Hypermutator neurons as a genomic reservoir that fuels adaptive synaptic rewiring under energetic constraints in the aging brain

Background Somatic SNVs and CNVs accumulate in cortical neurons at a rate of ~16‑40 per neuron per year, reaching ~2,500 by age 80 1 2. Contrary to the idea of selective eviction of inefficient neurons, hypermutator neurons—those carrying 2.3‑2.5‑fold higher mutation loads—persist and can undergo clonal expansion in aged brains 3 4. This raises the possibility that the brain retains, rather than discards, genetically diverse neurons as a substrate for adaptation when energy supply declines.

Hypothesis We propose that hypermutator neurons are not passive by‑products of DNA damage but constitute a functional genomic reservoir. Their elevated mutational burden creates a spectrum of altered protein isoforms and regulatory sequences that, when expressed, modulate neuronal excitability, metabolic efficiency, and synaptic plasticity. Under declining ATP availability, the brain preferentially engages these variant‑rich neurons to rewire circuits in ways that conserve energy while preserving information processing. Thus, neuronal loss in aging reflects the retirement of low‑variance, high‑cost neurons, whereas hypermutator neurons are retained and repurposed.

Predictions

1. Hypermutator neurons will show enriched expression of activity‑dependent genes linked to metabolic adaptation (e.g., Pgc‑1α, Ucp2) compared with neighboring low‑mutation neurons.
2. Silencing clonal expansion of hypermutator neurons (via CRISPR‑based lineage ablation) in aged mice will accelerate decline in energy‑sensitive cognitive tasks without affecting baseline motor function.
3. Transcriptomic single‑nucleotide resolution profiling will reveal a higher frequency of splice variants and novel open reading frames in hypermutator neurons that correlate with increased synaptic gene network flexibility.
4. Providing an exogenous energy substrate (e.g., ketone bodies) will reduce the relative contribution of hypermutator neurons to network activity, shifting usage back toward low‑mutation neurons.

Experimental Approach

• Use inducible Cre‑loxP reporters driven by a somatic mutation‑sensitive promoter (e.g., p53‑responsive element) to label hypermutator neurons in mice aged 18‑24 mo.
• Perform patch‑seq or Patch‑seq‑like assays to link mutational load (via targeted single‑cell DNA sequencing) with electrophysiological properties and transcriptome.
• Apply chemogenetic silencing (hM4Di) specifically to the labeled hypermutator population during a delayed‑alternation T‑maze task that taxes working memory under caloric restriction.
• Measure local ATP levels with luciferase‑based sensors and assess network calcium imaging to evaluate shifts in neuronal participation.
• Parallel human post‑mortem analysis: isolate laser‑captured neurons from prefrontal cortex, quantify SNV burden via duplex sequencing, and correlate with mitochondrial gene expression signatures.

Implications If validated, this hypothesis reframes neuronal loss in aging as a metabolic trade‑off rather than a failure of quality control. It suggests that enhancing the brain’s ability to harness genetically diverse neurons—through metabolic interventions or neuromodulation—could mitigate cognitive decline without attempting to prevent the underlying mutational accumulation, which appears inevitable.