Systematic reconstruction of the cellular trajectories of mammalian embryogenesis (E8.5)

Mammalian embryogenesis is characterized by rapid cellular proliferation and diversification. Within a few weeks, a single cell zygote gives rise to millions of cells expressing a panoply of molecular programs, including much of the diversity that will subsequently be present in adult tissues. Although intensively studied, a comprehensive delineation of the major cellular trajectories that comprise mammalian development in vivo remains elusive. For mouse embryogenesis in particular, we and others have performed single cell or single nucleus RNA-seq data (scRNA-seq) during implantation, gastrulation and organogenesis. Here we set out to integrate several single cell RNA-seq datasets (scRNA-seq) that collectively span mouse gastrulation and organogenesis. However, a technical challenge that we faced is that the datasets that we sought to integrate were generated by different groups at different times using different scRNA-seq technologies. In particular, probably because there was no overlapping timepoint, the integration of scRNA-seq data generated at E8.5 (cells, 10X Genomics) and E9.5 (nuclei, sci-RNA-seq3) was challenging (Cao et al. 2019; Pijuan-Sala et al. 2019). To address this, we set out to generate new data at E8.5 that might serve to “bridge” these two datasets. Because of how quickly changes are occurring during this window of development, we focused on individual, somite-resolved E8.5 embryos using a simplified, optimized version of sci-RNA-seq3. We selected 12 embryos from 2 separate litters harvested at E8.5, including a single primitive streak stage embryo (prior to somitogenesis) and 11 embryos staged in 1-somite increments from 2 to 12 somites. The optimized sci-RNA-seq3 method markedly improved data quality, with 9-fold higher UMIs and 6-fold higher gene detection per nucleus, relative to (Cao et al. 2019). Overall, we collected published data (Cheng et al. 2019; Mohammed et al. 2017; Pijuan-Sala et al. 2019), the new E8.5 data, and published data from one study spanning E9.5 to E13.5 but with deeper sequencing of those libraries (Cao et al. 2019). Altogether, we define cell states at each of 19 successive stages spanning E3.5 to E13.5, heuristically connect them to their pseudo-ancestors and pseudo-descendants. Despite being constructed through automated procedures, the resulting trajectories of mammalian embryogenesis (TOME) are largely consistent with our contemporary understanding of mammalian development. In addition, the new E8.5 data itself comprises a foundational resource for mammalian developmental biology (especially for the early somitogenesis), and are made available in a way that will facilitate their ongoing annotation by the research community. Overall design: Around 240 thousand cells derived from 12 individual, somite-resolved E8.5 mouse embryos. The 12 samples are from 2 separate litters harvested at E8.5, including a single primitive streak stage embryo (prior to somitogenesis) and 11 embryos staged in 1-somite increments from 2 to 12 somites.

哺乳动物胚胎发生以快速的细胞增殖与分化为特征。短短数周内，单个细胞的受精卵便可发育为数百万个细胞，这些细胞表达着多样的分子程序，涵盖了后续成体组织中几乎全部的细胞多样性。尽管该领域已得到深入研究，但全面阐明体内哺乳动物发育所涉及的主要细胞轨迹，仍是尚未解决的难题。尤其针对小鼠胚胎发生，我们与其他团队已在着床期、原肠胚形成期及器官发生期开展了单细胞或单细胞核RNA测序（single cell RNA-seq，scRNA-seq）相关工作。本研究旨在整合多组覆盖小鼠原肠胚形成与器官发生阶段的单细胞RNA测序数据集。然而，我们面临的一项技术挑战在于：待整合的数据集由不同团队在不同时期、采用不同的scRNA-seq技术生成。特别是，由于缺乏重叠的时间节点，E8.5阶段（细胞样本，10X Genomics平台）与E9.5阶段（细胞核样本，sci-RNA-seq3技术）产生的scRNA-seq数据整合难度极大（Cao等，2019；Pijuan-Sala等，2019）。为解决这一问题，我们计划在E8.5阶段生成新的测序数据，以作为衔接上述两组数据集的“桥梁”。鉴于该发育窗口内细胞变化极为迅速，我们采用经简化优化的sci-RNA-seq3技术，针对单个、经体节解析的E8.5胚胎开展研究。我们从两窝分别收获的E8.5胚胎中选取了12个样本，其中包括1个原条期胚胎（体节发生前），以及11个以1个体节为增量分期、从2体节到12体节的胚胎。经优化的sci-RNA-seq3技术显著提升了数据质量：与Cao等（2019）的研究相比，每个细胞核的唯一分子标识符（Unique Molecular Identifiers，UMIs）数量提升了9倍，基因检出率提升了6倍。总体而言，我们收集了已发表的数据集（Cheng等，2019；Mohammed等，2017；Pijuan-Sala等，2019）、本研究新增的E8.5阶段数据，以及另一项覆盖E9.5至E13.5阶段且进行了更深层测序的已发表数据集（Cao等，2019）。综上，我们明确了从E3.5到E13.5共19个连续阶段的细胞状态，并通过启发式方法将其与伪祖先及伪后代关联。尽管该哺乳动物胚胎发生轨迹（TOME）是通过自动化流程构建的，但其整体与当前我们对哺乳动物发育的认知高度一致。此外，本研究新增的E8.5阶段数据本身即可作为哺乳动物发育生物学（尤其是早期体节发生研究）的基础资源，并以便于科研群体持续注释的形式公开。整体实验设计：本研究获取了来自12个经体节解析的E8.5小鼠胚胎的约24万个细胞。这12个样本来自两窝分别收获的E8.5胚胎，其中包括1个原条期胚胎（体节发生前），以及11个以1个体节为增量分期、从2体节到12体节的胚胎。