Whole-embryo spatial transcriptomics at subcellular resolution from gastrulation to organogenesis.

Gene expression patterns underlie development, but their systematic detection in whole embryos has remained elusive. We introduce a whole-embryo imaging platform using multiplexed error-robust fluorescent in situ hybridization (weMERFISH). We quantified the expression of 495 genes in zebrafish embryos at subcellular resolution and generated an online atlas detailing the expression of 25,872 genes and accessibility of 294,954 chromatin regions during embryogenesis. Expression patterns often corresponded to composites of tissue-specific accessible elements, and expression changes aligned with cellular maturation and morphogenesis. Integration with live imaging revealed how similar expression patterns can emerge through different dynamics and showed that sharp boundaries develop through changes in gene expression rather than through cell sorting. These results establish multiplexed whole-embryo spatial transcriptomics and reveal the regulation and dynamics of embryonic gene expression patterns. Editor's summary: A central challenge in developmental biology is to visualize how gene expression patterns are organized across embryos and how they change during morphogenesis. Previous approaches have often been limited to small sets of genes or lacked sufficient spatial resolution. Wan et al. profiled whole zebrafish embryos at subcellular resolution, and placed the transcripts of hundreds of genes into precise spatial context. A comprehensive atlas of embryogenesis was created by linking this information with genome-wide single-cell expression and chromatin accessibility data. Together with live imaging, this resource elucidates how gene regulation and cell movements are coordinated to establish tissues and boundaries during development. —Stella M. Hurtley INTRODUCTION: Visualizing gene expression patterns has been essential to reveal the principles that underlie vertebrate development. Most analyses have been based on relatively small numbers of genes or rest on single-cell genomics approaches with limited spatial information. It has therefore been difficult to systematically study the full diversity and dynamics of vertebrate gene expression patterns and dissect their link to genomic elements and the movement of cells. RATIONALE: Earlier studies have demonstrated the potential of spatial transcriptomics technologies to map the expression of multiple genes in the same tissue. The goals of this study were to extend these technologies to systematically detect mRNA molecules with subcellular resolution across entire embryos and to create genome-wide atlases for exploration of gene expression, regulatory regions, and morphogenetic movements. RESULTS: We developed a whole-embryo imaging platform using multiplexed error-robust fluorescent in situ hybridization (weMERFISH). This spatial transcriptomics technology detected the transcripts of 495 genes at subcellular resolution in whole zebrafish embryos from gastrulation to early organogenesis. Through computational imputation with single-cell multiomics data, we reconstructed the spatial expression of more than 25,000 genes and the spatial accessibility of almost 300,000 chromatin regions, resulting in the online atlas MERFISHEYES (https://schier.merfisheyes.com). We found that subcellular RNA localization was gene-specific and dynamic and that combinatorial gene expression provided high positional information. Genes were often expressed in multiple related cell types and locations. Analysis of accessible chromatin regions supported the concept that the diversity of expression patterns is generated by a combination of dedicated control elements but also revealed chromatin regions that were accessible in several related domains. Moreover, cellular maturation and morphogenetic movements paralleled gene expression dynamics. Integration with in vivo cell tracking data through MERFISH-FATE revealed how similar expression patterns could emerge through diverse regulatory dynamics and that sharp boundaries developed through changes in gene expression rather than cell sorting. CONCLUSION: weMERFISH enables high-resolution spatial transcriptomics of hundreds of genes in whole embryos. Genome-wide atlases help explore the spatial patterns of gene expression and regulatory regions and their link to morphogenetic movements. These tools and observations lay the foundation for the integration of multiple modalities to reconstruct and analyze embryogenesis at global scales. Overview of the weMERFISH approach and key findings.: weMERFISH enables spatial transcriptomics of whole zebrafish embryos at subcellular resolution, measuring 495 genes and imputing 25,872 additional expression patterns and 294,954 chromatin accessibility profiles through integration with single-cell multiome data. The resulting atlas of gene expression and chromatin accessibility across developmental stages is accessible at https://schier.merfisheyes.com. Genes are frequently expressed in multiple tissues, supported by distinct tissue-specific accessible cis-regulatory elements. RNA velocity shows that transcriptional dynamics parallel cellular maturation and morphogenetic movements. Integration with live-imaging data reveals gene expression dynamics and demonstrates that sharp tissue borders form through local transcriptional changes rather than cell sorting. [ABSTRACT FROM AUTHOR]