The Lid/KDM5 histone demethylase complex activates a critical effector of the oocyte-to-zygote transition

Abstract

Following fertilization of a mature oocyte, the formation of a diploid zygote involves a series of coordinated cellular events that ends with the first embryonic mitosis. In animals, this complex developmental transition is almost entirely controlled by maternal gene products. How such a crucial transcriptional program is established during oogenesis remains poorly understood. Here, we have performed an shRNA-based genetic screen in Drosophila to identify genes required to form a diploid zygote. We found that the Lid/KDM5 histone demethylase and its partner, the Sin3A-HDAC1 deacetylase complex, are necessary for sperm nuclear decompaction and karyogamy. Surprisingly, transcriptomic analyses revealed that these histone modifiers are required for the massive transcriptional activation of deadhead (dhd), which encodes a maternal thioredoxin involved in sperm chromatin remodeling. Unexpectedly, while lid knock-down tends to slightly favor the accumulation of its target, H3K4me3, on the genome, this mark was lost at the dhd locus. We propose that Lid/KDM5 and Sin3A cooperate to establish a local chromatin environment facilitating the unusually high expression of dhd, a key effector of the oocyte-to-zygote transition.

Fig 1. Lid is required for sperm nuclear decompaction at fertilization.

A—Scheme of shRNA screen in the female germline. B—Left: Zygotic expression of a paternally-derived P[g-GFP::cid] transgene in a control embryo. GFP::Cid marks centromeric chromatin and is visible as nuclear dots. Right: lid KD females produce embryos that fail to express the paternally-inherited transgene. Scale bar: 25 μm. C—Maternal Lid is required for SNBP removal and sperm nuclear decompaction at fertilization. Left: a control egg at pronuclear apposition. Both pronuclei (inset) appear similar in size and shape and the SNBP marker ProtA::GFP is not detected. Middle: A representative lid KD egg containing a needle-shaped sperm nucleus (inset) still packaged with ProtA::GFP. Right: A fertilized lid KD egg with the sperm nucleus apposed to the female pronucleus. Scale bar: 10 μm. D—SNBP replacement with histones is impaired in lid KD eggs. Left: Confocal images of representative sperm nuclei in lid KD eggs. Partially decondensed nuclei are positive for histones. Scale bar: 5 μm. Right: Quantification of sperm nuclear phenotype in control and lid KD eggs.

Fig 2. deadhead is strongly downregulated in lid KD and Sin3A KD ovaries.

A—Scheme of a pair of adult ovaries with two isolated ovarioles and an egg chamber (inset). Germline nuclei are in blue. Oo: Oocyte, Nc: Nurse cells, Fc: Follicle cells. B—Venn diagram showing the number of differentially expressed genes in lid KD and Sin3A KD ovarian transcriptomes (FDR<0.001). C—Comparison of RNA seq normalized reads per gene (DESeq2) are shown for lid KD vs Control (left) and Sin3A KD vs Control (right). Genes with a negative foldchange (downregulated in KD) are in green (FDR<0.001). Genes with a positive foldchange (upregulated in KD) are in red (FDR<0.001). D—Integrative Genomics Viewer (igv) view of Control, dhd^[J5], lid KD and Sin3A KD ovarian RNA Seq signal on the dhd region. The Df(1)J5 deficiency is indicated as an interrupted baseline on the dhd^[J5] track.

Fig 3. H3K4me3 ovarian ChIP-Seq analysis.

A–H3K4me3 enrichment around peak center for Control and lid KD ovaries. Upper panels show the average profile around detected peak centers. Lower panels show read density heatmaps around the detected peak centers. B—H3K4me3 enrichment around gene loci for Control and lid KD ovaries. Upper panels show the average signal profile on genomic loci defined as 3kb upstream of annotated TSS to 3kb downstream of annotated TES. Lower panels show read density heatmaps around the same genomic loci. C—igv view of H3K4me3 occupancy on the dhd genomic region in Control and lid KD ovaries.

Fig 4. Forced expression of dhd partially rescues the lid KD phenotype.

A—Left: RT-qPCR quantification of dhd mRNA levels in ovaries of indicated genotypes (normalized to rp49 and relative to expression in w¹¹¹⁸). Data are presented as mean ± SD of 2 biological replicates. P values indicate one-way ANOVA with Dunnett’s multiple comparisons test to a control (**** P < 0.0001; ** P < 0.01, * P < 0.03; n.s = not significant). Right: Western blot analysis of DHD expression in ovaries of indicated genotypes. Alpha-tubulin detection is used as a loading control in Western-blotting. B—Quantification of sperm nuclear phenotype in eggs laid by females of indicated genotypes. C—Confocal images of a MTD>lid22, P[gnu-dhd] haploid embryo during the third nuclear division with incompletely remodeled male nucleus. Karyogamy has failed and the embryo contains four haploid nuclei of presumably maternal origin. The paternal nucleus (inset) still contains a region packaged with ProtA::GFP (arrow). Note that DNA positive dots at the spindle poles are Wolbachia endosymbionts. Scale bar: 10 μm.