TYPE Original Research
PUBLISHED 17 April 2023
DOI 10.3389/fgene.2023.1069871
R-loop landscape in mature
OPEN ACCESS human sperm: Regulatory and
EDITED BY
Jorg Tost, evolutionary implications
Commissariat à l’Energie Atomique et aux
Energies Alternatives, France
REVIEWED BY Maurice Scheuren*, Jonas Möhner and Hans Zischler*
Marta Olszewska,
Institute of Human Genetics, Polish Division of Anthropology, Faculty of Biology, Institute of Organismic and Molecular Evolution, Johannes
Academy of Sciences, Poland Gutenberg University Mainz, Mainz, Germany
Belén Gomez-Gonzalez,
Andalusian Center of Molecular Biology
and Regenerative Medicine (CSIC), Spain
*CORRESPONDENCE R-loops are three-stranded nucleic acid structures consisting of an RNA:DNA
Maurice Scheuren, hybrid and a displaced DNA strand. While R-loops pose a potential threat to
mscheur@uni-mainz.de
Hans Zischler, genome integrity, they constitute 5% of the human genome. The role of R-loops in
zischler@uni-mainz.de transcriptional regulation, DNA replication, and chromatin signature is becoming
SPECIALTY SECTION increasingly clear. R-loops are associated with various histone modifications,
This article was submitted to suggesting that they may modulate chromatin accessibility. To potentially
Epigenomics and Epigenetics,
harness transcription-coupled repair mechanisms in the germline, nearly the
a section of the journal
Frontiers in Genetics entire genome is expressed during the early stages of male gametogenesis in
14 October 2022 mammals, providing ample opportunity for the formation of a transcriptome-RECEIVED
ACCEPTED 03 April 2023 dependent R-loop landscape in male germ cells. In this study, our data
PUBLISHED 17 April 2023 demonstrated the presence of R-loops in fully mature human and bonobo
CITATION sperm heads and their partial correspondence to transcribed regions and
Scheuren M, Möhner J and Zischler H chromatin structure, which is massively reorganized from mainly histone to
(2023), R-loop landscape in mature
human sperm: Regulatory and mainly protamine-packed chromatin in mature sperm. The sperm R-loop
evolutionary implications. landscape resembles characteristic patterns of somatic cells. Surprisingly, we
Front. Genet. 14:1069871. detected R-loops in both residual histone and protamine-packed chromatin
doi: 10.3389/fgene.2023.1069871
and localize them to still-active retroposons, ALUs and SINE-VNTR-ALUs
COPYRIGHT (SVAs), the latter has recently arisen in hominoid primates. We detected both
© 2023 Scheuren, Möhner and Zischler.
This is an open-access article distributed evolutionarily conserved and species-specific localizations. Comparing our DNA-
under the terms of the Creative RNA immunoprecipitation (DRIP) data with published DNA methylation and
Commons Attribution License (CC BY). histone chromatin immunoprecipitation (ChIP) data, we hypothesize that
The use, distribution or reproduction in
other forums is permitted, provided the R-loops epigenetically reduce methylation of SVAs. Strikingly, we observe a
original author(s) and the copyright strong influence of R-loops on the transcriptomes of zygotes from early
owner(s) are credited and that the original developmental stages before zygotic genome activation. Overall, these findings
publication in this journal is cited, in
accordance with accepted academic suggest that chromatin accessibility influenced by R-loops may represent a
practice. No use, distribution or system of inherited gene regulation.
reproduction is permitted which does not
comply with these terms.
KEYWORDS
inherited gene regulation, primate evolution, retrotransposon, epigenetics, human
sperm, R-loop
1 Introduction
Since the discovery of R-loops, one of the most abundant non-B-DNA structures in
mammalian genomes (Al-Hadid and Yang, 2016), diverse types of biological functions and
consequences have been characterized, most notably and recently their regulatory potential.
Initially, R-loops were considered transcriptional byproducts that negatively affect genome
stability by causing replication stress when they collide with the replication machinery
(Aguilera and García-Muse, 2012). However, genomic results from antibody-based scans for
the presence of R-loops suggest their occurrence in non-transcribed regions and
Frontiers in Genetics 01 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
consequently, various non-exclusive models for their biogenesis the defense mechanisms against the mobility of TEs, e.g.,
(Crossley et al., 2019; Sanz and Chédin, 2019). Current methylation of the source gene, are imperfect. In consequence
knowledge suggests that R-loops play key roles in transcriptional this might lead to de novo integrations, especially of younger
regulation and replication, genomic instability independent of TEs, e.g. of ALUs, the most abundant hominoid SINE, which
replication stress, class switch recombination in B cells, and DNA occurs in about 1/40 births (Schmid, 1991; Belyeu et al., 2021).
damage and repair (Niehrs and Luke, 2020). In addition, R-loops When corrected for copy number, hominoid-specific SVA (SINE-
modulate chromatin architecture through non-B-DNA structure VNTR-Alus) exhibit even greater mobility. The SVA is a composite
and depletion of nucleosomes, becoming regulators of chromatin of a (CCCTCT)n hexamer repeat, two antisense ALU fragments, a
state and thus genes (Powell et al., 2013; Bayona-Feliu and Aguilera, VNTR and the env-gene plus 3′LTR fromHERV-K10 (Cordaux and
2021). Batzer, 2009). SVAs are preferentially found in gene-rich regions
Recently, Adrian-Kalchhauser et al. (2020) proposed ‘inherited and are of ever-increasing interest due to their co-evolution with TF
gene regulation’ as general feature of epigenetic inheritance. With and thus the potential regulation of expression in nearby genes
R-loops potentially modulating gene expression in somatic cells, (Savage et al., 2013; Gianfrancesco et al., 2019; Senft and Macfarlan,
they could hypothetically act as intergenerationally epigenetic active 2021; Barnada et al., 2022). SVAs are one of the youngest TEs in
elements when present in the germline. The persistence of R-loops primates, and some copies are still active in the lineages that give rise
in the genome of oocytes at metaphase II and in embryos at the to the extant primate representatives, bonobos and humans (Wang
preimplantation stage and at all cell cycles was documented by et al., 2005). Interestingly, the SVAs showed strong hypomethylation
immunofluorescence. Maternal and paternal pronuclei showed the in human spermatozoa and are reported to be overrepresented in
highest intensities at pronuclear stage 1. The authors hypothesize somatic R-loops (Molaro et al., 2011; Zeng et al., 2021).
that the R-loops detected in the early pronuclear stages may be In this study, we correlated both sperm head transcript profiles
related to epigenetic mechanisms such as histone modifications and and supposedly transcriptionally inactive protamine-covered
chromatin structure and may contribute to the genomic regions with R-loops (D’Occhio et al., 2007; Rathke et al., 2007),
reprogramming that occurs in separate pronuclei during this with transcription during spermatogenesis as main source for
early zygotic development, and put forward that the dynamics of R-loop biogenesis. We could furthermore highlight the role of
the R-loop landscape are likely mechanistically linked to zygotic hominoid-specific, still actively transposing SVA, in R-loop
genome activation (ZGA) (Lee et al., 2022). This leaves open the formation in both humans (referred to as Hsa in figures and
question of whether R-loops occur in mature sperm cells and thus tables) and Pan paniscus (referred to as bonobo in the text; Ppa
the search for their possible evolutionary consequences in the in figures and tables) sperm heads and demonstrate that R-loops
context of ‘non-genetic or epigenetic inheritance’ has not yet possess the potential to represent a system of ‘inherited gene
been addressed. To investigate a possible paternal contribution to regulation’ (Adrian-Kalchhauser et al., 2020).
epigenetic inheritance effective across evolutionary time scales, we
focused on the male germline and examined sperm heads for the
presence of R-loops. Sperm heads can be regarded as the precursors 2 Material and methods
of the male pronucleus in the zygote because after the sperm has
fertilized the egg, the decondensation of the sperm nucleus begins as 2.1 Sperm head preparation
the first step towards the formation of the male pronucleus (Lassalle
and Testart, 1991). During spermiogenesis, testis-specific histone All ejaculate samples were non-invasively obtained. Human
variants are incorporated into nucleosomes and subsequently samples were provided by volunteers by masturbation after
hyperacetylated and removed, leading to the replacement of 2–3 days of sexual abstinence, with informed consent. The
histones by transition proteins, and eventually to the transition bonobo sample was obtained at Zoo Wuppertal. The sample was
to smaller protamines (Rathke et al., 2014; Bao and Bedford, 2016). collected opportunistically (from the cage floor after the animal
This global remodeling together with DNA methylation and masturbated) and shipped at ambient temperature. All samples were
reprogramming of the genome could allow transcription and stored at −25°C/-80°C until further processing.
subsequent reintegration of transposable elements (TE), which The total ejaculate was centrifuged (16,000 x g, 5 min at room
mainly fall into two types of retroposons, namely, SINEs (short temperature) and the supernatant was discarded. The cell pellet was
interspersed nuclear elements) and LINEs (long interspersed nuclear resuspended in lysis buffer (10 mM TRIS pH 8; 10 mM EDTA;
elements). This poses a threat to the integrity of the genome in the 100 mM NaCl; 4% SDS). The suspension was centrifuged, the
male germline at various stages of spermiogenesis (Soumillon et al., supernatant discarded and the previous step repeated. The pellet
2013; Ward et al., 2013). Although this histone-to-protamine was resuspended in 1.35 mL lysis buffer, 150 µL 1 M dithiothreitol
transition is apparently incomplete and varies between different (DTT) was added for optimal sperm lysis, and the lysate was
mammalian taxa (Torres-Flores and Hernández-Hernández, 2020), incubated at 55°C for 30 min.
the resulting chromosome condensation is thought to coincide with
gradual termination of transcription (D’Occhio et al., 2007; Rathke
et al., 2007). In contrast, NGS data suggest that many different types 2.2 DNA-RNA immunoprecipitation (DRIP)
of mRNA and non-coding RNA are present in fully differentiated assay
sperm and can be transmitted intergenerationally into zygotes and
early embryos (Wei et al., 2014; Santiago et al., 2021; Conine and DRIP was performed as previously described with minor
Rando, 2022). Consistent with the histone-to-protamine transition, modifications (Halász et al., 2017; Bou-Nader et al., 2022).
Frontiers in Genetics 02 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
Briefly, SureBeads™ Protein A Magnetic Beads (BioRad, was used to calculate fragments per kilobase of transcript per million
United States) were pre-blocked with PBS containing 0.5% BSA fragments mapped (FPKM) data, and transcripts with an
and 5 mM EDTA and subsequently washed with wash buffer (PBS; FPKM <1 were discarded. RSeQC (version 5.0.1) was used for
1% Triton X-100; 1 mM EDTA). To pre-immobilize the calculation of transcript integrity number (TIN), transcripts with
S9.6 antibody (Active Motif, United States), 100 µL pre-blocked TIN >75% were considered to be intact (Wang et al., 2012; Wang
beads were incubated with 5 µg of S9.6 antibody in binding buffer et al., 2016). GeneIDs of transcripts were used for annotation with
(50 mM TRIS; 0.14 M NaCl; 5 mM EDTA; 1% Triton X-100) at 4 °C Ensembl Release 108 (Cunningham et al., 2022).
for 2 h with rotation. DNA was isolated from sperm head pellets,
lysed with DTT, using phenol/chloroform/isoamyl alcohol (25:24:1),
precipitated with ethanol and resuspended in TE buffer. The MboI 2.5 Identification of genomic regions with
digested DNA was added to the beads and incubated overnight at GC skew
4°C. Bound beads were recovered and washed three times with wash
buffer (PBS; 1% Triton X-100; 1 mM EDTA). Precipitates were To define the regions displaying a GC skew in the human and
eluted in elution buffer (10 mM Tris pH 8; 1 mM EDTA; 1% SDS) bonobo genomes, we applied the SkewR pipeline 1.00 b using the
and 5 µL proteinase K (10 mg/mL) in 100 µL for 15 min at 65 °C. most stringent model for GRCh38 and Mhudiblu_PPA_
DNA was purified using a QIAquick® PCR purification Kit v0 respectively (Ginno et al., 2012; Ginno et al., 2013).
(QIAGEN, Netherlands). Libraries were prepared by Novogene
using the NEB Next Ultra DNA Library Prep Kit and sequenced
on Illumina NovaSeq 6000 PE150 (Hsa n = 2, Ppa n = 1). 2.6 Gene ontology enrichment analysis
Gene ontology (GO) Biological Process 2021 enrichments were
2.3 Bioinformatic methodology of the DRIP- analyzed by using Enrichr (https://maayanlab.cloud/Enrichr/)
seq analysis (Chen et al., 2013; Kuleshov et al., 2016; Xie et al., 2021).
DRIP-seq reads were aligned to GRCh38 genome andMhudiblu_
PPA_v0 respectively, using STAR (version 2.7.9a) with parameters 2.7 Motif analysis
according to Teissandier et al. (2019) for a TE-sensitive mapping
(Teissandier et al., 2019), PCR-duplicated reads were removed using The Simple Enrichment Analysis software (version 5.5.0) from
Picard (‘Picard Toolkit.’ 2019. Broad Institute, GitHub Repository. MEME Suite (http://meme-suite.org/) was used with default settings
https://broadinstitute.github.io/picard/; Broad Institute) and to analyze the motifs present in R-loops (Bailey and Grant, 2021).
converted using SAMTools (version 1.10). To assign TE-derived
reads to individual peaks, the CSEM workflow (Chung et al., 2011)
was used, and MACS (version 3.0.0a7) with parameters according to 2.8 Retrieval of public datasets
Zeng et al. (2021) was used for peakcalling with digested gDNA Input
as control. We selected the output file ‘. broadPeak’ for the final peak We obtained the datasets (GSE57095, GSE40195,
detection results. Blacklisted regions of hg38 (obtained from http:// GSE144283 and GSE30340) from the GEO database (Table 1). If
mitra.stanford.edu/kundaje/akundaje/release/blacklists/hg38-human/ necessary, data was converted to GRCh38 using Liftover. We
hg38.blacklist.bed.gz) were removed using BEDTools (version 2.30.0) conducted the authors’ workflow for datasets PRJNA715579 and
(Quinlan and Hall, 2010). Subsequently, the identified peaks were GSE44183 (Table 1).
annotated using ‘annotatePeaks.pl’ from the Homer software (version
4.11) with hg38 for the human sample and a customized reference for
bonobo respectively (Heinz et al., 2010). The customized reference 3 Results
was built from RefSeq (O’Leary et al., 2016) and RepeatMasker (Smit
et al., 2008) datasets. Metaplots were created using R package 3.1 R-loops as revealed by DRIP overlap with
ChIPseeker (Yu et al., 2015; Wang et al., 2022) and Venn RNA-seq profiles in human sperm heads
diagrams were created using Intervene (version 0.6.1) (Khan and
Mathelier, 2017). Since sperm heads best reflect the situation in the male
pronucleus of the zygote, it is important to determine the
R-loops and transcripts localized in the sperm head.
2.4 RNA-seq and bioinformatic analysis Moreover, the respective localization suggests that this
paternal information could be contributed to the oocytes
Total RNA was isolated from prepared sperm heads using the during fertilization (Peng et al., 2012). Therefore, we isolated
standard TRIzol-based protocol (Thermo Fisher Scientific, sperm heads by applying the standard protocol for differential
United States), followed by poly-A enrichment and sequencing extraction of sperm heads, first in the absence of DTT as a
by BGI Biotechnology using BGISEQ-500 Transcriptome PE100 reducing agent with repeated intermittent centrifugation and
(n = 1). RNA-seq reads were aligned to GRCh38 transcriptome subsequent lysis of sperm heads in the presence of DTT
using STAR (version 2.7.9a) with ‘--quantMode (Jankova et al., 2019). The mature sperm transcriptome is
TranscriptomeSAM’. The RSEM Workflow (Li and Dewey, 2011) either derived from residual RNA transcripts generated during
Frontiers in Genetics 03 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
TABLE 1 The information of the used datasets.
Dataset Library type Explanation Species Edit References
GSE57095 ChIP-Seq Residual histones H3K4me1, H3K27ac, H3.3 Homo sapiens Liftover to GRCh38 Hammoud et al. (2014)
GSE40195 ChIP-Seq Residual histones H3K14ac Homo sapiens Liftover to GRCh38 N/A
PRJNA715579 ChIP-Seq Residual histones TH2B Homo sapiens Liftover to GRCh38 Patankar et al. (2021)
GSE144283 ChIP-Seq H3S10p marks in IMR-5 cells Homo sapiens Liftover to GRCh38 Roeschert et al. (2021)
GSE44183 RNA-Seq RNA expression in early human embryos Homo sapiens Xue et al. (2013)
GSE30340 Bisulfite-Seq Methylome of mature human sperm Homo sapiens Liftover to GRCh38 Molaro et al. (2011)
PRJNA890147 DRIP-Seq R-loop landscape of mature human sperm Homo sapiens This study
PRJNA890147 DRIP-Seq R-loop landscape of mature chimp sperm Pan paniscus This study
PRJNA890147 RNA-Seq Transcriptome of mature human sperm Homo sapiens This study
the early stages of spermatogenesis or is produced through active detected 6278 ± 639 peaks covering 4.94 ± 1.37 Mbp of the
transcription (Ren et al., 2017). Regardless of the transcript human sperm genome of which 52.76% ± 2.64% were located
origin, the existing RNA can interact with sperm chromatin in in genes. The average GC content of R-loops was 54.99% ± 0.02%.
a variety of ways and can be introduced into the oocyte as a To further investigate the correlation between R-loops and the base
substantial paternal contribution of diverse populations of RNAs composition of the corresponding genomic region, we checked for
(Sharma et al., 2018; Li and Klungland, 2020; Kretschmer and GC skews in genes associated with R-loops (Table 2). The
Gapp, 2022). Thus, the transcriptome is a major key for annotation by SkewR revealed that more than 75% of genes
understanding R-loops in sperm. Therefore, we first associated with R-loops showed a GC skew, which can facilitate
investigated the overlap between human sperm head nascent the formation of R-loops (Figure 2A). Furthermore, the density of
transcripts and corresponding R-loops. To this end RNA was R-loops per chromosome strongly correlates (Pearson’s r = 0.7468,
isolated from sperm heads lysed with DTT according to standard p < 0.001) with the chromosome specific gene density (Figure 2B).
TRIZOL-based protocols with subsequent precipitation and Both the strong GC skew and the gene dependency resembles the
subjected to deep sequencing with 45,811,118 clean reads R-loops landscape of somatic cells, indicating transcription as the
obtained from RNA-seq. Applying bioinformatic routines as main contributor to R-loop formation. Because transcription is mostly
implemented in STAR, we detected and mapped inactivated in mature sperm, transcription during spermatogenesis
32,534 expressed protein-coding and RNA genes in the sperm and residual transcripts could facilitate R-loop formation. Therefore,
transcriptome (genes with FPKM >1 were considered to be we investigated the genome-wide relationship between the
expressed). The annotated transcript population mainly transcriptome of mature sperm and the R-loop landscape. We
consists of mRNAs and long non-coding (lnc)RNAs, followed checked for overlaps between the two features and found that
by smaller fractions of transcribed pseudogenes and source more than 60% of genomic R-loops had corresponding transcripts
transcripts of diverse small non-coding RNAs (Figure 1A; in the sperm head (p < 0.001, Hypergeometric test). Regarding the
Supplementary Table S1). The complex mRNA, lncRNA and correlation between source genes of the transcripts and R-loops, we
pseudogene landscape results from the global transcription detected a strong bias towards protein-coding genes, in which 89,14 ±
during spermatogenesis. All thirty most abundant and intact 0.07% were associated with R-loops, contrasting just 31,83 ± 10,78%
(TIN >75%) transcripts are mRNAs and their translated proteins of ncRNA genes (Figure 2C).
correspond mainly to the biological processes of oxidative To further investigate the contribution of transcription to the
phosphorylation, sperm DNA condensation and cytoplasmic R-loop landscape, we used Simple Enrichment Analysis for a motif
translation (Figure 1B). enrichment analysis and detected motifs for the transcription factors
Next, we isolated sperm heads from the ejaculates and lysed (TF) ZNF460 and ZNF135 in more than 20% of human R-loops
them by adding DTT to a final concentration of 100 mM. (Figure 2D; Supplementary Table S2). Both, ZNF460 and ZNF135,
Subsequently phenol-chloroform extraction and DNA isolation function as TF for RNA polymerase II. RNA polymerase II is mainly
by precipitation were used to prepare an antibody-based ChIP-like responsible for the transcription of mRNAs, micro-RNAs (miRNAs)
genome-wide analysis of R-loops (DNA-RNA- and small nuclear RNAs (snRNAs). Both motifs were also enriched
immunoprecipitation, DRIP). To scrutinize the possible in the R-loop landscape of the bonobo. Motifs of two closely related
conservation of R-loop profiles in closely related primate TFMEF2A andMEF2D, which are able to form a heterodimer, were
representatives, we isolated sperm heads from a bonobo enriched in human R-loops, too. The MEF2A motif was also more
ejaculate and DRIP - analyzed the sample as well. Using DRIP- abundant in R-loops of the bonobo. The motif enrichment and
Seq and evaluating the data with STAR-mapping and CSEM to R-loop bias towards protein-coding genes reinforces transcription
incorporate repetitive DNA regions into peaks annotation we during the spermatogenesis, especially by RNA polymerase II, as the
Frontiers in Genetics 04 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
FIGURE 1
Transcriptome of mature human sperm (A) Biotype annotation of transcripts. (B) Biological processes of the thirty most abundant intact transcripts
(TIN >75%).
TABLE 2 Basic analysis of the DRIP-Seq experiments. Error shows ±standard error of the mean (SEM) among the biological replicates.
Number of peaks Coverage of peaks [Mbp] Average GC content [%] R-Loops associated with genes [%]
R-loops Hsa 6278 ± 639 4.94 ± 1.37 54.99 ± 0.02 52.76 ± 2.64
R-loops Ppa 17,560 6.31 51.55 60.18
main source of R-loops in the mature sperm. The finding of mature sperm head, suggesting their presence in non-transcribed
common enriched motifs in human and bonobo R-loops may regions or non-polyadenylated RNA as well. The applied RNAseq
primarily reflect the expected similarity of transcriptional protocols are based on an enrichment of poly-A RNA, but the exact
patterns in closely related primate species and global comparison of sperm head transcriptome data in a meaningful
transcription in the early stages of spermatogenesis (Xia et al., quantifiable approach to estimate the relative proportion of
2020). On the other hand, the other 39.19% ± 0.81% of the sequenced polyadenylated transcripts versus non-polyadenylated
genome wide R-loops show no corresponding transcripts in the RNA is pending.
Frontiers in Genetics 05 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
FIGURE 2
R-loop Distribution across the sperm genome and transcriptome. (A) Pie charts showing the distribution of human and bonobo genes with respect
to their GC skew according to SkewR. (B) Relative proportion of R-loop sequences per chromosome compared to the gene density of the corresponding
chromosome. (C) Venn diagrams showing the overlap between the human protein-coding transcriptome and R-loops, indicated are values for each
replicate. (D) Most enriched motifs in R-loops in human and bonobo.
3.2 R-loop formation is facilitated in gene R-loop formation. The highest enrichment was observed for
bodies CpG islands (CGI). Similar to human sperm, R-loops in
bonobo sperm heads also tend to form in genes, especially in
By associating peaks with genomic features including TSS, exons, and TTS, but also in CGIs albeit to a lesser extent as
different classes of TEs on a multilocus level, we found an compared to the human situation (Figures 2A, 3A).
enriched (log2 (observed/expected)) R-loop formation in gene
bodies (Figures 3A,B), as can be observed in somatic cells, too
(Sanz et al., 2016; Chen et al., 2017). Owing to their potentially 3.3 Only a minor fraction of R-loops
different modes of biogenesis and regulatory effects, the R-loop localizes to human-bonobo orthologous
peaks were further differentiated according to frequencies of loci
their occurrence along the gene body (Figure 3A). Thus, we
examined R-loops coinciding with transcription start sites To examine a possible locus-specific conservation of R-loops, we
(TSS), exons, introns, and transcription termination sites pairwise compared single loci for humans with the orthologue in
(TTS). As a result, R-loop formation in human sperm heads bonobos for conserved R-loop formation. To this end we subdivided
was in general strongly favored in TSS, exons and TTS the loci under scrutiny into the abovementioned gene components
particularly in genes showing a strong GC skew, whereas and regulatory regions. Neither in TSS, nor within exons or TTS a
introns and intergenic regions showed no enrichment in significant number of orthologs being identical between humans and
Frontiers in Genetics 06 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
FIGURE 3
R-loop Distribution across gene bodies. (A)Genomic distribution of R-loop peaks across genomic annotations compared to the expected genomic
distribution in human and bonobo sperm. Error bars shows ±SEM among the biological replicates. (B)Metaplot of the R-loop distribution in human and
bonobo sperm across gene bodies, relative to transcription start site and transcription termination site. (C) Pie charts showing the proportion of R-loops in
in human sperm with an orthologous gene feature in bonobo. (D) Summary of GO Biological Process 2021 enrichment results of conserved introns
between human and bonobo, ordered by p-values.
Frontiers in Genetics 07 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
FIGURE 4
R-loop Distribution across transposable elements. (A) Genomic distribution of R-loop peaks across repetitive elements compared to the expected
genomic distribution in human and bonobo sperm. (B) Distribution of R-loops in different ALU subfamilies in human and bonobo sperm. The bisection
indicates an equal distribution between human and bonobo (C) Distribution of R-loop peaks in SVA subfamilies compared to the expected distribution in
human and bonobo sperm. (D) Relative methylation rate of SVA subfamilies. Data in (A), (C) and (D) represent mean ± SEM among the biological
replicates.
Frontiers in Genetics 08 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
bonobos could be seen. Contrasting to this, 10% of the human resulting in a human-specific signature of intronic R-loop
introns with R-loops shared an intronic R-loop orthologue in the formation in SVAs.
bonobo (Figure 3C). Interestingly, Gene Ontology (GO) Previous studies have described the absence of methylation at
enrichment analysis of the corresponding genes using Enrichr CGI associated with R-loops and protection from de novo
(Kuleshov et al., 2016; Chen et al., 2013; Xie et al., 2021) displayed methylation by DNMT3B1 during early development.
enriched GO terms including positive regulation of epithelial cell Interestingly, some TEs tend to evade the re-methylation during
migration, positive regulation of substrate adhesion-dependent spermatogenesis after a genome wide erasure of epigenetic marks
cell spreading, positive regulation of cell-substrate adhesion and (Molaro et al., 2011; Ginno et al., 2012; Rodriguez-Terrones and
positive regulation of cell morphogenesis involved in Torres-Padilla, 2018; Dietmann et al., 2020). During
differentiation. These biological processes could all be relevant spermatogenesis nearly the whole genome is transcribed in an
for the early development of a zygote in both species (Figure 3D). extreme global transcription, including TEs (Xia et al., 2020).
In summary, the picture emerges that potentially regulatory Therefore, some highly transcribed TEs might adopt R-loops
R-loops act in a probabilistic manner rather than as discrete through excessive transcription and their composition thus
epigenetic character states at defined orthologous loci (Adrian- escaping a subsequent re-methylation through their RNA:DNA
Kalchhauser et al., 2020). hybrid structure. Dietmann et al. (2020) stated that most of the
young SVAs stay hypomethylated in human primordial germ cells,
after an erasure of epigenetic marks and subsequent re-methylation.
3.4 Transposable elements in introns might Therefore, we compared our DRIP data to sperm methylomes
represent hotspots of R-loop formation (Molaro et al., 2011). Noticeably, more than 60% of SVAs
adopting an R-loop structure are hypomethylated compared to a
TEs constitute a major part of primate genomes, and thus, genome-wide 40% of SVAs without a R-loop association. The
intronic sequences. With the likely detrimental effect of human-specific, actively transposing SVA_E and SVA_F are less
insertional mutagenesis into exonic sequences, reduced methylated than the evolutionary old SVAs common to all
evolutionary constraints hold for introns, although TEs shape the homininae, with the exception of SVA_C (Figure 4D). These
function of their corresponding introns by triggering differential findings of hypomethylated young SVAs in sperm correlate with
splicing, premature stop codons and maintaining an open the results of Dietmann et al. (2020) analyzing human primordial
chromatin state (Cordaux and Batzer, 2009; Zhang et al., 2011; germ cells. Escaping the re-methylation, similar to R-loop associated
Ohtani and Iwasaki, 2021). Therefore, in particular larger TEs such CGIs, could promote transcription of the hypomethylated SVAs and
as composite retroposons, for example, SVAs, and LINEs are more thus enhance the mobility of young SVAs in the male germline.
likely to be conserved in introns and intergenic regions as compared Moreover, the hypomethylation of intronic SVAs could facilitate
to TE insertions in coding sequences. Moreover, due to the high GC transcription in corresponding genes, creating a co-transcriptional
content of primate-specific SINEs and their copy number, they are influence on genes.
likely to facilitate the formation of R-loops during transcription at
one site and potentially also the re-hybridization of abundant SINE
transcripts, respectively (Zeng et al., 2021). R-loops were enriched in 3.5 Only a small fraction of sperm head
rRNA genes, satellite DNA and SVAs, whereas they were R-loops coincides with residual histones
underrepresented in LINEs and long terminal repeat
retrotransposon (LTRs) (Figure 4A). Therefore, we analyzed two During spermiogenesis and beginning with the elongating
actively transposing and thus transcribed TEs in more detail and spermatid stage, human sperm chromatin undergoes a dramatic
looked at hominoid ALUs and SVAs. transition in which histones are largely replaced by protamines.
Interestingly, the enrichment of SINEs concomitant with Residual histones in spermatogenesis remain because of an
R-loops is different in humans and bonobos (Figure 4A). incomplete erasure of somatic and transitional histones, such
Therefore, we analyzed the proportion of ALU-subfamilies as H3.3 and TH2B. It is hypothesized that they play a potential
contributing to the R-loop landscape. Most of the ALU- role in epigenetic inheritance due to their impact on the
subfamilies showed similar association with R-loops in both chromatin state (Wang et al., 2019). In contrast to the tight
human and bonobo sperm. Interestingly, the older subfamilies protamine packaging, residual histones create open chromatin
ALUSz, ALUSx and ALUJb, showed human-specific enrichment, poised for transcription and facilitating a re-hybridization of
whereas the much younger ALUY (Kapitonov and Jurka, 1996), transcripts and genomic DNA. Therefore, we investigated the
showed a strong enrichment in bonobo sperm (Figure 4B). Upon coincidence of R-loops in regions with residual histones H3.3 and
comparing the respective TE-R-loop coincidence in both human THB2 from spermatogenesis and residual somatic histones,
and bonobo sperm, we observed that in both taxa SVAs were typically marking open chromatin for active transcription.
enriched with R-loops. Strikingly, the human-specific families Previous studies described genomic regions tagged by
SVA_E and SVA_F, showed the highest R-loop formation rate, transcriptionally active chromatin states with the occurrence
whereas the youngest common SVA_D was most enriched in the of H3K27ac, H3K14ac and H3K4me1. Interestingly, no
bonobo (Figure 4C). Compared with bonobo sperm, where all significant overlap was detected between R-loops and marks of
R-loop-covered SVAs were intergenic, 60% of human SVAs transcriptional activity with the above-mentioned modified
associated with R-loops were found in introns. The SVAs in histones. We further investigated the combined occurrence of
introns were mostly human-specific integrations (81%), H3K27ac and H3K4me1, which marks an active enhancer, due to
Frontiers in Genetics 09 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
FIGURE 5
Potential epigenetic effective R-loops in sperm (A) Venn diagrams showing the overlap between R-loops and residual histones H3K14ac,
H3K4me1 and H3K27ac in human sperm. The overlapping peaks of DRIP replicates were used. (B) Venn diagrams showing the overlap between R-loops
and residual histones H3.3 and TH2B in human sperm. The overlapping peaks of DRIP replicates was used. (C) Fraction of early transcribed genes in the
human zygote by developmental stage associated with R-loops in human sperm. Error bars shows ±SEM among the biological replicates. Two-
tailed Student’s t-test was used to test for significance (p < 0.05).
their interaction with various lncRNAs. Marks of active Although residual histones, regardless of their origin during
enhancers also showed no significant correlation with R-loops spermatogenesis, should facilitate the formation of R-loops because
(Figure 5A). Genomic regions with an incomplete histone-to- of potential transcription and an open chromatin state (Hammoud
protamine transition, retaining the residual histones H3.3 and et al., 2011; Schagdarsurengin et al., 2012; Wang et al., 2019), only a
TH2B, showed no significant overlap with R-loops as well fraction of R-loops can be found in regions of residual histones. This
(Figure 5B). leads to the hypothesis that R-loops occur even in the tightly
Frontiers in Genetics 10 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
protamine-packed chromatin of sperm or in regions depleted from potentially early developmental processes. The motif enrichment
both protamines and residual histones. analysis revealed conserved motifs in R-loops of both taxa. The most
enriched motifs are binding sites for TF of RNA polymerase II,
which primarily transcribes mRNAs and miRNAs. Furthermore,
3.6 R-loop containing genes might trigger more than half of the human R-loops show a corresponding
intergenerationally effective epigenetic transcript in the transcriptome of the mature sperm, which
marks results from transcription during early spermatogenesis, because
of a mostly inert transcription in sperm (D’Occhio et al., 2007;
Like canonical histone modifications, R-loops can influence Rathke et al., 2007). These findings implicate the virtually genome-
DNA-protein-interactions, and reconstruct the spatial epigenetic wide transcription during spermatogenesis as the main contributor
landscape due to its three-dimensional structure (Al-Hadid and to the sperm R-loop landscape. Moreover, R-loop formation as a
Yang, 2016). This global spatial rearrangement of sperm chromatin byproduct of transcription is favored by GC skew of transcribed
could potentially play a major role in early transcription and DNA genes. Both the transcription-coupled R-loop formation leading to
interactions in the pronucleus of the zygote, also implementing an an R-loop pattern and the fact that R-loop RNA oligonucleotides
intergenerational epigenetic effect mediated by R-loops. Before the constitute a significant part of the non-coding RNA population in
major wave of autonomous transcription during ZGA beginning at sperm remind them of somatic cells.
the 4-cell-stage of the human zygote, minor transcriptional activities
can be detected as early as the formation of the pronuclei directly
after fertilization (Vassena et al., 2011; Xue et al., 2013; Rodriguez- 4.2 TEs as evolutionary hotspots for R-loop
Terrones and Torres-Padilla, 2018). This early transcription is formation
assumed to be strongly influenced by epigenetic factors of the
sperm and oocyte. In particular, transcription in the male Hypothetically and because of their GC skew and repetitive
pronucleus would essentially be dependent on the paternal nature, TEs could function as species-specific anchors for
transferred epigenetic state of the sperm genome. This could lead R-loops across evolutionary time scales (Zeng et al., 2021).
to a male influence on the early zygotic transcription pattern just The primate-specific ALUs and SVAs show both species-
hours after fertilization, influencing major developmental decisions specific and evolutionary conserved trends with respect to the
(Fraser and Lin, 2016). associated R-loops. Overall ALUs associated with R-loops were
To correlate the stage specific transcription patterns following enriched in bonobo but not in human sperm. We observed a very
zygote formation with our sperm head R-loop data, we compared similar distribution of R-loops in most of the ALU-subfamilies,
the presence of R-loops in different genes with their corresponding whereas older Subfamilies ALUSz, ALUSx and ALUJb show an
stage-dependent expression levels as outlined in Xue et al. (2013). enrichment bias towards human sperm. Interestingly, the
Remarkably, the transcription of genes associated with R-loops youngest ALUY in association with R-loops shows a strong
tends to start early in the zygote, especially directly after enrichment in bonobo. This association causes a genome-wide
fertilization in the 1-cell and 2-cell stages, whereas genes enrichment of ALUs covered in R-loops and therefore a species-
expressed days after fertilization show just few R-loops specific feature. The SVA is of ever-increasing interest due to its
(Figure 5C). The transcription patterns of the 1-cell and 2-cell co-evolution with TF and thus the potential regulation of
stages showed the highest similarity to the sperm R-loop profile, expression in nearby genes (Savage et al., 2013; Barnada et al.,
descending to the ZGA and following the autonomous 8-cell stage 2022). We detected an enrichment of SVAs associated with
and morula. The overlap of R-loop-associated genes and oocyte- R-loops as a conserved feature in both human and bonobo
specific expressed genes was comparable to that of sperm. Interestingly, the human-specific SVA_F, which is
Morula–expressed genes. This trend indicates a strong impact of primarily located in introns, showed the strongest enrichment
the R-loop profile to the epigenetic payload of the sperm. in R-loops. The humans-specific SVA_E and SVA_F are involved
in genomic changes during recent human evolution (Cordaux
and Batzer, 2009; Gianfrancesco et al., 2019). The SVA_F-
4 Discussion subfamily, the youngest and still actively transposing in the
human genome, even produced recent fusions with exons and
4.1 R-loops form during transcription in CGIs of multiple genes (Gianfrancesco et al., 2019). These
spermatogenesis characteristics make SVA_F an evolutionary interesting
human-specific feature in R-loops of mature sperm.
In the present study, we unveiled the presence of R-loops in Interestingly, our data suggests that the R-loop covered SVAs
mature sperm of humans and bonobo. In a genome-wide read out tend to remain hypomethylated, which could well influence
from DRIP-Seq experiments, we observed a strong correlation transcription and the epigenetic state of adjacent genes during
between R-loop abundance and gene density regarding early embryonal development, and therefore function as cis-
chromosomes. The R-loops present in both primates show a GC regulatory elements comparable to previous described findings
bias and an enrichment in gene bodies, particularly in protein- of co-transcription of genes in human stem cells (Barnada et al.,
coding genes. Most of the R-loop associated genes even show a 2022). We detected transcripts of each SVA-subfamily in the
strong GC skew. We observed evolutionary conserved R-Loops in transcriptome of mature sperm, validating their expression
orthologous introns of human and bonobo corresponding to during spermatogenesis. The transcription of the TEs is
Frontiers in Genetics 11 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
mainly suppressed during spermatogenesis by a multitude of zygote, particularly in the male pronucleus. Although, we have
mechanism (Reznik et al., 2019), although some TEs can still be provided evidence for the male side of the regulatory R-loop
mobilized causing new integrations. We speculate that actively landscape, it will be important to perform similar and
transposing SVAs could theoretically also be suppressed by the infraspecific comparative NGS experiments for the female side.
formation of regional R-loops after global demethylation and In this way the early zygotic R-loop landscape can be
before the PIWI-piRNA-system is effective, highlighting the disentangled regarding the relative contribution of paternal and
double-edged task of fostering gene transcription during maternal germ cell R-loops to the zygote. In vitro gametogenesis
gametogenesis without mobilizing TEs (Dietmann et al., 2020). starting from pluripotent stem cells may be an obvious strategy to
Until now, only few data on the quantitative and qualitative complement sperm with oocyte data and to characterize the role of
aspects of the association between SVA and R-loops are regulatory R-loops in epigenetic inheritance more completely
published. To support our conclusion of SVA silencing by (Saitou and Hayashi, 2021). Finally, we propose a new
R-loops, we exploited the link between H3S10 phosphorylation intergenerationally effective epigenetic mechanism influencing
and R-loops as a proxy to check H3S10p-ChIP data from human expression in the early embryo through paternally transmitted
IMR-5 cells for the possible enrichment of SVA sequences R-loops. Taken together, R-loops, the non-coding RNAs that
(Castellano-Pozo et al., 2013; Roeschert et al., 2021). As a constitute them, and the respective consequences on chromatin
result, we obtained a slight enrichment of SVAs in general structure should be added to the spectrum of the sperm cell’s
and more specifically an enrichment of the youngest human- payload to the zygote.
specific SVA_F in somatic cells. This pattern of H3S10p
enrichment of SVA sequences in somatic cells, though to a
lesser extent, is strikingly similar to what we detected in Data availability statement
sperm heads for the R-loop-SVA intersection thus supporting
the hypothesis of R-loops promoting the silencing of SVAs from a Original datasets are publicly available in SRA and can be found
somatic perspective. here: PRJNA890147. Publicly available datasets were analyzed in this
study. The names of the repository/repositories and accession
number(s) can be found in the article/supplementary material.
4.3 R-loops as epigenetic marks
During spermatogenesis the histone-to-protamine transition Ethics statement
ensures the integrity of the paternal DNA in the spermhead. An
incomplete histone-to-protamine transition leaves genomic regions Ethical review and approval was not required for the study on
associated with residual histones, leading to an open chromatin human participants in accordance with the local legislation and
state, poised to DNA-RNA interactions (Patankar et al., 2021). We institutional requirements. The patients/participants provided
differentiated between residual histones in marks of active their written informed consent to participate in this study.
transcription, like H3K14ac, H3K4me1 and H3K27ac and Ethical review and approval was not required for the animal
spermatogenesis relevant histones H3.3 and TH2B. Interestingly study because sample was collected non-invasively. The sample
we observed neither a significant overlap of H3K14ac, was collected opportunistically from the cage floor after the
H3K4me1 and H3K27ac with R-loops nor with H3.3 and TH2B. animal masturbated.
Previous studies hypothesized residual histones as an important
paternal epigenetic contribution to the early transcription of the
zygote (Torres-Flores and Hernández-Hernández, 2020; Patankar Author contributions
et al., 2021). Similarly, R-loops are like canonical histone
modifications-able to influence DNA-protein-interactions, and Conceptualization MS, JM, and HZ; methodology MS and JM;
can reconstruct the spatial epigenetic landscape (Al-Hadid and bioinformatic analysis MS; data interpretation MS, JM, and HZ;
Yang, 2016). Moreover, the three-stranded nature of R-loops figures, tables, graphics, MS; manuscript writing MS and HZ;
affects the broader chromatin confirmation of the sperm genome manuscript review and editing MS, JM, and HZ. All authors read
and could thus play a locally antagonistic role to the tight protamine and approved the final manuscript.
packaging. Therefore, we investigated the correlation between
R-loop associated genes in mature sperm and paternal influenced
early transcription before the zygotic genome activation. From these Acknowledgments
analyses we conclude that genes which are transcribed in the 1- and
2-cell stadium of the zygote tend to be associated with R-loops in We thank N. Mangazeev for carrying out pilot experiments
mature sperm. This association decreases in later developmental and the Zoo Wuppertal for bonobo sperm samples. HZ
stages closer to the zygotic genome activation and the final acknowledges the support of the RTG GenEvo funded by the
autonomous transcription in the 8-cell stage (Zhang et al., 2009; Deutsche Forschungsgemeinschaft (DFG, German Research
Vassena et al., 2011). We hypothesize that these R-loops facilitate Foundation)—GRK2526/1—Projectnr. 407023052. This
interaction between DNA and the transcriptional machinery publication is based on data collected as part of MS’s
through an open chromatin state and therefore play an dissertation project at the Department of Biology, Johannes
important role in fine tuning the early transcription in the Gutenberg University, Mainz, Germany.
Frontiers in Genetics 12 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
Conflict of interest organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or
The authors declare that the research was conducted in the claim that may be made by its manufacturer, is not guaranteed or
absence of any commercial or financial relationships that could be endorsed by the publisher.
construed as a potential conflict of interest.
Supplementary material
Publisher’s note
The Supplementary Material for this article can be found online
All claims expressed in this article are solely those of the authors at: https://www.frontiersin.org/articles/10.3389/fgene.2023.1069871/
and do not necessarily represent those of their affiliated full#supplementary-material
References
Adrian-Kalchhauser, I., Sultan, S. E., Shama, L. N. S., Spence-Jones, H., Tiso, S., Keller D’Occhio, M. J., Hengstberger, K. J., and Johnston, S. D. (2007). Biology of sperm
Valsecchi, C. I., et al. (2020). Understanding ’non-genetic’ inheritance: Insights from chromatin structure and relationship to male fertility and embryonic survival. Anim.
molecular-evolutionary crosstalk. Trends Ecol. Evol. 35, 1078–1089. doi:10.1016/j.tree. Reprod. Sci. 101, 1–17. doi:10.1016/j.anireprosci.2007.01.005
2020.08.011
Fraser, R., and Lin, C.-J. (2016). Epigenetic reprogramming of the zygote in mice and
Aguilera, A., and García-Muse, T. (2012). R loops: From transcription byproducts to men: On your marks, get set, go. Reproduction 152, R211–R222. doi:10.1530/REP-16-
threats to genome stability. Mol. Cell 46, 115–124. doi:10.1016/j.molcel.2012.04.009 0376
Al-Hadid, Q., and Yang, Y. (2016). R-Loop: An emerging regulator of chromatin Gianfrancesco, O., Geary, B., Savage, A. L., Billingsley, K. J., Bubb, V. J., and Quinn,
dynamics. Acta Biochim. Biophys. Sin. (Shanghai) 48, 623–631. doi:10.1093/abbs/ J. P. (2019). The role of SINE-VNTR-alu (SVA) retrotransposons in shaping the human
gmw052 genome. Int. J. Mol. Sci. 20, 5977. doi:10.3390/ijms20235977
Bailey, T. L., and Grant, C. E. (2021). Sea: Simple enrichment analysis of motifs. Ginno, P. A., Lim, Y. W., Lott, P. L., Korf, I., and Chédin, F. (2013). GC skew at the 5’
and 3’ ends of human genes links R-loop formation to epigenetic regulation and
Bao, J., and Bedford, M. T. (2016). Epigenetic regulation of the histone-to-protamine
transcription termination. Genome Res. 23, 1590–1600. doi:10.1101/gr.158436.113
transition during spermiogenesis. Reproduction 151, R55–R70. doi:10.1530/REP-15-
0562 Ginno, P. A., Lott, P. L., Christensen, H. C., Korf, I., and Chédin, F. (2012). R-loop
formation is a distinctive characteristic of unmethylated human CpG island promoters.
Barnada, S. M., Isopi, A., Tejada-Martinez, D., Goubert, C., Patoori, S., Pagliaroli, L.,
Mol. Cell 45, 814–825. doi:10.1016/j.molcel.2012.01.017
et al. (2022). Genomic features underlie the co-option of SVA transposons as cis-
regulatory elements in human pluripotent stem cells. PloS Genet. 18, e1010225. doi:10. Halász, L., Karányi, Z., Boros-Oláh, B., Kuik-Rózsa, T., Sipos, É., Nagy, É., et al.
1371/journal.pgen.1010225 (2017). RNA-DNA hybrid (R-loop) immunoprecipitation mapping: An analytical
workflow to evaluate inherent biases. Genome Res. 27, 1063–1073. doi:10.1101/gr.
Bayona-Feliu, A., and Aguilera, A. (2021). The role of chromatin at transcription-
219394.116
replication conflicts as a genome safeguard. Biochem. Soc. Trans. 49, 2727–2736. doi:10.
1042/BST20210691 Hammoud, S. S., Low, D. H. P., Yi, C., Carrell, D. T., Guccione, E., and Cairns, B.
R. (2014). Chromatin and transcription transitions of mammalian adult germline
Belyeu, J. R., Brand, H.,Wang, H., Zhao, X., Pedersen, B. S., Feusier, J., et al. (2021). De
stem cells and spermatogenesis. Cell Stem Cell 15, 239–253. doi:10.1016/j.stem.2014.
novo structural mutation rates and gamete-of-origin biases revealed through genome
04.006
sequencing of 2,396 families. Am. J. Hum. Genet. 108, 597–607. doi:10.1016/j.ajhg.2021.
02.012 Hammoud, S. S., Nix, D. A., Hammoud, A. O., Gibson, M., Cairns, B. R., and Carrell,
D. T. (2011). Genome-wide analysis identifies changes in histone retention and
Bou-Nader, C., Bothra, A., Garboczi, D. N., Leppla, S. H., and Zhang, J. (2022).
epigenetic modifications at developmental and imprinted gene loci in the sperm of
Structural basis of R-loop recognition by the S9.6 monoclonal antibody. Nat. Commun.
infertile men. Hum. Reprod. 26, 2558–2569. doi:10.1093/humrep/der192
13, 1641. doi:10.1038/s41467-022-29187-7
Heinz, S., Benner, C., Spann, N., Bertolino, E., Lin, Y. C., Laslo, P., et al. (2010). Simple
Castellano-Pozo, M., Santos-Pereira, J. M., Rondón, A. G., Barroso, S., Andújar,
combinations of lineage-determining transcription factors prime cis-regulatory
E., Pérez-Alegre, M., et al. (2013). R loops are linked to histone
elements required for macrophage and B cell identities. Mol. Cell 38, 576–589.
H3 S10 phosphorylation and chromatin condensation. Mol. Cell 52, 583–590.
doi:10.1016/j.molcel.2010.05.004
doi:10.1016/j.molcel.2013.10.006
Jankova, R., Jakovski, Z., and Janevski, R. (2019). Differential extraction method as a
Chen, E. Y., Tan, C. M., Kou, Y., Duan, Q., Wang, Z., Meirelles, G. V., et al. (2013).
golden standard in analyzing semen stains in sexual-assault cases. Forensic Sci. Int.
Enrichr: Interactive and collaborative HTML5 gene list enrichment analysis tool. BMC
Genet. Suppl. Ser. 7, 838–840. doi:10.1016/j.fsigss.2019.10.196
Bioinforma. 14, 128. doi:10.1186/1471-2105-14-128
Kapitonov, V., and Jurka, J. (1996). The age of Alu subfamilies. J. Mol. Evol. 42, 59–65.
Chen, L., Chen, J.-Y., Zhang, X., Gu, Y., Xiao, R., Shao, C., et al. (2017). R-ChIP
doi:10.1007/BF00163212
using inactive rnase H reveals dynamic coupling of R-loops with transcriptional
pausing at gene promoters. Mol. Cell 68, 745–757.e5. doi:10.1016/j.molcel.2017. Khan, A., andMathelier, A. (2017). Intervene: A tool for intersection and visualization
10.008 of multiple gene or genomic region sets. BMC Bioinforma. 18, 287. doi:10.1186/s12859-
017-1708-7
Chung, D., Kuan, P. F., Li, B., Sanalkumar, R., Liang, K., Bresnick, E. H., et al. (2011).
Discovering transcription factor binding sites in highly repetitive regions of genomes Kretschmer, M., and Gapp, K. (2022). Deciphering the RNA universe in sperm in its
with multi-read analysis of ChIP-Seq data. PloS Comput. Biol. 7, e1002111. doi:10.1371/ role as a vertical information carrier. Environ. Epigenet 8, dvac011. doi:10.1093/eep/
journal.pcbi.1002111 dvac011
Conine, C. C., and Rando, O. J. (2022). Soma-to-germline RNA communication. Nat. Kuleshov, M. V., Jones, M. R., Rouillard, A. D., Fernandez, N. F., Duan, Q., Wang, Z.,
Rev. Genet. 23, 73–88. doi:10.1038/s41576-021-00412-1 et al. (2016). Enrichr: A comprehensive gene set enrichment analysis web server
2016 update. Nucleic Acids Res. 44, W90–W97. doi:10.1093/nar/gkw377
Cordaux, R., and Batzer, M. A. (2009). The impact of retrotransposons on human
genome evolution. Nat. Rev. Genet. 10, 691–703. doi:10.1038/nrg2640 Lassalle, B., and Testart, J. (1991). Sequential transformations of human sperm
nucleus in human egg. J. Reprod. Fertil. 91, 393–402. doi:10.1530/jrf.0.0910393
Crossley, M. P., Bocek, M., and Cimprich, K. A. (2019). R-loops as cellular regulators
and genomic threats. Mol. Cell 73, 398–411. doi:10.1016/j.molcel.2019.01.024 Lee, H., You, S.-Y., Han, D. W., La, H., Park, C., Yoo, S., et al. (2022). Dynamic change
of R-loop implicates in the regulation of zygotic genome activation in mouse. Int. J. Mol.
Cunningham, F., Allen, J. E., Allen, J., Alvarez-Jarreta, J., Amode, M. R., Armean, I.
Sci. 23, 14345. doi:10.3390/ijms232214345
M., et al. (2022). Ensembl 2022. Nucleic Acids Res. 50, D988–D995. doi:10.1093/nar/
gkab1049 Li, B., and Dewey, C. N. (2011). Rsem: Accurate transcript quantification from RNA-
seq data with or without a reference genome. BMC Bioinforma. 12, 323. doi:10.1186/
Dietmann, S., Keogh, M. J., Tang, W., Magnusdottir, E., Kobayashi, T., Chinnery, P.
1471-2105-12-323
F., et al. (2020). Transposable elements resistant to epigenetic resetting in the human
germline are epigenetic hotspots for development and disease. bioRxiv. [Preprint]. Li, M., and Klungland, A. (2020). Modifications and interactions at the R-loop. DNA
doi:10.1101/2020.03.19.998930 Repair (Amst) 96, 102958. doi:10.1016/j.dnarep.2020.102958
Frontiers in Genetics 13 frontiersin.org
Scheuren et al. 10.3389/fgene.2023.1069871
Molaro, A., Hodges, E., Fang, F., Song, Q., McCombie, W. R., Hannon, G. J., et al. Senft, A. D., and Macfarlan, T. S. (2021). Transposable elements shape the evolution
(2011). Sperm methylation profiles reveal features of epigenetic inheritance and of mammalian development. Nat. Rev. Genet. 22, 691–711. doi:10.1038/s41576-021-
evolution in primates. Cell 146, 1029–1041. doi:10.1016/j.cell.2011.08.016 00385-1
Niehrs, C., and Luke, B. (2020). Regulatory R-loops as facilitators of gene expression Sharma, U., Sun, F., Conine, C. C., Reichholf, B., Kukreja, S., Herzog, V. A., et al.
and genome stability. Nat. Rev. Mol. Cell Biol. 21, 167–178. doi:10.1038/s41580-019- (2018). Small RNAs are trafficked from the epididymis to developing mammalian
0206-3 sperm. Dev. Cell 46, 481–494.e6. doi:10.1016/j.devcel.2018.06.023
O’Leary, N. A., Wright, M. W., Brister, J. R., Ciufo, S., Haddad, D., McVeigh, R., et al. Smit, A. F. A., Hubley, R., and Green, P. (2008). RepeatMasker Open-4.0. In:
(2016). Reference sequence (RefSeq) database at NCBI: Current status, taxonomic RepeatMasker Project [Internet]. 2008 [cited 2021]. Available at: http://www.
expansion, and functional annotation. Nucleic Acids Res. 44, D733–D745. doi:10.1093/ repeatmasker.org (accessed on March 03, 2022).
nar/gkv1189
Soumillon, Magali, Necsulea, Anamaria, Weier, Manuela, Brawand, David, Zhang,
Ohtani, H., and Iwasaki, Y. W. (2021). Rewiring of chromatin state and gene Xiaolan, Gu, Hongcang et al. (2013): Cellular source and mechanisms of high
expression by transposable elements. Dev. Growth Differ. 63, 262–273. doi:10.1111/ transcriptome complexity in the mammalian testis. In: Cell reports 3 (6), S.
dgd.12735 2179–2190. doi:10.1016/j.celrep.2013.05.031
Patankar, A., Gajbhiye, R., Surve, S., and Parte, P. (2021). Epigenetic landscape of Teissandier, A., Servant, N., Barillot, E., and Bourc’his, D. (2019). Tools and best
testis specific histone H2B variant and its influence on sperm function. Clin. Epigenetics practices for retrotransposon analysis using high-throughput sequencing data. Mob.
13, 101. doi:10.1186/s13148-021-01088-4 DNA 10, 52. doi:10.1186/s13100-019-0192-1
Peng, H., Shi, J., Zhang, Y., Zhang, H., Liao, S., Li, W., et al. (2012). A novel class of Torres-Flores, U., and Hernández-Hernández, A. (2020). The interplay between
tRNA-derived small RNAs extremely enriched in mature mouse sperm. Cell Res. 22, replacement and retention of histones in the sperm genome. Front. Genet. 11, 780.
1609–1612. doi:10.1038/cr.2012.141 doi:10.3389/fgene.2020.00780
Powell, W. T., Coulson, R. L., Gonzales, M. L., Crary, F. K., Wong, S. S., Adams, S., Vassena, R., Boué, S., González-Roca, E., Aran, B., Auer, H., Veiga, A., et al. (2011).
et al. (2013). R-loop formation at Snord116 mediates topotecan inhibition of Ube3a- Waves of early transcriptional activation and pluripotency program initiation during
antisense and allele-specific chromatin decondensation. Proc. Natl. Acad. Sci. U.S.A. human preimplantation development. Development 138, 3699–3709. doi:10.1242/dev.
110, 13938–13943. doi:10.1073/pnas.1305426110 064741
Quinlan, A. R., and Hall, I. M. (2010). BEDTools: A flexible suite of utilities for Wang, H., Xing, J., Grover, D., Hedges, D. J., Han, K., Walker, J. A., et al. (2005). SVA
comparing genomic features. Bioinformatics 26, 841–842. doi:10.1093/bioinformatics/ elements: A hominid-specific retroposon family. J. Mol. Biol. 354, 994–1007. doi:10.
btq033 1016/j.jmb.2005.09.085
Rathke, C., Baarends, W. M., Awe, S., and Renkawitz-Pohl, R. (2014). Chromatin Wang, L., Nie, J., Sicotte, H., Li, Y., Eckel-Passow, J. E., Dasari, S., et al. (2016).
dynamics during spermiogenesis. Biochim. Biophys. Acta 1839, 155–168. doi:10.1016/j. Measure transcript integrity using RNA-seq data. BMC Bioinforma. 17, 58. doi:10.1186/
bbagrm.2013.08.004 s12859-016-0922-z
Rathke, C., Baarends, W. M., Jayaramaiah-Raja, S., Bartkuhn, M., Renkawitz, R., and Wang, L., Wang, S., and Li, W. (2012). RseQC: Quality control of RNA-seq
Renkawitz-Pohl, R. (2007). Transition from a nucleosome-based to a protamine-based experiments. Bioinformatics 28, 2184–2185. doi:10.1093/bioinformatics/
chromatin configuration during spermiogenesis in Drosophila. J. Cell Sci. 120, bts356
1689–1700. doi:10.1242/jcs.004663
Wang, Q., Li, M., Wu, T., Zhan, L., Li, L., Chen, M., et al. (2022). Exploring
Ren, X., Chen, X.,Wang, Z., andWang, D. (2017). Is transcription in sperm stationary epigenomic datasets by ChIPseeker. Curr. Protoc. 2, e585. doi:10.1002/cpz1.585
or dynamic? J. Reprod. Dev. 63, 439–443. doi:10.1262/jrd.2016-093
Wang, T., Gao, H., Li, W., and Liu, C. (2019). Essential role of histone
Reznik, Boris, Cincotta, Steven A., Jaszczak, Rebecca G., Mateo, Leslie J., Shen, Joel, replacement and modifications in male fertility. Front. Genet. 10, 962. doi:10.
Cao, Mei et al. (2019). Heterogeneity of transposon expression and activation of the 3389/fgene.2019.00962
repressive network in human fetal germ cells. In: Development (Cambridge, England)
Wei, Y., Yang, C.-R., Wei, Y.-P., Zhao, Z.-A., Hou, Y., Schatten, H., et al.
146 (12). doi:10.1242/dev.171157
(2014). Paternally induced transgenerational inheritance of susceptibility to
Rodriguez-Terrones, D., and Torres-Padilla, M.-E. (2018). Nimble and ready to diabetes in mammals. Proc. Natl. Acad. Sci. U. S. A. 111, e585. doi:10.1002/
mingle: Transposon outbursts of early development. Trends Genet. 34, 806–820. doi:10. cpz1.585
1016/j.tig.2018.06.006
Ward, Michelle C., Wilson, Michael D., Barbosa-Morais, Nuno L., chmidt, Dominic,
Roeschert, I., Poon, E., Henssen, A. G., Garcia, H. D., Gatti, M., Giansanti, C., et al. Stark, Rory, Pan, Qun et al. (2013). Latent regulatory potential of human-specific
(2021). Combined inhibition of Aurora-A and ATR kinase results in regression of repetitive elements. In: Molecular cell 49 (2), S. 262–272. doi:10.1016/j.molcel.2012.
MYCN-amplified neuroblastoma. Nat. Cancer 2, 312–326. doi:10.1038/s43018-020- 11.013
00171-8
Xia, B., Yan, Y., Baron, M., Wagner, F., Barkley, D., Chiodin, M., et al. (2020).
Saitou, M., and Hayashi, K. (2021). Mammalian in vitro gametogenesis. Science 374, Widespread transcriptional scanning in the testis modulates gene evolution rates. Cell
eaaz6830. doi:10.1126/science.aaz6830 180, 248–262.e21. doi:10.1016/j.cell.2019.12.015
Santiago, J., Silva, J. V., Howl, J., Santos, M. A. S., and Fardilha, M. (2021). All you Xie, Z., Bailey, A., Kuleshov, M. V., Clarke, D. J. B., Evangelista, J. E., Jenkins, S. L.,
need to know about sperm RNAs. Hum. Reprod. Update 28, 67–91. doi:10.1093/ et al. (2021). Gene set knowledge discovery with Enrichr. Curr. Protoc. 1, e90. doi:10.
humupd/dmab034 1002/cpz1.90
Sanz, L. A., Hartono, S. R., Lim, Y. W., Steyaert, S., Rajpurkar, A., Ginno, P. A., et al. Xue, Z., Huang, K., Cai, C., Cai, L., Jiang, C., Feng, Y., et al. (2013). Genetic programs
(2016). Prevalent, dynamic, and conserved R-loop structures associate with specific in human and mouse early embryos revealed by single-cell RNA sequencing. Nature
epigenomic signatures in mammals. Mol. Cell 63, 167–178. doi:10.1016/j.molcel.2016. 500, 593–597. doi:10.1038/nature12364
05.032
Yu, G., Wang, L.-G., and He, Q.-Y. (2015). ChIPseeker: An R/bioconductor package
Sanz, Lionel A., and Chédin, Frédéric (2019). High-resolution, strand-specific R-loop for ChIP peak annotation, comparison and visualization. Bioinformatics 31, 2382–2383.
mapping via S9.6-based DNA-RNA immunoprecipitation and high-throughput doi:10.1093/bioinformatics/btv145
sequencing. In: Nature protocols 14 (6), S. 1734–1755. doi:10.1038/s41596-019-0159-1
Zeng, C., Onoguchi, M., and Hamada, M. (2021). Association analysis of repetitive
Savage, A. L., Bubb, V. J., Breen, G., and Quinn, J. P. (2013). Characterisation of the elements and R-loop formation across species. Mob. DNA 12, 3. doi:10.1186/s13100-
potential function of SVA retrotransposons to modulate gene expression patterns. BMC 021-00231-5
Evol. Biol. 13, 101. doi:10.1186/1471-2148-13-101
Zhang, P., Zucchelli, M., Bruce, S., Hambiliki, F., Stavreus-Evers, A., Levkov, L., et al.
Schagdarsurengin, U., Paradowska, A., and Steger, K. (2012). Analysing the sperm (2009). Transcriptome profiling of human pre-implantation development. PLoS One 4,
epigenome: Roles in early embryogenesis and assisted reproduction. Nat. Rev. Urol. 9, e7844. doi:10.1371/journal.pone.0007844
609–619. doi:10.1038/nrurol.2012.183
Zhang, Y., Romanish, M. T., and Mager, D. L. (2011). Distributions of transposable
Schmid, C. W. (1991). Human Alu subfamilies and their methylation revealed by blot elements reveal hazardous zones in mammalian introns. PLoS Comput. Biol. 7,
hybridization. Nucleic Acids Res. 19, 5613–5617. doi:10.1093/nar/19.20.5613 e1002046. doi:10.1371/journal.pcbi.1002046
Frontiers in Genetics 14 frontiersin.org