Amyotrophic lateral sclerosis (ALS) destroys the motor neurons that control movement. In roughly 97% of cases, clumps of the protein TDP-43 turn up in cells, but some patients with a long disease course have none of these clumps. Why some neurons die while others hold out for years remained unclear. This work sits among studies that hunt for molecular traces of disease in the brain in dementia and other neurodegenerative conditions, but it focuses on a rare group: veterans with ALS and no obvious genetic mutations. That approach makes it possible to see what happens inside cells when the disease develops sporadically rather than from a known defective gene.
The researchers took frozen and paraffin-embedded samples from three regions: the motor cortex, the cervical spinal cord, and the lumbar spinal cord. First they measured the activity of hundreds of genes using NanoString panels. Then, on adjacent sections, they applied GeoMx spatial proteomics: a laser cut out patches around individual motor neurons tagged with fluorescent antibodies, and from each patch proteins were collected for quantitative analysis. In total, about 3500 neurons were processed. To separate the effect of phosphorylated TDP-43 from the overall amount of the protein, for each neuron they calculated how far its signal deviated from the case average, then matched those deviations against the levels of other proteins.
In the motor cortex, ALS with TDP-43 switched on immune response genes and suppressed synapse genes. In the spinal cord, ALS without TDP-43 altered genes for immune receptors and metabolism, while ALS with TDP-43 additionally turned on cell cycle and aging genes. Spatial proteomics confirmed it: in the spinal cord, both forms of ALS pushed neurons to behave like senescent cells, though the cortex showed nothing of the kind. What's more, in the lumbar region of patients without TDP-43, the link between cell cycle and metabolism programs, normally negative (correlation coefficient −0.97), weakened to −0.20. This means neurons stopped keeping these programs clearly apart. Phosphorylated TDP-43 reshuffled a neuron's protein networks more strongly than total TDP-43 did: it was linked to 4–5 times as many proteins that change behavior only in disease.
The work shows that motor neuron aging may be a common denominator across different forms of ALS, even when the pathological TDP-43 is absent. That sets it apart from neighboring studies, which mostly tie brain changes to the buildup of specific proteins. Here, for the first time at the level of individual neurons, the aging program is seen switching on in the spinal cord regardless of TDP-43, and in the cortex only when it is present. This regional contrast hints that the disease may follow different scenarios in different parts of the nervous system.
If neuronal aging really does make neurons vulnerable, one could try to slow the disease by clearing out senescent cells or suppressing their inflammatory signals. In models of Alzheimer's and Parkinson's disease, such approaches have already improved the condition of animals. For ALS this is still a hypothesis, but it is testable: it would be enough to compare how senolytics (drugs that kill senescent cells) affect motor neurons in culture or in mice with ALS-causing mutations.
The results could influence drug development for ALS, shifting the focus from clearing TDP-43 to keeping aging neurons healthy. They are also useful for understanding other diseases where neurons resist death for a long time, such as Alzheimer's disease.
We need to check whether senescent motor neurons really do release inflammatory molecules, and whether they can be safely removed in ALS models. It is also worth studying why aging is tied to TDP-43 in the cortex but not in the spinal cord.
One question in neurodegeneration remains unresolved: why do some neurons die quickly while others live for years, despite the same genetic background. This work links survival to aging, but it doesn't explain what starts the aging program when no mutations are present. The key may lie in chronic stress or the buildup of damage that ordinary staining can't reveal.
Senescent neurons are rare: in the human brain they make up just 2–3% of all cells. To catch such one-of-a-kind cells, the team had to measure proteins in individual neurons rather than in a chunk of tissue, where their signal gets lost among thousands of other cells.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.