Fig. 1. Identification of PRDM9-dependent H3K4me3 modifications in the mouse genome.

a Schematic workflow. Male mice carrying both Vasa-mCherry and Lin28-YFP alleles were treated with WIN 18,446/retinoic acid to synchronize spermatogenesis. Testes were then digested and the synchronous spermatogenic cells were sorted by flow cytometry. In total, spermatogenic cells at ten stages, including mitotic cells (Undiff undifferentiated spermatogonia, A1 type A1 spermatogonia, B type B spermatogonia), meiotic S phase cells (pL preleptotene spermatocytes) and meiotic prophase I cells (L leptotene spermatocytes, mZ mid-zygotene spermatocytes, lZ late-zygotene spermatocytes, eP early-pachytene spermatocytes, mP mid-pachytene spermatocytes, D diplotene spermatocytes) were collected for ChIP-seq and NOMe-seq analyses. b Numbers of the newly generated (new) and common (common) H3K4me3 peaks in leptotene (L) and mid-zygotene (mZ) spermatocytes. c Venn diagram showing the overlap of the newly generated H3K4me3 peaks in mid-zygotene with PRDM9-binding sites (PRDM9 affinity-seq data). d Heatmap of H3K4me3 normalized tag density on PRDM9-binding sites. Each row represents a PRDM9-binding site of ± 5 kb around the center and ranked by PRMD9 tag density from the highest to the lowest. H3K4me3 and PRDM9 tag densities were calculated using H3K4me3 ChIP-seq reads or PRDM9 affinity-seq reads with 50-bp resolution. PRDM9 PRDM9-binding sites, e1P early 1-pachytene.