In the sacral spinal cord of a patient with amyotrophic lateral sclerosis, something strange holds on. Surrounding motor neurons are dying, their interiors a mess of misfolded protein, their nuclei emptied of the machinery they need. But a small population called Onuf’s nucleus keeps its architecture intact. TDP-43, the protein that has abandoned its post and clumped into toxic aggregates in the neighbouring cells, stays where it belongs in Onuf’s neurons: tucked inside the nucleus, doing its job. The question is why. And the answer, according to new research from the Instituto de Neurociencias in Spain, may tell us something important about why ALS kills the cells it does and spares the ones it doesn’t.
The finding points to a cellular cleaning mechanism called chaperone-mediated autophagy, and its apparent absence in the motor neurons that ALS destroys.
Autophagy, broadly speaking, is a process by which cells break down and recycle their own components. Think of it less as rubbish collection and more as a triage system that identifies damaged or surplus proteins and routes them to lysosomes for disposal. There are several variants. Macroautophagy is something like bulk waste disposal: it scoops up whole sections of cytoplasm. Chaperone-mediated autophagy (CMA) is considerably fussier. It works by recognising specific proteins that carry a particular chemical signature (a short sequence of amino acids sometimes called the KFERQ motif), binding to them via a chaperone protein called HSC70, and threading them one by one through a receptor called LAMP2A into the lysosome’s interior. CMA selectively degrades roughly 30 to 35 percent of the proteins drifting around in the cellular cytoplasm, including proteins involved in stress response, metabolism, and immune function.
TDP-43 carries the KFERQ signature. Which means, in theory, that CMA should be able to clear it. The problem is what happens when CMA fails.
A Cleaning System Gone Quiet
Salvador Martínez and colleagues at the UMH-CSIC Institute for Neurosciences examined spinal cord tissue from ten sporadic ALS patients and six healthy controls, using LAMP2A as a proxy for CMA activity. The results were stark. “In our study, we have shown that motor neurons require very high levels of chaperone-mediated autophagy to survive,” says Martínez, who directed the research. “When this mechanism declines, as occurs in ALS, these are precisely the cells that are first affected and eventually die.”
In healthy spinal cords, motor neurons showed intense, consistent LAMP2A expression concentrated around the cell nucleus, exactly where you’d expect if CMA were running at high capacity. In the ALS patients’ tissue, LAMP2A was markedly reduced in almost all motor neurons across the length of the spinal cord. About 5 to 10 percent of motor neurons in the ALS cases retained some LAMP2A signal, possibly reflecting normal variation or cells in earlier stages of decline. Critically, the other autophagy pathway, macroautophagy, appeared unaffected. LC3, its marker, was just as abundant in ALS motor neurons as in controls. The lysosomal enzyme GBA was similarly unchanged. This wasn’t a generalised failure of cellular housekeeping. It was specific, and it was specific to the selective pathway that handles TDP-43.
“ALS is a devastating disease whose cause remains unknown in the vast majority of patients, which greatly hampers the development of effective treatments,” Martínez says. “Identifying cellular mechanisms directly involved in neuron survival is a key step towards advancing new therapeutic strategies.”
The Clue Hidden in Resistant Cells
The Onuf’s nucleus result is the piece of the puzzle that makes the story cohere. Onuf’s nucleus, a group of motor neurons controlling sphincter and pelvic floor muscles, has been known to be relatively spared in ALS for decades; patients typically lose bladder and bowel function late in the disease, or sometimes not at all, even as limb and respiratory muscles fail. Nobody has had a satisfying explanation for why. The new study found that Onuf’s neurons in sALS patients had strong LAMP2A expression and no TDP-43 pathology. Their neighbours in the same spinal cord sections, the motor neurons that were dying, had depleted LAMP2A and cytoplasmic TDP-43 aggregates. One patient, identified as ALS40, showed a pattern somewhere in between: higher LAMP2A than the other patients and correspondingly less TDP-43 proteinopathy, suggesting a slower or less severe trajectory of CMA decline.
The team also found supporting evidence in the tissue itself: in healthy motor neurons, TDP-43 and LAMP2A were observed together in the cytoplasm, colocalized in the way you’d expect if CMA were actively processing TDP-43 as part of normal maintenance. “We have been able to observe this mechanism directly in human tissue, something we had not achieved in animal models,” Martínez notes.
It isn’t entirely obvious why motor neurons would be so much more dependent on CMA than other cell types. The paper points to their unusual physiology: they are among the largest neurons in the nervous system, firing at high rates for a lifetime, which makes enormous demands on protein quality control. The bigger and more active the cell, perhaps, the more it needs the specific, targeted cleaning that CMA provides. Neurons in Clarke’s thoracic column, another spinal cord population that doesn’t degenerate in ALS, showed low LAMP2A in the control tissue too, hinting that different neuron types have fundamentally different autophagy requirements. What confers resilience in one population may not travel across cell types.
A Potential Target, with Caveats
Glial cells in the ALS spinal cords told a different story. Rather than losing LAMP2A, they showed increased expression in the white matter tracts alongside degenerating axons, apparently ramping up CMA as part of an inflammatory response. The motor neurons seem unable to do the same. Whether that’s because of some intrinsic limitation in how they regulate LAMP2A, or because the pathological TDP-43 itself is somehow suppressing the pathway, remains unclear.
The findings, published in Acta Neuropathologica Communications, position CMA as a therapeutic target. “Our goal is to try to modulate this pathway to increase its activity,” says Martínez. That ambition is easier to state than to achieve. Current ALS treatments have a dismal track record: riluzole, the first approved drug, extends survival by perhaps 6 to 19 months. Edaravone failed to show benefit in an independent Italian trial after FDA approval. Tofersen, targeting the SOD1 mutation found in some familial cases, moved biomarkers encouragingly without producing meaningful clinical improvement. Relyvrio was withdrawn after a phase III trial failure. The field has learned to be cautious about mechanisms that look promising in tissue.
Still, there is something unusually direct about this result. The researchers could see it in the cells: a cleaning system that healthy neurons run at high capacity, absent in the ones that die, preserved in the ones that survive. “These findings indicate that chaperone-mediated autophagy activity is clearly decreased in motor neurons from ALS patients,” says Daniel Garrigós García, the paper’s first author. The study was made possible in part by tissue donations from ALS patients enrolled in clinical trials, people who agreed to contribute their spinal cords to research they would not live to see completed. Whether what Martínez’s team found in those tissues eventually becomes a treatment will depend on whether CMA can be safely boosted in motor neurons specifically, and whether, even then, it would be enough to shift the trajectory of a disease whose cruelty lies partly in how fast and how completely it moves.
Source: Garrigos et al., Acta Neuropathologica Communications 14, 67 (2026)
Frequently Asked Questions
Why do some motor neurons survive ALS while others die?
This study suggests one significant factor may be the activity of a cellular cleaning system called chaperone-mediated autophagy (CMA). Motor neurons that survive ALS, such as those in a region called Onuf’s nucleus, appear to maintain high levels of a protein called LAMP2A, which drives CMA activity, and they show no toxic TDP-43 accumulation. Motor neurons that die in ALS show markedly reduced LAMP2A and are full of misfolded TDP-43 protein. Whether preserving CMA activity is the cause of their survival or simply a correlate remains an open question, but the pattern across patients was consistent enough to be striking.
What exactly is chaperone-mediated autophagy and why does it matter for ALS?
CMA is a highly selective protein disposal pathway that operates inside cells. Rather than bulk-clearing material the way standard autophagy does, it identifies specific proteins using a chemical tag and delivers them individually to lysosomes for breakdown. One of those tagged proteins is TDP-43, which accumulates in toxic clumps inside motor neurons in roughly 95% of ALS cases. If CMA is running properly, it should be clearing TDP-43 before it aggregates; when CMA fails, that clearance stops and the protein builds up. The new study found the broader autophagy pathway was intact in ALS motor neurons, making the CMA-specific failure look particularly significant.
Is it possible to boost this cleaning system as a treatment?
That is the direction the research is heading, though it remains early. Stimulating CMA in general has been explored in models of Parkinson’s and Alzheimer’s disease, where similar protein clearance problems occur. The challenge specific to ALS would be targeting the boost selectively to motor neurons, since the same pathway is involved in many other cell types and tissues. The researchers at UMH-CSIC say their goal is to modulate the pathway to increase its activity, but practical therapies are some way off, and ALS drug development has a long history of promising results in tissue that have not translated to effective treatments.
Why did it take until now to see this in human tissue?
Animal models of ALS, while useful, are mostly based on genetic mutations (particularly SOD1 mutations) that account for a small fraction of actual ALS cases. Sporadic ALS, which makes up over 90% of cases and has an unknown cause, is difficult to model faithfully in mice. Access to human spinal cord tissue from sporadic ALS patients requires not only donor willingness but also well-organised biobank systems and ethical infrastructure around clinical trials. This study drew on tissue from patients enrolled in clinical trials at the UMH-CSIC and IMIB centres, a resource built over years, which is part of why the finding could be made directly in the cells that matter rather than inferred from a model.
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.