Single-cell genomics to study developmental cell fate decisions in zebrafish

Briefings in Functional Genomics, 20(6), 2021, 420–426 https://doi.org/10.1093/bfgp/elab018 Advance Access Publication Date: 30 March 2021 Review Paper Single-cell genomics to study developmental cell fate decisions in zebrafish Corresponding author: J.P. Junker, Max Delbrück Center for Molecular Medicine, Berlin Institute for Medical Systems Biology, 10115 Berlin, Germany. Tel: +49 30 9406 1860; Fax: +49 30 9406 1779; E-mail: Abstract New developments in single-cell genomics have transformed developmental biology in recent years by enabling systematic analysis of embryonic cell types and differentiation trajectories. Ongoing efforts in experimental and computational method development aim to reveal gene-regulatory mechanisms and to provide additional spatio-temporal information about developmental cell fate decisions. Here, we discuss recent technological developments as well as biological applications of single-cell genomics, with a particular focus on analysis of developmental cell fate decisions. Although the approaches described here are generally applicable to a broad range of model systems, we focus our discussion on applications in zebrafish, which has proven to be a particularly powerful model organism for establishing novel methods in single-cell genomics. Key words: zebrafish; cell fate decisions; development; single-cell genomics; spatial transcriptomics; lineage tracing One of the main outcomes of embryonic development is the acquisition of cell identity and function. Identifying and categorizing the many cell types present in an organism has been a slow and laborious process in the past. Single-cell genomics technologies constitute an important advancement for the characterization of the cellular heterogeneity in a sample by allowing identification of transcriptomic and chromatin accessibility profiles in thousands of single cells. Importantly, these approaches not only enable systematic identification of embryonic cell types, but they also yield insight into developmental differentiation trajectories, lineage trees and regulatory mechanisms [1–3]. Single-cell transcriptomics is by far the most advanced of the single-cell omics technologies and will hence take up the largest part of this review. The zebrafish has been one of the protagonist models in the emergence of the single-cell technologies. In this review, we summarize the applications used to this day in this model organism. Technologies for single-cell transcriptomics Single-cell genomics experiments typically start with dissociation of the tissue of interest into a single-cell solution, and the quality of the single-cell suspension is a decisive factor for the success of the downstream experiment. Incomplete dissociation, loss of specific cell types and triggering of cellular stress response are typical challenges in single-cell genomics experiments. The use of a psychrophilic protease during this critical step has been reported to alleviate possible artifacts [4]. Once dissociation has been optimized, individual cells need to be processed into sequencing libraries. The two most widely used experimental approaches are plate-based processing using liquid handling robotics and droplet microfluidics. Most early studies in single-cell transcriptomics were plate based, i.e. cells are sorted into and lysed in individual wells of a microwell plate [5, 6]. Droplet-based methods, in which cell lysis and reverse transcription happen in nanoliter-sized droplets containing reagents and cellular barcodes, have gained prominence in recent years due to their higher throughput and lower cost per cell [7, 8]. However, both approaches have distinct advantages and disadvantages: although plate-based methods are limited to lower numbers of cells, they typically provide higher quality data and full transcript coverage, whereas current droplet microfluidics approaches capture only 3 or 5 tags of transcripts. Roberto Moreno-Ayala is a postdoctoral researcher in Jan Philipp Junker’s laboratory working on early zebrafish developmental variability and its phenotypic outcomes. Jan Philipp Junker is a group leader at the Max Delbrück Center in Berlin. Using the zebrafish as their primary model system, his group develops and uses methods in single-cell genomics in order to understand cell fate decisions in health and disease. © The Author(s) 2021. Published by Oxford University Press. All rights reserved. For Permissions, please email: 420 Roberto Moreno-Ayala and Jan Philipp Junker Single-cell genomics 421 into a single-cell suspension, and the transcriptomics profiles are used to build cell fate trajectories for some embryonic structures. (B) Workflow of a linage tracing experiment: a one-cell stage embryo is injected with Cas9/gRNA to introduce indels in specific loci of the genome during the first cell divisions (colored lines). These cellular barcodes will give information about the history of each cell to construct a lineage tree. Understanding cell fate decisions by single-cell genomics Single-cell RNA sequencing (scRNA-seq) has emerged as a powerful method for systematic identification of cell types [9]: single-cell profiles can be clustered by transcriptome similarity, and the identified clusters correspond to the different cell types in the sample. However, clustering results may differ depending on the algorithm and the metrics that are used, which leads to a certain level of ambiguity in cell type identification. Characterizing cell fate dynamics during embryogenesis is an ongoing endeavor. By sampling embryos at different stages, and by ordering single-cell transcriptomic profiles by similarity, a systematic landscape of developmental differentiation trajectories can be reconstructed (Figure 1A) [10, 11]. In this way, the transcriptional changes that cells undergo during differentiation can be measured in a systematic and continuous way, which may lead to identification of marker genes for previously uncharacterized intermediate states. By tracing back the earliest origin of an embryonic structure, the likely progenitors of this cell type and the branch points of cell fate decisions can be determined [10]. Identifying the gene regulatory mechanisms that underlie cell fate decisions is one of the major questions in developmental biology, and single-cell genomics data are a powerful basis for computational prediction of regulatory networks. However, inference of gene regulatory networks remains a challenging task, and methods based purely on transcriptomic data have only moderate performance [12]. Therefore, approaches that include transcription factor-binding information or open chromatin data [13, 14] are required for reliable identification of the gene regulatory networks that underlie developmental cell fate decisions. Beyond analysis of wild-type animals, single-cell genomics also provides the means to better understand mutant phenotypes by comparing their cell state composition to wild-t (...truncated)