B). The ERVB4 yDMRs showed strain-specific variation in methylation in females. There were also significant differences between males carrying the Y of musculus vs domesticus origin (Fig. 2B; Table S1). The caveat here is that the ERVB4 yDMRs map to the ERVB4_1_I_MM-int copy of the betaretrovirus 4 transposon in the intron 1 of the agouti gene in the C57BL/6J reference genome (Fig. 1A). Insertion of this retrotransposon is responsible for the black (non-agouti) coat color in C57BL/6J mice and is present in several other strains including C57BL/6N and DBA/1J, but absent in the C3H, FVB/NJ, MOLF/EiJ, and CAST/EiJ mice (Tanave et al. 2019) (Table 2). Bisulfite conversion of DNA that is required for WGBS and pyrosequencing methylation assays reduces the complexity of the DNA sequence, and it is likely that our primers amplify more than one copy of the ERVB4. If our primers amplified only the chromosome 2 target region, we would have observed PCR amplification in B6.YTIR, C57BL/6J, C57BL/6N, DBA/1J, and B6FVBF1 mice but not in C3H, CAST/EiJ, FVB/NCrl, or MOLF/EiJ mice. On one hand, the ERVB4 targets were amplified in all samples. On the other hand, variation in methylation levels that may reflect presence or absence of the ERVB4 copy in the agouti locus was observed in females from different strains. These results demonstrate that our yDMR assays interrogate several copies of the transposon, including the target sequence from the agouti locus in those strains that carry it.
Methylation of the Ch6qA1 yDMR showed sex bias in FVB/NCrl, B6FVBF1, MOLF/EiJ, and C3H mice (Fig. 2A). FVB/NCrl, B6FVBF1 and MOLF/EiJ males had lower methylation levels than females. The bias was reversed in the C3H mice who had higher methylation in males than females, albeit the differences did not exceed 10%. The B6.YTIR male mice had the lowest methylation levels at the Ch6qA1 yDMR compared to males from other strains (Fig. 2A–B). The Caprin1 yDMR showed sex bias in methylation in FVB/NCrl, B6FVBF1, MOLF/EiJ, and CAST/EiJ mice.
Y-linked gene candidates
We hypothesized that a Y-linked epigenetic modifier or transcription factor(s) was responsible for the Y effect on TE methylation and that genetic variation affecting its function would lead to loss/gain of methylation at yDMRs in XY mice. YDMRs are present in mouse organs derived from the three different germ layers. This may reflect the ubiquitous expression of the Y-linked gene responsible for yDMR methylation or, alternatively, point to a gene that is expressed early in embryonic development driving an early onset of Y-dependent methylation patterns before the germ layers are differentiated. Four Y-linked protein-coding genes residing on the short arm of the mouse Y chromosome, Kdm5d, Uty, Ddx3y, and Eif2s3y are expressed in different organs of our mice (Zhuang et al. 2020) and Bauermeister et al. (paper in preparation). If yDMRs are established independently in different somatic cell types, one of these commonly expressed genes must be involved. Uty and Kdm5d, but not Eif2s3y or Ddx3y harbor non-synonymous coding polymorphisms between the FVB/NJ and C57BL/6J strains (Mouse Genomes Project, https://www.sanger.ac.uk/data/mouse-genomes-project). The UTY amino acid variation D615N maps outside of the protein domains with known functions (https://beta.uniprot.org/uniprotkb/P79457), whereas the KDM5D variation G213A maps to the disordered domain of the protein (https://beta.uniprot.org/uniprotkb/Q62240). Due to its demonstrated effect on methylation, KDM5D was the most attractive gene candidate. We tested yDMR methylation in the livers and spleens of Kdm5d mutant and control mice but found no differences in methylation levels (Fig. 3A). It is worth noting that the knock-out has been generated in the ICR strain of mice, which carry the Y chromosome of domesticus origin (Table S2). We next compared the RNA levels of Kdm5d in the livers of adult C57BL/6J and B6.YTIR mice using RT-qPCR. No differences between the two strains were found (Fig. S1A). Hence, we conclude that KDM5D is unlikely to be responsible for the Y-dependent methylation levels.
Next, we asked if the expression levels of the other ubiquitously expressed gene candidates were different between mice with M. musculus musculus and M.musculus domesticus Y chromosomes. Expression levels of Eif2s3y, Uty, and Ddx3y were retrieved from our RNA-seq data for adult C57BL/6N (AlOgayil et al. 2021) and B6.YTIR livers (Zhuang et al. 2020) and normalized to Kdm5d RNA levels. Ddx3y and Eif2s3y expression levels were significantly lower, whereas Uty levels were significantly higher in B6.YTIR compared to C57BL/6N livers (Fig. S1B).
The q-arm of the mouse Y chromosome largely consists of amplified repeat units including protein-coding genes that are expressed in the testis and multiple long non-coding RNAs (Comptour et al. 2014; Hughes and Page 2015; Soh et al. 2014). The ampliconic structure of the Yq makes it prone to intrachromosomal rearrangements that poses a challenge for Yq mapping in different strains (Morgan and Pardo-Manuel de Villena 2017). To determine whether the Yq arm harbored gene(s) that influenced methylation, we assayed yDMR methylation in the testes of consomic B6.NPYq-2 mice lacking the Yq arm and several copies of Yp-arm encoded Rbmy gene. The remaining Yp-arm was genotyped and its M. musculus musculus origin was confirmed (Table S1). No difference in methylation levels between B6.NPYq-2 mice and wild type C57BL/6J mice were found (Fig. 3).
Discussion
To explore the contribution of the mouse Y chromosome to sexual dimorphism in DNA methylation in somatic cells, we conducted a WGBS data analysis followed by targeted analysis of methylation at several autosomal loci that showed different methylation levels in XY vs XO or XX mice. We show that DNA methylation levels at several loci are associated with the presence of the Y chromosome and independent of the gonadal sex. Firstly, sex-reversed females with two ovaries, hermaphrodites, and males with two testes have the same methylation levels at yDMRs, and secondly, phenotypic males that carry an Y chromosome of M. musculus musculus origin have yDMR methylation levels that are not different from those of phenotypic females.
The majority of yDMRs overlap with TEs and most of the TE-related yDMRs have lower methylation in XY compared to XX mice (Table 1). Previous studies have established that gonadal sex has a major impact on DNA methylation in mouse liver (AlOgayil et al. 2021; Hao and Waxman 2021; McCormick et al. 2017; Reizel et al. 2015; Zhuang et al. 2020), whereas the X chromosome dosage predominantly affects DNA methylation of X-linked loci (Zhuang et al. 2020). Here, we report that the Y chromosome influences DNA methylation of certain TEs. We demonstrate that genetic variation in both, the target loci and the Y chromosome itself, must be taken into account when interpreting the results of methylation analyses.
The yDMRs are not liver-specific and are present in different organs. This implies that in contrast to the tissue-specificity of gonadal-sex-dependent methylation (McCormick et al. 2017), the effect of the Y chromosome is not tissue-specific. Silencing of foreign DNA sequences, such as endogenous viruses or transgenes, is a highly conserved fundamental function of DNA methylation that is pivotal for genome integrity [(Conklin et al. 1982; Drahovsky et al. 1979), reviewed in (Bourque et al. 2018; Déléris et al. 2021; Naumova 2013)]. Demethylation and reactivation of endogenous retroviruses is necessary for zygotic genome activation in the early embryo but becomes detrimental in somatic or male germ cells as it facilitates retroviral transposition and hence mutagenesis (Beraldi et al. 2006; Bourc’his and Bestor 2004, reviewed in Jansz 2019). Interestingly, data from human studies support the link between the Y chromosome and TE methylation as certain human cancers, including hepatocellular carcinoma and lung cancer, are characterized by both loss of the Y chromosome and demethylation of TEs in tumor cells (Babaian and Mager 2016; Center et al. 1993; Iskow et al. 2010; Park et al. 2006; Qin et al. 2019; Shitani et al. 2012; Zheng et al. 2019). Examining the associations between human Y-haplotypes, TE methylation, and risks of developing certain diseases with male preponderance, will determine if in humans certain genetic variants of the Y chromosome contribute to both TE demethylation and elevated risk of disease.
Based on our data, we speculate that the mouse Y chromosome harbors a gene(s) that modifies TE methylation levels. Since most yDMRs overlapping with TEs have lower methylation in XY mice compared to XX mice, but XO and XX mice have similar methylation levels, it is a reasonable conjecture that the Y-linked gene product protects TEs from methylation. In an attempt to narrow down the list of gene candidates, we tested mice with a mutation in the Kdm5d gene, which was the best gene candidate based on its function, as well as mice lacking the whole Yq arm. These mutations did not affect methylation levels of yDMRs, suggesting that Kdm5d or the Yq genes did not contribute to Y-dependent variation in methylation. In principle, regulatory variation in the Y chromosome may influence the expression levels of the putative Y-linked modifier of methylation, which in turn may affect methylation at certain loci. Expression levels of three of the four ubiquitously expressed Y-linked genes, Eif2s3y, Uty, and Ddx3y differ between B6.YTIR and C57BL/6 mice. Therefore, at this point, we cannot exclude any of these three genes as potential contributors to Y-dependent methylation. It is also worth noting that these Y-linked genes share common regulatory elements: a gene-trap mutation in intron 4 of Uty perturbs transcription of the two neighboring genes, Ddx3y and Eif2s3y, in somatic tissues (Deschepper 2020). Therefore, in future studies targeting exonic mutations that affect the functions of each protein separately may be a more informative approach for identifying the Y-linked modifier of TE methylation.
Study limitations
One of the major limitations of this study is the lack of complete sequence assemblies of the Y chromosomes from different mouse strains. Sequencing assemblies are hindered by the very structure of the Y chromosome and its ampliconic nature. This adds to the challenges of narrowing down the list of Y-linked gene candidates that contribute to Y-dependent methylation and mapping Y-linked copies of TEs. Moreover, we cannot rule completely out the possibility that our analyses capture TE copies that are located on the Y chromosome itself in addition to autosomal TEs.
The advantage of using WGBS methylation analysis as a starting point is that it permits a relatively unbiased analysis of the mouse methylome and is expected to provide an accurate view of the genomic distribution and repertoire of Y-dependent differentially methylated regions. The caveat here is that the reduced complexity of the DNA sequence after bisulfite conversion may be particularly challenging for mapping repetitive elements. Our data show that after filtering low coverage reads and excluding CGs that overlap with common mouse SNPs, only about one third of the genomic CGs remain available for analysis. We cannot rule out the possibility that the number of TEs containing yDMRs is larger than the one found in our study. Conversely, the possibility of false-positive results is also higher. Therefore, alternative methodologies of genome-wide methylation analysis that do not require bisulfite conversion will provide a more accurate understanding of the impact of the Y chromosome on TE methylation in general and permit a more detailed characterization of its targets.
References
Akalin A, Kormaksson M, Li S, Garrett-Bakelman FE, Figueroa ME, Melnick A, Mason CE (2012) methylKit: a comprehensive R package for the analysis of genome-wide DNA methylation profiles. Genome Biol 13:R87
AlOgayil N, Bauermeister K, Galvez JH, Venkatesh VS, Zhuang QK, Chang ML, Davey RA, Zajac JD, Ida K, Kamiya A, Taketo T, Bourque G, Naumova AK (2021) Distinct roles of androgen receptor, estrogen receptor alpha, and BCL6 in the establishment of sex-biased DNA methylation in mouse liver. Sci Rep 11:13766
AlSiraj Y, Chen X, Thatcher SE, Temel RE, Cai L, Blalock E, Katz W, Ali HM, Petriello M, Deng P, Morris AJ, Wang X, Lusis AJ, Arnold AP, Reue K, Thompson K, Tso P, Cassis LA (2019) XX sex chromosome complement promotes atherosclerosis in mice. Nat Commun 10:2631
Alton M, Lau MP, Villemure M, Taketo T (2008) The behavior of the X- and Y-chromosomes in the oocyte during meiotic prophase in the B6.Y(TIR)sex-reversed mouse ovary. Reproduction (Cambridge, England) 135:241–252
Amleh A, Smith L, Chen H, Taketo T (2000) Both nuclear and cytoplasmic components are defective in oocytes of the B6.Y(TIR) sex-reversed female mouse. Dev Biol 219:277–286
Babaian A, Mager DL (2016) Endogenous retroviral promoter exaptation in human cancer. Mob DNA 7:24
Bellott DW, Hughes JF, Skaletsky H, Brown LG, Pyntikova T, Cho TJ, Koutseva N, Zaghlul S, Graves T, Rock S, Kremitzki C, Fulton RS, Dugan S, Ding Y, Morton D, Khan Z, Lewis L, Buhay C, Wang Q, Watt J, Holder M, Lee S, Nazareth L, Alföldi J, Rozen S, Muzny DM, Warren WC, Gibbs RA, Wilson RK, Page DC (2014) Mammalian Y chromosomes retain widely expressed dosage-sensitive regulators. Nature 508:494–499
Beraldi R, Pittoggi C, Sciamanna I, Mattei E, Spadafora C (2006) Expression of LINE-1 retroposons is essential for murine preimplantation development. Mol Reprod Dev 73:279–287
Bourc’his D, Bestor TH (2004) Meiotic catastrophe and retrotransposon reactivation in male germ cells lacking Dnmt3L. Nature 431:96–99
Bourque G, Burns KH, Gehring M, Gorbunova V, Seluanov A, Hammell M, Imbeault M, Izsvák Z, Levin HL, Macfarlan TS, Mager DL, Feschotte C (2018) Ten things you should know about transposable elements. Genome Biol 19:199
Burgoyne PS, Arnold AP (2016) A primer on the use of mouse models for identifying direct sex chromosome effects that cause sex differences in non-gonadal tissues. Biol Sex Differ 7:68
Cattanach BM, Pollard CE, Hawker SG (1971) Sex-reversed mice: XX and XO males. Cytogenetics 10:318–337
Center R, Lukeis R, Vrazas V, Garson OM (1993) Y chromosome loss and rearrangement in non-small-cell lung cancer. Int J Cancer 55:390–393
Chang PL, Kopania E, Keeble S, Sarver BAJ, Larson E, Orth A, Belkhir K, Boursot P, Bonhomme F, Good JM, Dean MD (2017) Whole exome sequencing of wild-derived inbred strains of mice improves power to link phenotype and genotype. Mamm Genome 28:416–425
Comptour A, Moretti C, Serrentino ME, Auer J, Ialy-Radio C, Ward MA, Touré A, Vaiman D, Cocquet J (2014) SSTY proteins co-localize with the post-meiotic sex chromatin and interact with regulators of its expression. FEBS J 281:1571–1584
Conforto TL, Waxman DJ (2012) Sex-specific mouse liver gene expression: genome-wide analysis of developmental changes from pre-pubertal period to young adulthood. Biol Sex Differ 3:9
Conklin KF, Coffin JM, Robinson HL, Groudine M, Eisenman R (1982) Role of methylation in the induced and spontaneous expression of the avian endogenous virus ev-1: DNA structure and gene products. Mol Cell Biol 2:638–652
Deegan DF, Karbalaei R, Madzo J, Kulathinal RJ, Engel N (2019) The developmental origins of sex-biased expression in cardiac development. Biol Sex Differ 10:46
Déléris A, Berger F, Duharcourt S (2021) Role of polycomb in the control of transposable elements. Trends Genet 37:882–889
Deschepper CF (2020) Regulatory effects of the Uty/Ddx3y locus on neighboring chromosome Y genes and autosomal mRNA transcripts in adult mouse non-reproductive cells. Sci Rep 10:14900
Drahovsky D, Boehm TL, Kreis W (1979) Distribution pattern and enzymic hypermethylation of inverted repetitive DNA sequences in P815 mastocytoma cells. Biochim Biophys Acta 563:28–35
ElMaghraby MF, Andersen PR, Pühringer F, Hohmann U, Meixner K, Lendl T, Tirian L, Brennecke J (2019) A heterochromatin-specific RNA export pathway facilitates piRNA production. Cell 178:964-979.e920
Engel N (2018) Sex differences in early embryogenesis: inter-chromosomal regulation sets the stage for sex-biased networks. BioEssays 40:1800073
Gatev E, Inkster AM, Negri GL, Konwar C, Lussier AA, Skakkebaek A, Sokolowski MB, Gravholt CH, Dunn EC, Kobor MS, Aristizabal MJ (2021) Autosomal sex-associated co-methylated regions predict biological sex from DNA methylation. Nucleic Acids Res 49:9097–9116
Grafodatskaya D, Chung BH, Butcher DT, Turinsky AL, Goodman SJ, Choufani S, Chen YA, Lou Y, Zhao C, Rajendram R, Abidi FE, Skinner C, Stavropoulos J, Bondy CA, Hamilton J, Wodak S, Scherer SW, Schwartz CE, Weksberg R (2013) Multilocus loss of DNA methylation in individuals with mutations in the histone H3 lysine 4 demethylase KDM5C. BMC Med Genom 6:1
Hao P, Waxman DJ (2021) STAT5 regulation of sex-dependent hepatic CpG methylation at distal regulatory elements mapping to sex-biased genes. Mol Cell Biol 41:e00166-e120
Ho B, Greenlaw K, Al Tuwaijri A, Moussette S, Martinez F, Giorgio E, Brusco A, Ferrero GB, Linhares ND, Valadares ER, Svartman M, Kalscheuer VM, Rodriguez Criado G, Laprise C, Greenwood CMT, Naumova AK (2018) X chromosome dosage and presence of SRY shape sex-specific differences in DNA methylation at an autosomal region in human cells. Biol Sex Differ 9:10
Hughes JF, Page DC (2015) The biology and evolution of mammalian Y chromosomes. Annu Rev Genet 49:507–527
Iskow RC, McCabe MT, Mills RE, Torene S, Pittard WS, Neuwald AF, Van Meir EG, Vertino PM, Devine SE (2010) Natural mutagenesis of human genomes by endogenous retrotransposons. Cell 141:1253–1261
Iwase S, Lan F, Bayliss P, de la Torre-Ubieta L, Huarte M, Qi HH, Whetstine JR, Bonni A, Roberts TM, Shi Y (2007) The X-linked mental retardation gene SMCX/JARID1C defines a family of histone H3 lysine 4 demethylases. Cell 128:1077–1088
Jansz N (2019) DNA methylation dynamics at transposable elements in mammals. Essays Biochem 63:677–689
Kosugi M, Otani M, Kikkawa Y, Itakura Y, Sakai K, Ito T, Toyoda M, Sekita Y, Kimura T (2020) Mutations of histone demethylase genes encoded by X and Y chromosomes, Kdm5c and Kdm5d, lead to noncompaction cardiomyopathy in mice. Biochem Biophys Res Commun 525:100–106
Lau-Corona D, Suvorov A, Waxman DJ (2017) Feminization of male mouse liver by persistent growth hormone stimulation: activation of sex-biased transcriptional networks and dynamic changes in chromatin states. Mol Cell Biol 37:e00301-17
Lo PK, Huang YC, Poulton JS, Leake N, Palmer WH, Vera D, Xie G, Klusza S, Deng WM (2016) RNA helicase Belle/DDX3 regulates transgene expression in drosophila. Dev Biol 412:57–70
Matsubara Y, Kato T, Kashimada K, Tanaka H, Zhi Z, Ichinose S, Mizutani S, Morio T, Chiba T, Ito Y, Saga Y, Takada S, Asahara H (2015) TALEN-mediated gene disruption on Y chromosome reveals critical role of EIF2S3Y in mouse spermatogenesis. Stem Cells Dev 24:1164–1170
Mazeyrat S, Saut N, Grigoriev V, Mahadevaiah SK, Ojarikre OA, Rattigan A, Bishop C, Eicher EM, Mitchell MJ, Burgoyne PS (2001) A Y-encoded subunit of the translation initiation factor Eif2 is essential for mouse spermatogenesis. Nat Genet 29:49–53
McCormick H, Young PE, Hur SSJ, Booher K, Chung H, Cropley JE, Giannoulatou E, Suter CM (2017) Isogenic mice exhibit sexually-dimorphic DNA methylation patterns across multiple tissues. BMC Genom 18:966
Mizukami H, Kim JD, Tabara S, Lu W, Kwon C, Nakashima M, Fukamizu A (2019) KDM5D-mediated H3K4 demethylation is required for sexually dimorphic gene expression in mouse embryonic fibroblasts. J Biochem 165:335–342
Morgan AP, Fu CP, Kao CY, Welsh CE, Didion JP, Yadgary L, Hyacinth L, Ferris MT, Bell TA, Miller DR, Giusti-Rodriguez P, Nonneman RJ, Cook KD, Whitmire JK, Gralinski LE, Keller M, Attie AD, Churchill GA, Petkov P, Sullivan PF, Brennan JR, McMillan L, Pardo-Manuel de Villena F (2015) The mouse universal genotyping array: from substrains to subspecies. G3 (Bethesda) 6:263–279
Morgan AP, Pardo-Manuel de Villena F (2017) Sequence and structural diversity of mouse Y chromosomes. Mol Biol Evol 34:3186–3204
Naumova AK (2013) Transgenerational epigenetic effects and complex inheritance patterns. In: Naumova AK, Greenwood CMT (eds) Epigenetics and complex traits. Springer, New York, pp 107–129
Park Y, Wu H (2016) Differential methylation analysis for BS-seq data under general experimental design. Bioinformatics (Oxford, England) 32:1446–1453
Park SJ, Jeong SY, Kim HJ (2006) Y chromosome loss and other genomic alterations in hepatocellular carcinoma cell lines analyzed by CGH and CGH array. Cancer Genet Cytogenet 166:56–64
Qin N, Li N, Wang C, Pu Z, Ma Z, Jin G, Zhu M, Dai M, Hu Z, Ma H, Shen H (2019) Association of mosaic loss of chromosome Y with lung cancer risk and prognosis in a Chinese population. J Thorac Oncol 14:37–44
Reizel Y, Spiro A, Sabag O, Skversky Y, Hecht M, Keshet I, Berman BP, Cedar H (2015) Gender-specific postnatal demethylation and establishment of epigenetic memory. Genes Dev 29:923–933
Sekiguchi T, Iida H, Fukumura J, Nishimoto T (2004) Human DDX3Y, the Y-encoded isoform of RNA helicase DDX3, rescues a hamster temperature-sensitive ET24 mutant cell line with a DDX3X mutation. Exp Cell Res 300:213–222
Shi B, Li W, Song Y, Wang Z, Ju R, Ulman A, Hu J, Palomba F, Zhao Y, Le JP, Jarrard W, Dimoff D, Digman MA, Gratton E, Zang C, Jiang H (2021) UTX condensation underlies its tumour-suppressive activity. Nature 597:726–731
Shitani M, Sasaki S, Akutsu N, Takagi H, Suzuki H, Nojima M, Yamamoto H, Tokino T, Hirata K, Imai K, Toyota M, Shinomura Y (2012) Genome-wide analysis of DNA methylation identifies novel cancer-related genes in hepatocellular carcinoma. Tumour Biol 33:1307–1317
Shpargel KB, Sengoku T, Yokoyama S, Magnuson T (2012) UTX and UTY demonstrate histone demethylase-independent function in mouse embryonic development. PLoS Genet 8:e1002964
Soh YQ, Alföldi J, Pyntikova T, Brown LG, Graves T, Minx PJ, Fulton RS, Kremitzki C, Koutseva N, Mueller JL, Rozen S, Hughes JF, Owens E, Womack JE, Murphy WJ, Cao Q, de Jong P, Warren WC, Wilson RK, Skaletsky H, Page DC (2014) Sequencing the mouse Y chromosome reveals convergent gene acquisition and amplification on both sex chromosomes. Cell 159:800–813
Tanave A, Imai Y, Koide T (2019) Nested retrotransposition in the East Asian mouse genome causes the classical nonagouti mutation. Commun Biol 2:283
Werner RJ, Schultz BM, Huhn JM, Jelinek J, Madzo J, Engel N (2017) Sex chromosomes drive gene expression and regulatory dimorphisms in mouse embryonic stem cells. Biol Sex Differ 8:28
Wijchers PJ, Yandim C, Panousopoulou E, Ahmad M, Harker N, Saveliev A, Burgoyne PS, Festenstein R (2010) Sexual dimorphism in mammalian autosomal gene regulation is determined not only by Sry but by sex chromosome complement as well. Dev Cell 19:477–484
Xu J, Burgoyne PS, Arnold AP (2002) Sex differences in sex chromosome gene expression in mouse brain. Hum Mol Genet 11:1409–1419
Yamauchi Y, Riel JM, Wong SJ, Ojarikre OA, Burgoyne PS, Ward MA (2009) Live offspring from mice lacking the Y chromosome long arm gene complement. Biol Reprod 81:353–361
Yamauchi Y, Riel JM, Stoytcheva Z, Ward MA (2014) Two Y genes can replace the entire Y chromosome for assisted reproduction in the mouse. Science 343:69–72
Zheng Y, Hlady RA, Joyce BT, Robertson KD, He C, Nannini DR, Kibbe WA, Achenbach CJ, Murphy RL, Roberts LR, Hou L (2019) DNA methylation of individual repetitive elements in hepatitis C virus infection-induced hepatocellular carcinoma. Clin Epigenet 11:145
Zhuang QK, Galvez JH, Xiao Q, AlOgayil N, Hyacinthe J, Taketo T, Bourque G, Naumova AK (2020) Sex chromosomes and sex phenotype contribute to biased DNA methylation in mouse liver. Cells 9:1436
Acknowledgements
This work was supported by the Natural Sciences and Engineering Research Council of Canada discovery grant and discovery acceleration supplement (to AKN). The authors are thankful to Drs. Danielle Malo, David Labbe, Loydie Majewska, and Fernando Pardo Manuel de Villena for providing mice or samples from different laboratory strains, to Tim Bell for technical assistance, and to Fernando Pardo Manuel de Villena and Sanny Moussette for helpful discussion.
Author information
Authors and Affiliations
Contributions
E.B. performed most of the experimental work and contributed to writing of the manuscript, N.A. performed RNA and DNA analyses and assay designs and contributed to writing of the manuscript, J.H.G. helped in most analyses of WGBS and RNA-seq data; Q.W.Z. performed WGBS analyses and contributed to writing of the manuscript, K.B. performed some of the genotyping experiments. K.N., T.K., and M.A.W. generated mutant mice and provided DNA from mutant mice. T.T. generated consomic B6.YTIR mice and their littermates as well as Paf mutant mice for experiments. M.A.W. and T.T. contributed to writing the manuscript. A.K.N. conceived the experimental design and wrote most of the main text of the manuscript. A.K.N. and G.B. provided guidance and oversight of the project. All authors reviewed the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors have no competing interests to declare that are relevant to the content of this article.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Batdorj, E., AlOgayil, N., Zhuang, Q.Kw. et al. Genetic variation in the Y chromosome and sex-biased DNA methylation in somatic cells in the mouse. Mamm Genome 34, 44–55 (2023). https://doi.org/10.1007/s00335-022-09970-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00335-022-09970-z