Mapping lineage relationships between cells in a multicellular organism is a fundamental question in developmental biology (Kretzschmar and Watt, 2012). Completed in the 1980s, the description of the entire cell lineage of the nematode Caenorhabditis elegans by direct observation established a paradigm for cell lineage tracing(Sulston and Horvitz, 1977; Sulston et al., 1983) However, direct observation may not be suitable for complex organisms without invariant lineages and containing millions to billions of cells(VanHorn and Morris, 2021). Advancements in cell marking technologies encompass various approaches such as viral barcoding, Cre-mediated recombination, integration of random barcodes, and the utilization of natural somatic mutations (Salipante and Horwitz, 2006; Barker et al., 2007; Lu et al., 2011; Pei et al., 2017; Rodriguez-Fraticelli et al., 2018; Wagner et al., 2018). In recent years, progression in high-throughput sequencing and genome editing technology has made it feasible to trace cell lineages in complex multicellular organisms comprehensively. Among these methods, the application of Cas9 genome editing on target transgenic cassettes has been widely adopted (Alemany et al., 2018; Raj et al., 2018; Spanjaard et al., 2018; Bowling et al., 2020; Li et al., 2023). Although Cas9 creates diverse editing on barcodes, these methods rely on indels as genetic scars, where large deletions could remove previously recorded lineage information (VanHorn and Morris, 2021). Indeed, in zebrafish, the recorded mutations per barcode are approximately three, significantly fewer than the tens of mitotic divisions occurring from a zygote to a fully developed complex organism indicating that these systems miss recording lots of mitotic divisions.

Recently, SMALT, a substitution mutation-aided lineage-tracing system, was used in mapping cell phylogeny for Drosophila (Liu et al., 2021). SMALT utilized AID10–iSceI to induce C-to-T transition mutations on a 3-kb synthetic cassette, successfully recording on average over 20 mutations in early-adult fly cells. Here, we applied this system to zebrafish, utilizing a more condensed 1-kb barcoding sequence. In the 1-day post-fertilization (dpf) embryo, we documented a median of 14 substitution mutations on this 1-kb barcode. This mutation count is comparable to the estimated number of mitotic cell divisions from zygote to 1dpf embryo, providing a detailed recording of cell lineage history. Thus, our barcoding system outperforms the state-of-the-art Cas9-based lineage tracing method in zebrafish by recording a much higher number of mutations.

To demonstrate the capabilities of our system in zebrafish lineage reconstruction, we applied it to fin regeneration. Despite decades of detailed research by scientists on cell types and molecular pathways involved in zebrafish fin development and regeneration (Poss et al., 2000; Whitehead et al., 2005; Petrie et al., 2014; Pfefferli and Jaźwińska, 2015; Sehring and Weidinger, 2020), there is still a lack of quantitative descriptions of original and regenerated fin cell relationships based on cell phylogenetic trees. Besides, previous studies on fin generation mainly focus on the caudal fin (Lee et al., 2005; Blum and Begemann, 2015). By utilizing the mutated 1-kb barcode as alleles, we reconstructed four nearly single-cell-resolution lineage trees using zebrafish fin cells from 8 parts of fins in four individuals. Each tree comprised hundreds of internal nodes with 99% median bootstrap values supported. Using these trees, we inferred that regenerated fin cells mainly originate from the cells of the same part of the fins.

Furthermore, with this system, we sampled germ cells from separate spawning events of the same individual. Based on the cell lineage trees, we observed the mixing of germ cells from separate spawning and an early separation between the progenitor of germ cells and somatic cells. Besides, by analyzing the monophyletic clades formed by germ cells and performing a targeting coalescent analysis (TarCA) (Deng et al., 2024), we revealed a relatively small germ cell progenitor number.