Recapitulation of gametic DNA methylation and its post-fertilization maintenance with reassembled DNA elements at the mouse Igf2/H19 locus

Imprinted DNA methylation of the H19 ICR is established and maintained in vivo in the absence of Kaiso

BACKGROUND

Paternal allele-specific DNA methylation of the imprinting control region (H19 ICR) controls genomic imprinting at the Igf2/H19 locus. We previously demonstrated that the mouse H19 ICR transgene acquires imprinted DNA methylation in preimplantation mouse embryos. This activity is also present in the endogenous H19 ICR and protects it from genome-wide reprogramming after fertilization. We also identified a 118-bp sequence within the H19 ICR that is responsible for post-fertilization imprinted methylation. Two mutations, one in the five RCTG motifs and the other a 36-bp deletion both in the 118-bp segment, caused complete and partial loss, respectively, of methylation following paternal transmission in each transgenic mouse. Interestingly, these mutations overlap with the binding site for the transcription factor Kaiso, which is reportedly involved in maintaining paternal methylation at the human H19 ICR (IC1) in cultured cells. In this study, we investigated if Kaiso regulates imprinted DNA methylation of the H19 ICR in vivo.

RESULTS

Neither Kaiso deletion nor mutation of Kaiso binding sites in the 118-bp region affected DNA methylation of the mouse H19 ICR transgene. The endogenous mouse H19 ICR was methylated in a wild-type manner in Kaiso-null mutant mice. Additionally, the human IC1 transgene acquired imprinted DNA methylation after fertilization in the absence of Kaiso.

CONCLUSIONS

Our results indicate that Kaiso is not essential for either post-fertilization imprinted DNA methylation of the transgenic H19 ICR in mouse or for methylation imprinting of the endogenous mouse H19 ICR.

Effects of Tcte1 knockout on energy chain transportation and spermatogenesis: implications for male infertility

STUDY QUESTION

Is the Tcte1 mutation causative for male infertility?

SUMMARY ANSWER

Our collected data underline the complex and devastating effect of the single-gene mutation on the testicular molecular network, leading to male reproductive failure.

WHAT IS KNOWN ALREADY

Recent data have revealed mutations in genes related to axonemal dynein arms as causative for morphology and motility abnormalities in spermatozoa of infertile males, including dysplasia of fibrous sheath (DFS) and multiple morphological abnormalities in the sperm flagella (MMAF). The nexin-dynein regulatory complex (N-DRC) coordinates the dynein arm activity and is built from the DRC1-DRC7 proteins. DRC5 (TCTE1), one of the N-DRC elements, has already been reported as a candidate for abnormal sperm flagella beating; however, only in a restricted manner with no clear explanation of respective observations.

STUDY DESIGN SIZE DURATION

Using the CRISPR/Cas9 genome editing technique, a mouse Tcte1 gene knockout line was created on the basis of the C57Bl/6J strain. The mouse reproductive potential, semen characteristics, testicular gene expression levels, sperm ATP, and testis apoptosis level measurements were then assessed, followed by visualization of N-DRC proteins in sperm, and protein modeling in silico. Also, a pilot genomic sequencing study of samples from human infertile males (n = 248) was applied for screening of TCTE1 variants.

PARTICIPANTS/MATERIALS SETTING METHODS

To check the reproductive potential of KO mice, adult animals were crossed for delivery of three litters per caged pair, but for no longer than for 6 months, in various combinations of zygosity. All experiments were performed for wild-type (WT, control group), heterozygous Tcte1+/- and homozygous Tcte1-/- male mice. Gross anatomy was performed on testis and epididymis samples, followed by semen analysis. Sequencing of RNA (RNAseq; Illumina) was done for mice testis tissues. STRING interactions were checked for protein-protein interactions, based on changed expression levels of corresponding genes identified in the mouse testis RNAseq experiments. Immunofluorescence in situ staining was performed to detect the N-DRC complex proteins: Tcte1 (Drc5), Drc7, Fbxl13 (Drc6), and Eps8l1 (Drc3) in mouse spermatozoa. To determine the amount of ATP in spermatozoa, the luminescence level was measured. In addition, immunofluorescence in situ staining was performed to check the level of apoptosis via caspase 3 visualization on mouse testis samples. DNA from whole blood samples of infertile males (n = 137 with non-obstructive azoospermia or cryptozoospermia, n = 111 samples with a spectrum of oligoasthenoteratozoospermia, including n = 47 with asthenozoospermia) was extracted to perform genomic sequencing (WGS, WES, or Sanger). Protein prediction modeling of human-identified variants and the exon 3 structure deleted in the mouse knockout was also performed.

MAIN RESULTS AND THE ROLE OF CHANCE

No progeny at all was found for the homozygous males which were revealed to have oligoasthenoteratozoospermia, while heterozygous animals were fertile but manifested oligozoospermia, suggesting haploinsufficiency. RNA-sequencing of the testicular tissue showed the influence of Tcte1 mutations on the expression pattern of 21 genes responsible for mitochondrial ATP processing or linked with apoptosis or spermatogenesis. In Tcte1-/- males, the protein was revealed in only residual amounts in the sperm head nucleus and was not transported to the sperm flagella, as were other N-DRC components. Decreased ATP levels (2.4-fold lower) were found in the spermatozoa of homozygous mice, together with disturbed tail:midpiece ratios, leading to abnormal sperm tail beating. Casp3-positive signals (indicating apoptosis) were observed in spermatogonia only, at a similar level in all three mouse genotypes. Mutation screening of human infertile males revealed one novel and five ultra-rare heterogeneous variants (predicted as disease-causing) in 6.05% of the patients studied. Protein prediction modeling of identified variants revealed changes in the protein surface charge potential, leading to disruption in helix flexibility or its dynamics, thus suggesting disrupted interactions of TCTE1 with its binding partners located within the axoneme.

LARGE SCALE DATA

All data generated or analyzed during this study are included in this published article and its supplementary information files. RNAseq data are available in the GEO database (https://www.ncbi.nlm.nih.gov/geo/) under the accession number GSE207805. The results described in the publication are based on whole-genome or exome sequencing data which includes sensitive information in the form of patient-specific germline variants. Information regarding such variants must not be shared publicly following European Union legislation, therefore access to raw data that support the findings of this study are available from the corresponding author upon reasonable request.

LIMITATIONS REASONS FOR CAUTION

In the study, the in vitro fertilization performance of sperm from homozygous male mice was not checked.

WIDER IMPLICATIONS OF THE FINDINGS

This study contains novel and comprehensive data concerning the role of TCTE1 in male infertility. The TCTE1 gene is the next one that should be added to the 'male infertility list' because of its crucial role in spermatogenesis and proper sperm functioning.

STUDY FUNDING/COMPETING INTERESTS

This work was supported by National Science Centre in Poland, grants no.: 2015/17/B/NZ2/01157 and 2020/37/B/NZ5/00549 (to M.K.), 2017/26/D/NZ5/00789 (to A.M.), and HD096723, GM127569-03, NIH SAP #4100085736 PA DoH (to A.N.Y.). The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Assisted reproductive technology and imprinting errors: analyzing underlying mechanisms from epigenetic regulation

With the increasing maturity and widespread application of assisted reproductive technology (ART), more attention has been paid to the health outcomes of offspring following ART. It is well established that children born from ART treatment are at an increased risk of imprinting errors and imprinting disorders. The disturbances of genetic imprinting are attributed to the overlap of ART procedures and important epigenetic reprogramming events during the development of gametes and early embryos, but the detailed mechanisms are hitherto obscure. In this review, we summarized the DNA methylation-dependent and independent mechanisms that control the dynamic epigenetic regulation of imprinted genes throughout the life cycle of a mammal, including erasure, establishment, and maintenance. In addition, we systematically described the dysregulation of imprinted genes in embryos conceived through ART and discussed the corresponding underlying mechanisms according to findings in animal models. This work is conducive to evaluating and improving the safety of ART.

Five nucleotides found in RCTG motifs are essential for post-fertilization methylation imprinting of the H19 ICR in YAC transgenic mice

Genomic imprinting at the mouse Igf2/H19 locus is controlled by the H19 ICR, within which paternal allele-specific DNA methylation originating in sperm is maintained throughout development in offspring. We previously found that a 2.9 kb transgenic H19 ICR fragment in mice can be methylated de novo after fertilization only when paternally inherited, despite its unmethylated state in sperm. When the 118 bp sequence responsible for this methylation in transgenic mice was deleted from the endogenous H19 ICR, the methylation level of its paternal allele was significantly reduced after fertilization, suggesting the activity involving this 118 bp sequence is required for methylation maintenance at the endogenous locus. Here, we determined protein binding to the 118 bp sequence using an in vitro binding assay and inferred the binding motif to be RCTG by using a series of mutant competitors. Furthermore, we generated H19 ICR transgenic mice with a 5-bp substitution mutation that disrupts the RCTG motifs within the 118 bp sequence, and observed loss of methylation from the paternally inherited transgene. These results indicate that imprinted methylation of the H19 ICR established de novo during the post-fertilization period involves binding of specific factors to distinct sequence motifs within the 118 bp sequence.

The transgenic IG-DMR sequence of the mouse Dlk1-Dio3 domain acquired imprinted DNA methylation during the post-fertilization period

BACKGROUND

Allele-specific methylation of the imprinting control region (ICR) is the molecular basis for the genomic imprinting phenomenon that is unique to placental mammals. We previously showed that the ICR at the mouse H19 gene locus (H19 ICR) was unexpectedly established after fertilization and not during spermatogenesis in transgenic mice (TgM), and that the same activity was essential for the maintenance of paternal methylation of the H19 ICR at the endogenous locus in pre-implantation embryos. To examine the universality of post-fertilization imprinted methylation across animal species or imprinted loci, we generated TgM with two additional sequences.

RESULTS

The rat H19 ICR, which is very similar in structure to the mouse H19 ICR, unexpectedly did not acquire imprinted methylation even after fertilization, suggesting a lack of essential sequences in the transgene fragment. In contrast, the mouse IG-DMR, the methylation of which is acquired during spermatogenesis at the endogenous locus, did not acquire methylation in the sperm of TgM, yet became highly methylated in blastocysts after fertilization, but only when the transgene was paternally inherited. Since these two sequences were evaluated at the same genomic site by employing the transgene co-placement strategy, it is likely that the phenotype reflects the intrinsic activity of these fragments rather than position-effect variegation.

CONCLUSIONS

Our results suggested that post-fertilization imprinted methylation is a versatile mechanism for protecting paternal imprinted methylation from reprogramming during the pre-implantation period.

Orientation of mouse H19 ICR affects imprinted H19 gene expression through promoter methylation-dependent and -independent mechanisms

The mouse Igf2/H19 locus is regulated by genomic imprinting, in which the paternally methylated H19 imprinting control region (ICR) plays a critical role in mono-allelic expression of the genes in the locus. Although the maternal allele-specific insulator activity of the H19 ICR in regulating imprinted Igf2 expression has been well established, the detailed mechanism by which the H19 ICR controls mono-allelic H19 gene expression has not been fully elucidated. In this study, we evaluated the effect of H19 ICR orientation on imprinting regulation in mutant mice in which the H19 ICR sequence was inverted at the endogenous locus. When the inverted-ICR allele was paternally inherited, the methylation level of the H19 promoter was decreased and the H19 gene was derepressed, suggesting that methylation of the H19 promoter is essential for complete repression of H19 gene expression. Unexpectedly, when the inverted allele was maternally inherited, the expression level of the H19 gene was lower than that of the WT allele, even though the H19 promoter remained fully hypomethylated. These observations suggested that the polarity of the H19 ICR is involved in controlling imprinted H19 gene expression on each parental allele, dependent or independent on DNA methylation of the H19 promoter.

[Autosomal dominant intellectual disability type 21 in a neonate]

This is a case report on a 1-day-old male neonate admitted due to a weak cry for 1 day and recurrent circumoral cyanosis for 2 hours. He had unusual facial features at birth, with a single transverse palmar crease on both hands, flat feet, weak cry, feeding difficulties, congenital heart disease, and abnormality on cerebral MRI. Whole exome sequencing showed a de novo mutation, c.778_781delAAAG(p.Lys260ValfsTer2), in exon 3 of the CTCF gene, which was considered a pathogenic mutation by protein function prediction and might damage the function of CTCF protein. He was diagnosed with autosomal dominant intellectual disability type 21 based on the clinical manifestations and genetic analysis results. This case suggests that genetic analysis should be performed as early as possible for neonates with feeding difficulties which cannot be explained by infection or hypoxia, so as to help with early diagnosis and genetic counselling.

Transient establishment of imprinted DNA methylation of transgenic human IC1 sequence in mouse during the preimplantation period

Monoallelic gene expression at the Igf2/H19 locus is controlled by paternal allele-specific DNA methylation of the imprinting control region (H19 ICR) that is established during spermatogenesis. We demonstrated that the H19 ICR fragment in transgenic mice acquires allele-specific methylation only after fertilization, which is essential for maintaining its allelic methylation during early embryogenesis. We identified a DNA element required for establishing postfertilization methylation within a 118 bp (m118) region. A previously generated knock-in mouse whose endogenous H19 ICR was substituted with the human H19 ICR (hIC1; 4.8 kb) sequence revealed that the hIC1 sequence was partially methylated in sperm, although this methylation was lost by the blastocyst stage, which we assume is due to a lack of an m118-equivalent sequence in the hIC1 transgene. To identify a cis sequence involved in postfertilization methylation within the hIC1 region, we generated three transgenic mouse lines (TgM): one carrying an 8.8 kb hIC1 sequence joined to m118 (hIC1+m118), one with the 8.8 kb hIC1 and one with the 5.8 kb hIC1 sequence joined to m118 (hIC1-3'+m118). We found that the hIC1-3' region was resistant to de novo DNA methylation throughout development. In contrast, the 5' portion of the hIC1 (hIC1-5') in both hIC1+m118 and hIC1 TgM were preferentially methylated on the paternal allele only during preimplantation. As DNA methylation levels were higher in hIC1+m118, the m118 sequence could also induce imprinted methylation of the human sequence. Most importantly, the hIC1-5' sequence appears to possess an activity equivalent to that of m118.

Maternal DNMT3A-dependent de novo methylation of the paternal genome inhibits gene expression in the early embryo

De novo DNA methylation (DNAme) during mammalian spermatogenesis yields a densely methylated genome, with the exception of CpG islands (CGIs), which are hypomethylated in sperm. While the paternal genome undergoes widespread DNAme loss before the first S-phase following fertilization, recent mass spectrometry analysis revealed that the zygotic paternal genome is paradoxically also subject to a low level of de novo DNAme. However, the loci involved, and impact on transcription were not addressed. Here, we employ allele-specific analysis of whole-genome bisulphite sequencing data and show that a number of genomic regions, including several dozen CGI promoters, are de novo methylated on the paternal genome by the 2-cell stage. A subset of these promoters maintains DNAme through development to the blastocyst stage. Consistent with paternal DNAme acquisition, many of these loci are hypermethylated in androgenetic blastocysts but hypomethylated in parthenogenetic blastocysts. Paternal DNAme acquisition is lost following maternal deletion of Dnmt3a, with a subset of promoters, which are normally transcribed from the paternal allele in blastocysts, being prematurely transcribed at the 4-cell stage in maternal Dnmt3a knockout embryos. These observations uncover a role for maternal DNMT3A activity in post-fertilization epigenetic reprogramming and transcriptional silencing of the paternal genome.

The paternal genome in mice undergoes widespread DNA methylation loss post-fertilization. Here, the authors apply allele-specific analysis of WGBS data to show that a number of genomic regions are simultaneously de novo methylated on the paternal genome dependent on maternal DNMT3A activity, which induces transcriptional silencing of this allele in the early embryo.