Single-Cell Assessment of Genomic Mosaicism and Transcriptomic Heterogeneity in the Human Brain

The human cerebral cortex makes up approximately 82% of the total brain mass, has 52 distinct Brodmann areas, and contains approximately 16 billion neurons. In recent years, neuroscientists, geneticists, bioengineers, and bioinformaticians, by working in collaboration have only begun to scratch the surface towards understanding the enormous cellular complexity and heterogeneity that exists in our brains. My thesis work in has focused on the investigation of the immense diversity that comprises both the genomic and the transcriptomic landscapes of the human brain through the use of traditional, and newly engineered, single-cell technologies. Neuronal genomic mosaicism – the phenomenon wherein neurons possess unique somatically altered genomes – was first identified as mosaic aneuploidies, a gain or loss of an entire chromosome. In recent years, multiple labs have now demonstrated that the somatic genomic changes also include LINE-1 retrotransposons, both large (>10 megabases (Mb)) and small (<1 Mb) copy number variations (CNVs), single nucleotide variants (SNVs), and regional patterns of total DNA content changes referred to as DNA content variation (DCV). Genomic mosaicism has potentially important implications for the development of somatic brain diseases of which there is no known cause. Sporadic Alzheimer’s disease (SAD), constitutes ~98% of cases, and bears a striking clinical and pathological resemblance to familial Alzheimer’s disease (FAD). This dissertation began by exploring the hypothesis that somatic genomic alterations occur in SAD brains to produce CNVs of AD related genes. We found that in the frontal cortex (FCTX) of SAD patients, neurons displayed an increase in total DNA content when compared to FCTX neurons from non-diseased patients. Importantly, this was accompanied by an increase in APP gene copy number, as determined by single-cell qPCR and peptide nucleic acid (PNA) fluorescent in situ hybridization (FISH), independent from chromosome 21 trisomy. In continuation of this work, I hypothesized that the DNA content increases observed may develop throughout the decades of life preceding Alzheimer’s disease symptoms. Down sydome (DS) patients, possess three copies of APP, a gene dosage effect strong enough to result in FAD. Therefore, to test this hypothesis I examined Down syndrome (DS) patient samples from ages 0-65 both with DSAD and without. We observed a steady increase in neuronal DNA content throughout life, that began to decrease again, at the onset of AD. Additionally, non-diseased neurons observed a similar increase in DNA content but, at a significantly lower rate. This age dependent increase in DNA content is highly significant, because it is some of the first evidence that neuronal mosaic increases are occurring post-mitotically. In addition, I have been an integral part of a highly collaborative single-cell transcriptome brain mapping initiative. As part of this team, we have sequenced the nuclear transcriptome of over 3,000 neurons from 5 different brain regions. We identified 16 different neuronal subtypes that were cortical layer and region specific. Together, these genomic and transcriptomic assessments are integral to understanding the normal functioning of the human brain, in order to better understand neurological diseases