Nascent transcriptome reveals orchestration of zygotic genome activation in early embryogenesis

Curr Biol. 2022 Aug 22;S0960-9822(22)01232-5. doi: 10.1016/j.cub.2022.07.078.

Hui Chen & Matthew C Good

Click here to view article at Current Biology.

Click here to view article on PubMed.

Click here to view article on Xenbase.

In Brief

By profiling the nascent transcriptome during zygotic genome activation (ZGA) from whole Xenopus blastula and dissected subregions, Chen and Good unveil predominant transcription from maternal-zygotic genes and distinct spatial patterns, reconcile regulatory mechanisms of ZGA, and discover a link to sequential activation of germ-layer-specific genes.

Highlights

- Whole-embryo and regional EU-RNA-seq determines timing and spatial patterns of ZGA

- Maternal-zygotic genes dominate transcriptional output during ZGA

- Manipulation of translation and cell division reconciles regulatory mechanisms of ZGA

- Timing of germ-layer-specific expression appears sequential in the blastula

Abstract

Early embryo development requires maternal-to-zygotic transition, during which transcriptionally silent nuclei begin widespread gene expression during zygotic genome activation (ZGA).1-3 ZGA is vital for early cell fating and germ-layer specification,3,4 and ZGA timing is regulated by multiple mechanisms.1-5 However, controversies remain about whether these mechanisms are interrelated and vary among species6-10 and whether the timing of germ-layer-specific gene activation is temporally ordered.11,12 In some embryonic models, widespread ZGA onset is spatiotemporally graded,13,14 yet it is unclear whether the transcriptome follows this pattern. A major challenge in addressing these questions is to accurately measure the timing of each gene activation. Here, we metabolically label and identify the nascent transcriptome using 5-ethynyl uridine (5-EU) in Xenopus blastula embryos. We find that EU-RNA-seq outperforms total RNA-seq in detecting the ZGA transcriptome, which is dominated by transcription from maternal-zygotic genes, enabling improved ZGA timing determination. We uncover discrete spatiotemporal patterns for individual gene activation, a majority following a spatial pattern of ZGA that is correlated with a cell size gradient.14 We further reveal that transcription necessitates a period of developmental progression and that ZGA can be precociously induced by cycloheximide, potentially through elongation of interphase. Finally, most ectodermal genes are activated earlier than endodermal genes, suggesting a temporal orchestration of germ-layer-specific genes, potentially linked to the spatially graded pattern of ZGA. Together, our study provides fundamental new insights into the composition and dynamics of the ZGA transcriptome, mechanisms regulating ZGA timing, and its role in the onset of early cell fating.

Figure 1. Nascent EU-RNA-seq to characterize the composition and dynamics of ZGA with high sensitivity
(A) Schematic of transcript composition during zygotic genome activation in early embryogenesis: from egg to late blastula. Red, transcripts of zygotic genes; orange, transcripts of maternal-zygotic genes; blue, transcripts of maternal genes.
(B) Schematic describing the EU-RNA-seq methodology. Nascent transcripts are metabolically labeled via 5-ethynyl uridine (5-EU) microinjected in 1-cell Xen- opus embryos. Total RNAs are isolated for biotinylation via click reaction. The 5-EU-labeled nascent transcripts (red) are captured by streptavidin beads; flow- through contains maternal transcripts.
(C) Distinguishing nascent transcriptome (‘‘bead’’) versus maternal transcriptome (‘‘flowthrough’’) reads via RNA-seq from 5 to 9 hpf in blastula embryos. Each dot represents individual genes with rlog reads averaged from replicates quantified by DESeq2. Dashed lines: 1.5-fold threshold for enrichment.
(D) Nascent reads enrich in ‘‘bead’’ library versus traditional total transcriptome (‘‘all’’) at 7 hpf from RNA-seq. Each dot represents individual genes with log2 reads averaged from replicates quantified by DESeq2. Dashed lines: 1.5-fold threshold for enrichment.
(E) Higher sensitivity for detection of zygotic expression in nascent transcriptome versus total transcriptome at all gene expression levels.
(F) Hundreds of transcripts are uniquely detected by nascent transcriptome. Mean reads from duplicates (mean ± SE) for 240 genes detected from nascent EU- RNA (red) and total RNA (orange), respectively.
(G) Percentage of genes expressed during ZGA that are classified as zygotic-only genes (Z) and maternal-zygotic gene (MZ).
(H and I) Percentage of total library reads from transcripts of zygotic-only genes (Z) and maternal-zygotic gene (MZ) from 5 to 9 hpf.
(J and K) Gene ontology (GO) analysis of MZ (J) and Z (K) genes.

Figure 2. EU-RNA-seq on segmented embryos uncovers spatial patterns of single-gene activation
(A) Schematic showing spatial patterns of ZGA observed from EU-RNA imaging, and strategy for classifying gene expression profiles from segmented embryos using regional EU-RNA-seq. AP, animal pole; VP, vegetal pole.
(B) Categorization of distinct spatial patterns of activation for zygotic genes. Left: schematic of each pattern for gene activation. Middle: reads from EU-RNA-seq performed on the segmented AP (red) and VP (blue) regions from early embryo at 5–9 hpf. Transcript levels were thresholded to those at 5 hpf. Mean ± SE from indicated number of genes. Right: comparison of the activation onset time for transcription of each gene for AP (red) and VP (blue). ****p < 0.0001; ns, not significant. Note that to compare and plot the activation onset time for those genes that were not activated during 5–9 hpf at AP or VP, their activation onset time was set to 9.1. (C) Percentage of zygotic genes that have distinct spatial patterns of expression during ZGA. A majority of genes (55%) show AP early and VP delayed expression.

Figure 4. Timing of germ-layer initiation is correlated to spatially graded onset of ZGA
(A) Average transcript levels for ectoderm and endoderm genes in blastula embryos. Nascent reads averaged from replicate embryos. Red, set of ectoderm genes (N = 111 genes); blue, set of endoderm genes (N = 172 genes) at 5–9 hpf. Mean ± SE.
(B) Distribution of time of onset for transcriptional activation of individual genes within ectoderm (red) and endoderm (blue) sets.
(C) Cumulative density of time of onset for transcriptional activation for ectoderm (red) and endoderm (blue) genes.
(D and E) Heatmaps showing Z scores for induction of ectoderm (D) and endoderm (E) genes in control embryos at 7.5 hpf versus embryos treated with CHX from 5 to 7.5 hpf. Data are selected genes that have detectable transcriptional induction in control embryos from 5 to 7.5 hpf. Pie chart shows fraction of genes similar (gray) or upregulated (red) or downregulated (blue), comparing CHX-treated versus control embryos; threshold 1.5-fold difference.
(F) Cumulative density showing induction relative to control of germ-layer genes in CHX-treated embryos at 7.5 hpf. Red, ectoderm; blue, endoderm.
(G) Profile plots for ChIP-seq peaks of RNA Pol II for ectoderm and endoderm genes, respectively, in embryos at stage 10.5. Data from Session et al.19 The regions of transcription start site (TSS) ±2kb from two replicates are shown. The level of RNA Pol II binding in the indicated regions is higher in ectoderm genes than in endoderm genes.
(H) Composite model and pathway describing regulated ZGA onset. Widespread ZGA is inhibited by rapid cell cycles in cleavage-stage embryos. As cell size reduces, due to cell division without cell growth, a DNA:cytoplasm ratio threshold is reached, promoting cell-cycle elongation and allowing for accumulation of nascent zygotic transcripts. Cell-cycle elongation is sufficient to promote large-scale ZGA in embryos that have achieved transcriptional competence. Translation of maternal TFs and histone acetyltransferases are necessary to generate transcriptionally competent embryos. Xenopus blastula embryos contain a gradient of cell sizes which achieve widespread ZGA onset at different times. A majority of the nascent transcriptome during ZGA displays early activation in smaller cells at the animal pole (AP)— the presumptive ectoderm—and later activation in larger cells at the vegetal pole (VP)—the presumptive endoderm. Chronological sequence of germ-layer induction may be linked to temporally graded onset of widespread ZGA in the blastula. Red, ectoderm genes; blue, endoderm genes. For clarity, the schematic of embryo is shown rotated 90 counterclockwise along the AV axis from the normal orientation of the embryo. Dotted arrow indicates that N/C ratio may affect ZGA transcription in a cell-cycle-independent manner.

Adapted with permission from Current Biology on behalf of Cell Press: Chen & Good. (2022). Nascent transcriptome reveals orchestration of zygotic genome activation in early embryogenesis. Curr Biol. 2022 Aug 22;S0960-9822(22)01232-5. doi: 10.1016/j.cub.2022.07.078.