Histone Methylation Participates in Gene Expression Control during the Early Development of the Pacific Oyster Crassostrea gigas

Abstract

Histone methylation patterns are important epigenetic regulators of mammalian development, notably through stem cell identity maintenance by chromatin remodeling and transcriptional control of pluripotency genes. But, the implications of histone marks are poorly understood in distant groups outside vertebrates and ecdysozoan models. However, the development of the Pacific oyster Crassostrea gigas is under the strong epigenetic influence of DNA methylation, and Jumonji histone-demethylase orthologues are highly expressed during C. gigas early life. This suggests a physiological relevance of histone methylation regulation in oyster development, raising the question of functional conservation of this epigenetic pathway in lophotrochozoan. Quantification of histone methylation using fluorescent ELISAs during oyster early life indicated significant variations in monomethyl histone H3 lysine 4 (H3K4me), an overall decrease in H3K9 mono- and tri-methylations, and in H3K36 methylations, respectively, whereas no significant modification could be detected in H3K27 methylation. Early in vivo treatment with the JmjC-specific inhibitor Methylstat induced hypermethylation of all the examined histone H3 lysines and developmental alterations as revealed by scanning electronic microscopy. Using microarrays, we identified 376 genes that were differentially expressed under methylstat treatment, which expression patterns could discriminate between samples as indicated by principal component analysis. Furthermore, Gene Ontology revealed that these genes were related to processes potentially important for embryonic stages such as binding, cell differentiation and development. These results suggest an important physiological significance of histone methylation in the oyster embryonic and larval life, providing, to our knowledge, the first insights into epigenetic regulation by histone methylation in lophotrochozoan development.

Keywords: H3K27; H3K36; H3K4; H3K9; embryos; epigenetics; histone modifications; methylstat; mollusk.

Figure 1

Histone methylation profiles of H3K4 (A), H3K9 (B), H3K27 (C) and H3K36 (D) during the early development of Crassostrea gigas. Asterisk in the legend indicates significant variation of the indicated mark along development A One-way ANOVA was used to determine whether methylation levels were statistically significantly different between development stage for each residue, and a p-value < 0.05 was considered significant.

Figure 2

(1) Hypermethylation of histone lysine residues in presence of 10 µM Methylstat. Global H3K4 methylation increase in presence of methylstat, but only H3K4me1 and H3K4me3 are significant. Global H3K9 methylation increase in presence of methylstat, but only H3K9me2 is significant. H3K27me and H3K27me2 increase in presence of methylstat. Asterisk in the legend indicates significant variation of the indicated mark (One-way ANOVA; p < 0.05 was considered significant (<0.05 (*), <0.001 (**), <0.0001 (***)). (2) Abnormal development. Different phenotypes observed in control condition and under methylstat treatment at 6 h after fertilization (a) and 24 h after fertilization (b).

Figure 3

(A) Gene clustering. KMC clustering of the 376 genes differentially expressed upon methylstat exposure independently of development time. Three main clusters were discriminated (From the top to the bottom): (1) overexpressed in MeS-treated animals (147 genes), (2) underexpressed in MeS-treated animals (90 genes) and (3) underexpressed in MeS-treated 6hpf animals (139 genes). (B) Principal component analysis. The first three Principal component (PCs) explained 55.417% of the total variance of the 376 genes significantly differentially expressed between sample pools.

Figure 4

RT-qPCR measurements of mRNA expression of selected marker genes from the different clusters. Examples of cluster 1 (Peptidase inhibitor 15-a (pi15a) (CU683400), overexpressed in MeS treatment) and cluster 2 (Sperm7 (FP004994), underexpressed in MeS-treated 6hpf animals) are shown. For each transcript, a student’s T tests (p < 0.05 was considered significant) has been used to compare control 6 h (Ctl6h) vs. Methylstat sample 6h (MeS6h) and control 24 h (Ctl24h) vs. Methylstat sample 24 h (MeS24h). (****: p < 0.0001).