Menu
Alzheimer’s disease (AD) is a progressive neurodegenerative disorder characterized by the pathological deposition of amyloid-beta and hyperphosphorylated tau (P-tau) proteins in the brain (1). P-tau accumulates in neurons and is the pathology most strongly associated with AD severity (2) and affected brain regions (3). However, only a subset of neurons in affected tissue exhibits tau pathology. The mechanism underlying heterogeneous tau pathology and its contribution to AD is not well understood.
During aging, neurons accumulate somatic mutations, despite their non-dividing state (4). In AD, neurons acquire even greater somatic mutations, with oxidative and other disease-associated mutational signatures (5), prompting the question of whether somatic mutations are driven by disease pathology. We hypothesized that neurons with pathological P-tau deposition will carry more somatic mutations than neurons without P-tau.
During aging, neurons accumulate somatic mutations, despite their non-dividing state4. In AD, neurons acquire even greater somatic mutations, with oxidative and other disease-associated mutational signatures5, prompting the question of whether somatic mutations are driven by disease pathology. We hypothesized that neurons with pathological P-tau deposition will carry more somatic mutations than neurons without P-tau.
We developed a method to isolate single neuronal nuclei based on cytoplasmic P-tau aggregation with fluorescence-activated nuclear sorting. Immunofluorescence microscopy showed tau adherent to the outside of nuclei, and an AD-control mixing experiment demonstrated that this sorting method effectively enriches P-tau+ nuclei (>380 fold) with high accuracy (98%).
We then performed single-cell whole genome sequencing (scWGS), by primary template-directed amplification, on nuclei from AD brain prefrontal cortex. We analyzed scWGS data of P-tau sorted neuronal nuclei from 5 AD prefrontal cortex samples, with a mean of 4 P-tau+ and 4 P-tau- nuclei per individual. We compared P-tau sorted neurons with 29 tau-undetermined neurons from the same AD samples and 52 neurons from neurotypical control individuals. We utilized SCAN2 to identify single-cell somatic mutations and performed mutational signature analysis.
We found that total somatic single-nucleotide variant (sSNV) burden and oxidation-related mutations are both elevated in the P-tau+ and P-tau- neurons (P-tau- vs control: +186 sSNV/neuron, p = 2.35e-3; P-tau+ vs control: +165 sSNV/neuron, p = 2.69e-3). Notably, P-tau+ and P-tau- neurons do not differ significantly in sSNV burden (P-tau+ vs P-tau-: +19 sSNV/neuron, p =0.75) and oxidative signature mutations.
Similarly, P-tau+ and P-tau- neurons carry more indels (insertions and deletions) than control neurons, but we observed no significant difference between P-tau+ and P-tau- neurons. We found that a specific indel mutational signature, ID4, is elevated in both P-tau+ and P-tau- neurons compared to control neurons. ID4 has previously been reported as a cancer mutation signature related to defective ribonucleotide excision repair.
We observed no significant differences in somatic SNV and indels between P-tau+ and P-tau- neurons in AD, with both accumulating mutations more than control neurons. This unexpected result suggests that P-tau aggregation, the most prominent hall marker for AD, occurs independently of somatic mutation. Paired with a recent report that gene expression profiles are similar between P-tau+ and P-tau- neurons, our findings suggest that DNA damage and tau accumulation may not have a causal relationship, but instead may each be a consequence of upstream etiological change in AD.
In summary, this study is the first to investigate the somatic genomic profiles of tau-specified single neurons and demonstrates an independent relationship between somatic mutation accumulation and P-tau aggregation.
1. PMID: 22101365.
2. PMID: 31780819.
3. PMID: 32572268.
4. PMID: 29217584.
5. PMID: 35444284.
