One person, many genomes

Mutations occur in cells of the body

Our genome is under attack. This assault comes from exogenous mutagens like UV light and polycyclic aromatic hydocarbons present in tobacco smoke, and also from endogenous forces such as reactive oxygen species, mobile DNA elements, and errors during normal cell division. As a result, each of us is a genetic mosaic, with each cell likely having a unique genome, different from every other cell in the body.

 

Since mosaic mutations exist in only some cells in a tissue, or even in a single cell, they are too rare to be studied comprehensively using standard genome sequencing techniques. Therefore, our group has used single-cell whole-genome sequencing to characterize the rates, characteristics, and consequences of somatic mutations in the human brain. These experiments have allowed us to use somatic mutations to study the birth and death of human neurons.

Somatic mutations as natural lineage tags

One of the great mysteries of biology is how a single fertilized zygote can give rise to the vast array of specialized cell types comprising the human body. Viral and other lineage tracing methods have been workhorses of developmental biology in model organisms, but cannot be used to interrogate human development for technical and ethical reasons. Interestingly, we found that naturally-occurring developmental somatic mutations can be used as endogenous lineage tags to probe human development (Lodato et al., 2015, Science). This work allowed us to draw a single-cell phylogeny of neurons in the human brain for the first time. Surprisingly, we found that the human neurons are deeply polyclonal, with some lineages comprising a small fraction of neurons but including cells of the heart, liver, lungs, pancreas, and other organs. Tracing human cell lineage with somatic mutations has made previously untestable hypotheses about human development tractable for the first time.

website lineage fig 1.tif

Somatic mutations demonstrate lineage relationships. (A) Phylogeny of 136 human cortical neurons from a 17 y/o male derived from 18 clonal somatic mutations. Neurons are placed into four distinct nested clades (pink, green, blue, purple) defined by one or more independent mutations. Cells are ordered within clades based on the presence of multiple somatic mutations. A few cells in each clade fail to manifest individual SNVs shared by other cells of the same clade (indicated by open squares), likely representing incomplete amplification. Dark gray boxes represent cells analyzed by scWGS, and light gray represents cells analyzed by targeted sequencing. Genomic locations of mutations are not shown. (B) Ultra-deep sequencing (~10,000x) of four mutated loci (B2, D2, A1, and C1) in bulk DNA isolated from across the brain and body. A1, C1, and D2 are detected in the cortex and across the body, while B2 is limited to the cortex. BA - Brodmann area. Modified from Lodato et al., 2015.

Somatic mutations increase in aging in human neurons

            Increased DNA damage is a well-known hallmark of aging, but whether that damage results in increased permanent somatic mutations in the brain has been an open question. We applied single-cell DNA sequencing to this question for the first time, showing that indeed in two areas of the human brain, the prefrontal cortex (PFC) and dentate gyrus of the hippocampus (DG), somatic SNVs accumulate over time (Lodato et al., 2018, Science). Importantly, mutation types were not random but comprised specific signatures that displayed age, brain region, and disease status specificity. These data suggest that specific pathways generate DNA damage and mutation in the brain during aging, suggesting pathways that can be intervened upon therapeutically to protect the genome during life.

Website aging fig 2.tif

Somatic mutations accumulate during aging and in disease. A. Somatic SNVs increase in PFC neurons (blue) and DG neurons (red) during life. Each point represents one neuron. N = 93 PFC neurons, 15 cases and 26 DG neurons, 6 cases. The mutation rate was higher in DG than PFC, P=8x10-4, mixed linear model. B. Somatic SNV rates are higher in two diseases of early-onset neurodegeneration, Cockayne Syndrome (CS) and Xeroderma Pigmentosum (XP). C, D. Two mutation signatures identified in single neurons. Both signatures comprise six cardinal substitution types, subdivided into 16 trinucleotide contexts. E. Signature A increases during aging in all neurons, regardless of brain area. F. Signature C, potentially linked to oxidative damage, is elevated in neurons from brains of neurodegenerative disease cases. * denotes p < 0.05, ANOVA. Modified from Lodato et al., 2018.