Dynamics of Chromatin Opening across Larval Development in the Urochordate Ascidian Ciona savignyi

Abstract

Ascidian larvae undergo tail elongation and notochord lumenogenesis, making them an ideal model for investigating tissue morphogenesis in embryogenesis. The cellular and mechanical mechanisms of these processes have been studied; however, the underlying molecular regulatory mechanism remains to be elucidated. In this study, assays for transposase-accessible chromatin using sequencing (ATAC-seq) and RNA sequencing (RNA-seq) were applied to investigate potential regulators of the development of ascidian Ciona savignyi larvae. Our results revealed 351 and 138 differentially accessible region genes through comparisons of ATAC-seq data between stages 21 and 24 and between stages 24 and 25, respectively. A joint analysis of RNA-seq and ATAC-seq data revealed a correlation between chromatin accessibility and gene transcription. We further verified the tissue expression patterns of 12 different genes. Among them, Cs-matrix metalloproteinase 24 (MMP24) and Cs-krüppel-like factor 5 (KLF5) were highly expressed in notochord cells. Functional assay results demonstrated that both genes are necessary for notochord lumen formation and expansion. Finally, we performed motif enrichment analysis of the differentially accessible regions in different tailbud stages and summarized the potential roles of these motif-bearing transcription factors in larval development. Overall, our study found a correlation between gene expression and chromatin accessibility and provided a vital resource for understanding the mechanisms of the development of ascidian embryos.

Keywords: ATAC-seq; RNA-seq; ascidian; gene expression; open chromatin; transcription factor.

Figure 1

Morphological characterization of embryonic development in C. savignyi and the landscape of DNA accessibility in three developmental stages. (A) St. 21, St. 24, and St. 25 C. savignyi embryos cultured at 16 °C were stained with phalloidin. The upper images present the intact ascidian embryo, and the lower images present the notochord tissue. Scale bar, 10 μm. (B) Peak distribution ratio of gene functional elements. Six regions were detected: 3’-UTRs, 5’-UTRs, exon, intron, promoter, and intergenic regions. (C,D) Scatter plot indicates the pattern of differentially accessible regions (DARs) in different developmental stages, including both FDR < 0.05 (red) and FDR ≥ 0.05 (blue) genes. (E,F) GO enrichment analysis of DAR-associated genes. C and E: St. 21 vs. St. 24; D and F: St. 24 vs. St. 25.

Figure 2

Visualization of DEGs. (A,C) Visualization of gene expression changes. padj is represented by −log₁₀(padj) in the ordinate, and gene expression is represented by log₂ (fold change) in the abscissa. Blue, red, and gray dots indicate downregulated DEGs, upregulated DEGs, and genes with no significant expression change, respectively. (B,D) GO enrichment analysis of DEGs at different stages. B: St. 21 vs. St. 24, D: St. 24 vs. St. 25.

Figure 3

Integrated analysis of ATAC-seq and RNA-seq data. (A,B) Intersection of different mRNAs and target genes of different peaks. DAR-associated genes (green and purple indicate St. 21 vs. St. 24 and St. 24 vs. St. 25, respectively) and DEGs (blue and orange indicate St. 21 vs. St. 24 and St. 24 vs. St. 25, respectively). (C) Chromatin opening and expression changes of associated genes. IGV plots present ATAC-seq signals in selected gene loci. Gene expression was determined by quantitative fluorescence PCR and transcriptomic analysis. Blue, red, and green represent chromatin fragments of St. 21, St. 24, and St. 25, respectively. Black and gray bars represent transcriptome data and quantitative fluorescence PCR results of genes, respectively.

Figure 4

Expression and functional assay of key genes in larval development. (A) Tissue expression patterns of the selected genes as determined by promoter assay. Scale bar, 50 μm. (B) Treatment with an MMP24 inhibitor (GM6001, 40 μM) arrested lumen expansion. Scale bar, 10 μm. The asterisk marks the lumen. (C) Quantification of lumen volume in DMSO- (n = 191) and GM6001-treated embryos (n = 165) embryos at 22 hpf. Statistical analysis was performed using Student’s t-test, p < 0.001 (***). (D) Overexpression of dominant-negative Cs-KLF5 led to failure of lumen formation. Schematic images located on the right side of confocal images present the phenotypes of Cs-KLF5 knockdown and control. The green, blue, and red areas represent the notochord cells, nuclei, and nuclei in cells expressing Bra>Cs-KLF5DN::tdtomato/Bra> tdtomato (red), respectively. Scale bar, 10 μm.

Figure 5

TF binding motif enrichment analysis. (A) Nine TF binding motifs enriched in accessible regions from closed to open in St. 21 versus St. 24. (B) Eight TF binding motifs enriched in accessible regions from closed to open in St. 24 versus St. 25. Red highlights the upregulation of TFs. (A,B). The X-axis represents −log10 (p-value) in the motif enrichment analysis. The Y-axis represents the predicted TF binding motifs. (C) Summary of the potential roles of TFs in Ciona larval development. Different colored arrows represent downstream genes that may be regulated by different transcription factors.