HENMT1 and piRNA Stability Are Required for Adult Male Germ Cell Transposon Repression and to Define the Spermatogenic Program in the Mouse

Abstract

piRNAs are critical for transposable element (TE) repression and germ cell survival during the early phases of spermatogenesis, however, their role in adult germ cells and the relative importance of piRNA methylation is poorly defined in mammals. Using a mouse model of HEN methyltransferase 1 (HENMT1) loss-of-function, RNA-Seq and a range of RNA assays we show that HENMT1 is required for the 2' O-methylation of mammalian piRNAs. HENMT1 loss leads to piRNA instability, reduced piRNA bulk and length, and ultimately male sterility characterized by a germ cell arrest at the elongating germ cell phase of spermatogenesis. HENMT1 loss-of-function, and the concomitant loss of piRNAs, resulted in TE de-repression in adult meiotic and haploid germ cells, and the precocious, and selective, expression of many haploid-transcripts in meiotic cells. Precocious expression was associated with a more active chromatin state in meiotic cells, elevated levels of DNA damage and a catastrophic deregulation of the haploid germ cell gene expression. Collectively these results define a critical role for HENMT1 and piRNAs in the maintenance of TE repression in adult germ cells and setting the spermatogenic program.

Fig 1. The Henmt1 ^PIN/PIN mutation resulted in exon 3 skipping, the production of unstable Henmt1 mRNAs and an absence of HENMT1 protein in the testis.

(A) A schematic representation of the three Henmt1 transcripts and the effect of the Henmt1 ^PIN/PIN on exon 3 splicing. Boxes represent exons, open bars represent introns. (B) RT-PCR analysis using primers located in exons 2 and 4 confirmed the skipping of exon 3 in the Henmt1 ^PIN allele. (C) qPCR analysis of Henmt1 transcript 1 and 2 levels in Henmt1 ^PIN/PIN and Henmt1 ^WT/WT spermatocytes and round spermatids (round). 8 weeks old, n = 3 /genotype (* p<0.05, mean ± SD). (D) The relative expression of Henmt1 transcripts 1 and 2, and transcript 3 in Henmt1 ^WT/WT spermatocytes and round spermatids. 8 weeks old. n = 3 /genotype (p*<0.05, ** p<0.01, mean ± SD). (E) HENMT1 localization (red) in Henmt1 ^WT/WT (upper two panels) and 8 week old Henmt1 ^PIN/PIN testes (lower two panels). DNA is labelled with DAPI (blue). (F) A representative western blot for HENMT1 in 8 week old Henmt1 ^WT/WT and Henmt1 ^PIN/PIN testes. Beta actin was used as a loading control. Where relevant a two-tailed unpaired student T-test was performed for statistical analyses. Scale bars = 50μm.

Fig 2. HENMT1 is required for mouse spermatogenesis.

(A-D) Periodic acid-Schiff (PAS) stained testis sections from Henmt1 ^WT/WT and Henmt1 ^PIN/PIN mice. Arrows indicate the presence of pinhead-shaped sperm heads. Arrow heads indicate symplasts composed of coalesced germ cells. (E) Testis weight in Henmt1 ^WT/WT and Henmt1 ^PIN/PIN mice (n = 6 / genotype +/- SEM, * p<0.05). (F) Daily sperm production in Henmt1 ^WT/WT and Henmt1 ^PIN/PIN mice expressed as percentage of Henmt1 ^WT/WT (n = 6 / genotype +/- SEM, ** p<0.01) (G-H) PAS stained sections of epididymides from Henmt1 ^WT/WT and Henmt1 ^PIN/PIN mice. (I) Epididymal sperm content in Henmt1 ^WT/WT and Henmt1 ^PIN/PIN mice expressed as percentage of Henmt1 ^WT/WT (n = 3 / genotype +/- SD, ****p<0.0001). (J) Hematoxylin and eosin stained sperm from Henmt1 ^WT/WT and Henmt1 ^PIN/PIN mice showing the pin-shaped heads and the absence of a mitochondrial sheath (open arrow) in Henmt1 ^PIN/PIN sperm. (K) The percentage of sperm showing any forms of motility in Henmt1 ^WT/WT and Henmt1 ^PIN/PIN mice (n = 3 / genotype +/- SD, ****p<0.0001). (L) The percentage of sperm showing progressive motility in Henmt1 ^WT/WT and Henmt1 ^PIN/PIN mice (n = 3 / genotype +/- SD, ****p<0.001). (M-N) Electron microscopy of stage VIII pachytene spermatocytes showing the presence of inter-mitochondrial cement (arrow) and the chromatoid body (red square) in Henmt1 ^WT/WT and Henmt1 ^PIN/PIN mice. All data was collected from 10 week-old mice. A two-tailed unpaired student T-test was performed for statistical analyses.

Fig 3. HENMT1 loss results in decreased piRNA bulk and the absence of 3’ 2’-O-methylation and stability.

(A-B) A representative image illustrating a ~51% reduction in the abundance of piRNAs in Henmt1 ^PIN/PIN and Henmt1 ^WT/WT 30 day-old testes as indicated on an acrylamide gel (total piRNAs, A) and a small RNA northern blotting for pachytene piRNA1 (B). 5.8S and 5S rRNAs were used as loading controls. Statistics were carried out on n-4 biological replicates per genotype. (C) The effects of β-elimination on piRNA length. 5.8S and 5S rRNAs were used as loading controls. (D-E) piRNA length distributions, expressed as a percentage of total, in Henmt1 ^WT/WT and Henmt1 ^PIN/PIN. (F-G) The percentage of piRNAs with (F) adenylation (last base of read was A when the reference base was not A) and (G) uridylation (last base of read was T, reference base was not T) at their 3’ end in spermatocytes and round spermatids (n = 2 / Henmt1 ^WT/WT spermatocytes, n = 3 Henmt1 ^WT/WT round, Henmt1 ^PIN/PIN spermatocytes and round +/- SD, * p<0.05; ** p<0.01; ***p<0.001; ****p<0.0001). Round = round spermatids. A two-tailed unpaired student T-test was performed for statistical analysis.

Fig 4. HENMT1 is required for the repression of Line-1 and IAP retrotransposons in spermatocytes and round spermatids.

(A-D) qPCR analyses of (A) L1_A (B) L1_TF14 (C) IAP_LTR (D) IAP_GAG in Henmt1 ^WT/WT and Henmt1 ^PIN/PIN spermatocytes and round spermatids (round) purified from 30 day old mice. A two-tailed unpaired student T-test was performed for statistical analysis, n = 5/genotype, +/- SEM, * p<0.05; ** p<0.01; ***p<0.001; ****p<0.0001. (E-F) RNA in situ hybridisation for Line-1 expression in Henmt1 ^WT/WT and Henmt1 ^PIN/PIN 10 weeks-old testes. Black arrows indicate pachytene spermatocytes. Red arrows indicate round spermatids. (G) Staining for LINE-1 in in Henmt1 ^WT/WT and Henmt1 ^PIN/PIN in 10 weeks-old testes. The boxed areas are magnified in the panel to the immediate right of each genotype. The scale bar = 50 μm. (H) Immunofluorescence staining for γH2AX (red) as a marker of DNA double stranded break respectively in Henmt1 ^WT/WT and Henmt1 ^PIN/PIN in 10 weeks-old testes. The white arrow indicates elevated γH2AX staining in spermatids compared to wild type cells. The boxed areas are magnified in the panel to the immediate right of each genotype. The scale bar = 50 μm.

Fig 5. Altered gene expression in Henmt1 ^PIN/PIN spermatocytes and round spermatids.

(A) A heatmap showing log2 fold change in differentially expressed genes in spermatocytes and their corresponding log2 fold change in round spermatids (round). Color coding represent log2 fold changes of genes. Prm1, Tnp1, Tnp2, Prm2 indicate the position of transcripts subjected to further analysis. (B) A volcano plot showing expression changes of all detected genes in spermatocytes and round spermatids. Dots in cyan represent significantly differential expressed genes (FDR < 0.05) based on edgeR. Fold changes and p-values were calculated with edgeR. (C-G) qPCR of spermiogenic genes including (C) Tnp1 (D) Tnp2 (E) Prm1 (F) Prm2 (G) Gapdhs in 28 day-old Henmt1 ^WT/WT and Henmt1 ^PIN/PIN spermatocytes and round spermatids (n = 5 / genotype +/- SEM, * p<0.05, **p<0.01). White bar represents Henmt1 ^WT/WT and black bar is Henmt1 ^PIN/PIN. S’cytes = spermatocytes, S’tids = round spermatids. A two-tailed unpaired student T-test was performed for statistical analyses.

Fig 6. Enrichment of active histone mark at the promoter regions of spermiogenic genes in Henmt1 ^PIN/PIN spermatocytes.

ChIP and qPCR analyses were performed on Henmt1 ^WT/WT and Henmt1 ^PIN/PIN spermatocytes (n = 5/genotypes, 3 biological replicates, * p<0.05, **p<0.01, *** p<0.001, **** p<0.0001). qPCR for the promoter regions of (A) Tnp1, (B) Tnp2, (C) Prm1, (D) Prm2, (E) Gapdhs, and (F) Ppia (as a house keeping control). Histone enrichment was normalised to histone H3. The data is presented in the ratio of Henmt1 ^PIN/PIN enrichment/ Henmt1 ^WT/WT enrichment. A two-tailed unpaired student T-test was performed for statistical analyses.