The testis-enriched histone demethylase, KDM4D, regulates methylation of histone H3 lysine 9 during spermatogenesis in the mouse but is dispensable for fertility

Abstract

Epigenetic modifications, and methylation of histones in particular, dynamically change during spermatogenesis. Among various methylations of histone H3, methylation of histone H3 lysine 9 (H3K9) and its regulation are essential for spermatogenesis. Trimethytransferases as well as dimethyltransferase are required for meiotic progression. In addition, didemethylase of H3K9 is also critical for spermatogenesis through transcriptional regulation of spermatid-specific genes. However, the requirement for demethylation of trimethylated H3K9 (H3K9me3) during spermatogenesis remains to be elucidated. Here, we report the targeted disruption of KDM4D, a testis-enriched tridemethylase of H3K9. Kdm4d-null mice are viable and fertile and do not show any obvious phenotype. However, H3K9me3 accumulates significantly in Kdm4d-null round spermatids, and the distribution of methylated H3K9 in germ cells is dramatically changed. Nevertheless, the progression of spermatogenesis and the number of spermatozoa are normal, likely secondary to the earlier nuclear localization of another H3K9 tridemethylase, KDM4B, in Kdm4d-null elongating spermatids. These results suggest that demethylation of H3K9me3 in round spermatids is dispensable for spermatogenesis but that possible defects in Kdm4d-null elongating spermatids could be rescued by functional redundancy of the KDM4B demethylase.

FIG. 1.

Spatiotemporal expression of Kdm4d. A and B) Transcripts of Kdm4d were examined by semiquantitative RT-PCR in multiple tissue samples (A) and during postnatal development of the testis (B). Hprt was used as an internal control. GS, germline stem cells. C) In situ hybridization of Kdm4d. An antisense probe (left) and a sense probe (right) specific for Kdm4d were hybridized.

FIG. 2.

Demethylation activity of KDM4D. A) Western blot of methylated histones in HeLa cells expressing KDM4D. Extracted histones from 1 × 10⁵ HeLa cells expressing mCherry or mCherry-KDM4D were loaded. Fold-exchanges of each histone modification in KDM4D-expressing cells are indicated at the right of each picture. Ponceau staining is shown as a loading control. AcH3, acetylated histone H3. B and C) Immunostaining of methylated histones in HeLa cells expressing KDM4D. HeLa cells expressing mCherry (B) or mCherry-KDM4D (C) were fixed and stained with indicated antibodies. An mCherry or an mCherry-KDM4D (red; left), methylated histone (green; middle), and a merged image (right) are shown.

FIG. 3.

Generation of Kdm4d-null mice. A) Strategy of targeting of the Kdm4d gene. The ORF of Kdm4d was replaced with a PGK1-Neo cassette. B and C) Southern blot analyses using 5′ (B) and 3′ (C) external probes. EcoRI-digested (B) and SphI-digested (C) tail DNA were hybridized with probes that detect a 13.3-kb wild-type (WT) and 9.2-kb null alleles at the 5′ end (B) or an 11.7-kb WT and a 7.4-kb null alleles at the 3′ end (C). D) Confirmation of the Kdm4d-null allele. RT-PCR using primers spanning exon 1 and the 5′ untranslated region of exon 2 and primers within the ORF of Kdm4d are shown. DTA, diphtheria toxin fragment A; KO, knockout (homozygote); +/–, heterozygote.

FIG. 4.

Alterations in methylated histone H3 in Kdm4d^−/− testis. A) Western blot analyses of H3K9me1, H3K9me2, H3K9me3, H3K4me3, and total histone H3 in Kdm4d^+/− and Kdm4d^−/− germ cell-enriched testes, pachytene spermatocytes (Pach), round spermatids (RS), and elongated spermatids (ES) are shown. B) Quantitative changes of H3K9me1, H3K9me2, H3K9me3, and H3K4me3 are shown in A. Relative fold-exchanges of H3K9me1, H3K9me2, H3K9me3, and H3K4me3 in Kdm4d^−/− germ cells are shown. *P < 0.05. C) Low- and high-magnification of immunofluorescence images of H3K9me1, H3K9me2, and H3K9me3 in Kdm4d^+/− and Kdm4d^−/− testes at 3 wk and 10 wk of age. Roman numerals indicate spermatogenic stages.

FIG. 5.

Fertility analysis and sperm counts of Kdm4d knockout mice. A) Average number of pups produced by Kdm4d^+/− (+/−) and Kdm4d^−/− (KO) males over 9 mo of breeding (n = 10 per genotype). B and C) Average litter size (B) and litters per month (C) produced by Kdm4d^+/− (+/−) and Kdm4d^−/− (KO) males. D) Number of total epididymal sperm in 6-mo-old Kdm4d^+/− and Kdm4d^−/− males (n = 4 per genotype).

FIG. 6.

Phenotype of Kdm4d^−/− testes. A and B) Gross pictures of Kdm4d^+/− (+/−) and Kdm4d^−/− (KO) testes at 6 wk (A) and 12 wk (B) of age. C and D) Testis (C) and body (D) weights of Kdm4d^+/− mice (green bar) and Kdm4d^−/− mice (red bar) at indicated ages (n ≥ 5 mice). E–J) Histological analysis. Testes from Kdm4d^+/− (E–G) and Kdm4d^−/− (H–J) mice at 3 wk (E and H), 12 wk (F and I), and 1 yr (G and J) are shown. Roman numerals indicate spermatogenic stages. Arrows indicate apoptotic cells. *P < 0.01, **P < 0.05.

FIG. 7.

Methylation of H3K9 in spermatocytes. Kdm4d^+/− and Kdm4d^−/− zygotene and pachytene spermatocytes were stained with anti-H3K9me2 or anti-H3K9me3 (green) and anti-SCP3 (red) antibodies and DAPI (blue). Images shown are merged images of each staining and DAPI.

FIG. 8.

Expression of KDM4B in Kdm4d^+/− and Kdm4d^−/− testes. Testes from Kdm4d^+/− and Kdm4d^−/− mice at 25 days and 10 wk of age were stained with anti-KDM4B (green), anti-H2AFX (red), and DAPI (blue).