Clonal dynamics in early human embryogenesis inferred from somatic mutation

Abstract

SummaryThe trillions of cells that constitute the human body are developed from a fertilized egg through embryogenesis. However, cellular dynamics and developmental outcomes of embryonic cells in humans remain to be largely unknown due to the technical and ethical challenges. Here, we explored whole-genomes of 334 single-cell expanded clones and targeted deep-sequences of 379 bulk tissues obtained from various anatomical locations from seven individuals. Using the discovered 1,688,652 somatic mutations as an intrinsic barcode, we reconstructed cellular phylogenetic trees that provide novel insights into early human embryogenesis. Our findings suggest (1) endogenous mutational rate that is higher in the first cell division of life but decreases to ~1 per cell per cell division later in life, (2) universal unequal contribution of early cells into embryo proper resulting from early cellular bottlenecks that stochastically separate epiblasts from embryonic cells (3) uneven differential outcomes of early cells into three germ layers, left-right and cranio-caudal tissues, (4) emergence of a few ancestral cells that will contribute to the substantial fraction of adult blood cells, and (5) presence of mitochondrial DNA heteroplasmy in the fertilized egg. Our approach additionally provides insights into the age-related mutational processes including UV-mediated mutagenesis and loss of chromosome X or Y in normal somatic cells. Taken together, this study scrutinized somatic mosaicism, clonal architecture, and cellular dynamics in human embryogenesis at an unprecedented level and provides a foundation for future studies to complete cellular phylogenies in human embryogenesis.

Multi-organ analysis of low-level somatic mosaicism reveals stage- and tissue-specific mutational features in human development

“…The increased relative portion of SBS1 in late-stage somatic mutations appears to represent active proliferation and clonal expansion during late-embryonic and post-natal or aging periods 22,23 . Although the etiologies associated with most indel signatures remain unknown, the higher contributions of ID1 and ID2 in early-stage SNVs and ID5 and ID8 in late-stage SNVs were consistent with a previous finding 24 .…”

Effects of Different Mutation Rates and Randomness on the Distribution of Linkage Disequilibrium

“…Mutation is considered as the only source to provide genetic variation, and has been extensively studied in fields like evolution, agriculture and human diseases, partially due to the relatively easy and large-scale identification and functional exploration by genome sequencing [15,[34][35][36][37][38]. As we found here, different mutation types and rates, and their combination with factors like effective population size, recombination and selection, do interfere with the dynamics of LD distribution.…”

“…It is not uncommon that mutations are indicated to be involved in the complex trait of importance, e.g. agriculture or human disease studies [15,[34][35][36][37][38]. Many questions remained to be answered.…”

“…[2]-31]. Novel insights on mutation and its role in population genetics [32,33] and disease incidence (especially driver mutations in cancer progression) were obtained, in large-scale cancer genomics and in-depth data mining studies [34][35][36][37][38].…”

Detection of Low-Level Parental Somatic Mosaicism for Clinically Relevant Snvs and Indels Identified in a Large Exome Sequencing Dataset

“…Of note, the levels of mosaicism may vary greatly in whole blood as a result of somatic clonal expansion, especially in subjects with advanced age [26]. Furthermore, recent work has shown that blood cell lineages branch off from other cell lineages early in embryogenesis, and that certain bottlenecks in embryonic development lead to heterogeneity in genetic mosaicism in different regions of the brain [27][28]. Thus, it has been suggested that somatic mosaicism may not be measured in the most accurate fashion through sampling whole blood samples.…”