One thousand somatic SNVs per skin fibroblast cell set baseline of mosaic mutational load with patterns that suggest proliferative origin

Abstract

Few studies have been conducted to understand post-zygotic accumulation of mutations in cells of the healthy human body. We reprogrammed 32 skin fibroblast cells from families of donors into human induced pluripotent stem cell (hiPSC) lines. The clonal nature of hiPSC lines allows a high-resolution analysis of the genomes of the founder fibroblast cells without being confounded by the artifacts of single-cell whole-genome amplification. We estimate that on average a fibroblast cell in children has 1035 mostly benign mosaic SNVs. On average, 235 SNVs could be directly confirmed in the original fibroblast population by ultradeep sequencing, down to an allele frequency (AF) of 0.1%. More sensitive droplet digital PCR experiments confirmed more SNVs as mosaic with AF as low as 0.01%, suggesting that 1035 mosaic SNVs per fibroblast cell is the true average. Similar analyses in adults revealed no significant increase in the number of SNVs per cell, suggesting that a major fraction of mosaic SNVs in fibroblasts arises during development. Mosaic SNVs were distributed uniformly across the genome and were enriched in a mutational signature previously observed in cancers and in de novo variants and which, we hypothesize, is a hallmark of normal cell proliferation. Finally, AF distribution of mosaic SNVs had distinct narrow peaks, which could be a characteristic of clonal cell selection, clonal expansion, or both. These findings reveal a large degree of somatic mosaicism in healthy human tissues, link de novo and cancer mutations to somatic mosaicism, and couple somatic mosaicism with cell proliferation.

Figure 1.

Conceptual diagram of our approach. Our cohort consisted of four families, each having a proband with autism, while the other family members were phenotypically normal. Family 03 includes a normal male sibling. Three hiPSC lines were generated from the fibroblast samples of each person in the cohort. As hiPSC lines are clonally derived from single cells, comparison (STAGE 1) of their genomes to the germline genome uncovers mosaic variants present in the founder fibroblast cells of hiPSC colonies (green, orange, and purple variants). Germline variants for children were inferred from corresponding parents and those for parents from corresponding children and fibroblast samples. Analysis at STAGE 1 yields a list of putative mosaic variants manifested in hiPSC lines. In STAGE 2, the mosaic candidates are scrutinized by additional experiments in founder fibroblasts to confirm their presence and to determine their tissue allele frequency (TAF). Naming pattern for hiPSC lines is as follows: family-person#hiPSC, e.g., S1120-01#2.

Figure 2.

Discovery and confirmation of LM-SNVs in children. (A) Discovered LM-SNVs in hiPSC lines were divided into two groups: low tissue allele frequency (with no evidence in fibroblasts, orange bars), and high TAF (with some evidence, yellow bars). Site reanalysis in fibroblasts with DNA capture and deep sequencing confirmed that, on average, 74 LM-SNVs in each hiPSC line are mosaic SNVs in fibroblasts (green bars) or 235 when adjusted for discovery sensitivity and ascertained fraction (light blue bars). (B) Virtually all LM-SNVs were present at around 50% AF in hiPSC lines as detected by amplicon-seq experiments. (C) Capture-seq experiment in fibroblasts revealed that mosaic SNVs were present in the fibroblast tissue with TAF ranging from 0.25% to 35%. Distributions of TAF have clear peaks. (D) Amplicon-seq experiment for 57 LM-SNVs sites results in better sensitivity and confirms an additional six LM-SNVs with low TAF as mosaic (black dots), including two with no supporting read in the data from capture (shown with TAF of 10⁻⁴ for capture). Germline and confirmed mosaic SNVs by capture experiments are in red and green circles, respectively. (E) ddPCR reactions revealed excellent concordance in TAF estimates with the capture-seq and amplicon-seq experiments. Dashed green bars show SNV sites for which capture experiments were conducted, but support for the alternative allele was consistent with background sequencing noise. Additional ddPCR assays confirmed mosaic SNV at even lower TAFs that could not be accessed by the other two experiments.

Figure 3.

(A) On a large scale, the distance between neighboring SNVs is distributed according to the power law, i.e., frequency decreases exponentially with increasing distance. This is consistent with a uniform distribution of SNVs across the genome (such simulated distributions are shown by dashed lines). There is an enrichment of short (i.e., <20 bp) distances (see inset). All data pertain to four children. (B) Distribution of trinucleotide motifs of the reference genome around SNVs defines mutational signature. The signature of mosaic SNVs is similar to signature 5 and signature 8 observed in cancers (Alexandrov et al. 2013). (C) Comparison of mutational signature in this study (solid bars) with signature 5 from cancers (empty bars).

Figure 4.

(A) Mosaic SNVs in fibroblasts (detected in this study) exhibit negative correlations (except for the H4K20me1 mark) with histone marks from skin fibroblasts (blue bars). This is similar to what was observed for somatic SNVs in liver cancer (external data set) with histone marks from hepatocytes (green) and hepatocellular carcinoma cell line (cyan). However, the absolute values of correlations are lower. Data for certain histone marks are not available, and the corresponding bars are not shown. In contrast, mosaic SNVs in fibroblasts correlate positively with histone marks in stem cells. Correlations for the H1 cell line are shown in red (see also Supplemental Fig. S9). (B) Example of correlation between density of mosaic SNVs in fibroblasts and two histone marks in fibroblasts.