Why Some Neurons Survive: Identity Maintenance May Set the Threshold of Neurodegeneration

Neurodegenerative diseases are strikingly selective. In Parkinson’s disease, dopamine-producing neurons in the substantia nigra pars compacta die early and extensively, while their close neighbors in the ventral tegmental area, which share the same neurotransmitter and arise from related developmental programs, survive for far longer. In Alzheimer’s disease, neurons of layer II of the entorhinal cortex are among the first casualties, and in amyotrophic lateral sclerosis, the fast-fatigable motor neurons fail before other motor neuron subtypes. This phenomenon, known as selective vulnerability, has been studied for decades, yet a provocative new essay in PLOS Biology argues that the field may be asking the wrong question. Instead of focusing exclusively on why some neurons die, researchers should also ask why others, exposed to largely the same pathological pressures, manage to live.

The essay, written by Dong Won Kim of the Danish Research Institute of Translational Neuroscience at Aarhus University, proposes a candidate mechanism that could help set the boundary between survival and degeneration: the capacity of a mature neuron to maintain the gene-regulatory machinery that sustains its subtype identity. In this view, a neuron is not a finished product bequeathed by development. Its defining features, including its transmitter enzymes, receptors, channels, and projection-specific molecules, may require continuous transcriptional and chromatin-based upkeep. If that upkeep weakens under stress, the neuron may not die immediately; it may first become less precisely itself. Kim proposes that the robustness of this maintenance machinery determines where the tipping point lies: above a critical stress level, the identity program recovers, whereas below it, regulatory loss becomes self-sustaining and degeneration follows.

The idea rests on two distinct claims that Kim is careful to separate. The first is that mature neuronal identities require active maintenance rather than passive inheritance. This claim has direct experimental support. In the roundworm Caenorhabditis elegans, so-called terminal selector transcription factors regulate batteries of genes that define mature neuronal features, and several of them are required throughout life, not just during development. When maintenance fails, neurons can retain broad neuronal character while losing the molecular features that made them a particular type. Work on the motor neuron selector UNC-3 shows that such factors also regulate transcriptional, functional, and metabolic programs that shift across the life course, complicating any simple division between identity genes and housekeeping genes.

Vertebrate examples reinforce the point. The transcription factors Pet-1 and Lmx1b support the maintenance of the serotonergic program, while FEZF2 illustrates how postmitotic cortical neurons actively repress inappropriate alternative identities. Equally important is repression by chromatin machinery. Loss of the Polycomb repressive complex 2, or PRC2, which silences inappropriate gene programs, causes de-repression and destabilization of subtype-specific expression in differentiated dopaminergic and serotonergic neurons. In adult striatal neurons, PRC2 loss produces early identity disruption followed by progressive, fatal degeneration. Together, these findings suggest that mature identity requires both the continued expression of subtype-defining programs and the continued repression of programs belonging to other cell types.

The second, more speculative claim is that differences in the robustness or stress sensitivity of these maintenance programs contribute to selective vulnerability. Kim emphasizes that evidence for this is limited and must be treated as a cell-type- and disease-specific hypothesis rather than a universal principle. The contrast between substantia nigra and ventral tegmental area dopamine neurons illustrates why the question is tractable. Classic explanations for nigral vulnerability point to large axonal arbors, autonomous pacemaking, calcium handling, and mitochondrial demand. The identity-maintenance framework does not replace these mechanisms. Instead, it proposes that their effects depend on how long a neuron can preserve its subtype-defining regulatory architecture as stress rises. Two populations facing comparable insults could fail at different stress levels because their maintenance programs differ in organization, buffering, or redundancy.

Notably, the transcription factor Nurr1, often invoked in Parkinson’s research, is expressed in both vulnerable and resistant dopamine populations, so its mere presence cannot explain their divergent fates. What may differ is how such factors are wired into networks, or how sensitive their chromatin landscape is to stress. Experimental work shows that Nurr1 is continuously required in adult dopamine neurons: conditional deletion causes progressive dopamine loss, motor impairment, and dystrophic changes in axons and dendrites. In a rat model of alpha-synuclein overexpression, pathological alpha-synuclein reduced Nurr1 and disrupted GDNF signaling, but forced Nurr1 expression restored Ret-dependent signaling and protected neurons despite ongoing stress. Whether this protection reflects preserved identity or separable survival outputs remains unresolved, and Kim outlines experiments that could distinguish the two models.

A second dopaminergic example strengthens the case that related populations organize shared identity programs differently. The transcription factor Pitx3 potentiates Nurr1-dependent transcription, and its loss devastates substantia nigra neurons while leaving ventral tegmental area neurons largely spared. When Pitx3 was deleted from mature dopamine neurons, mice developed age-dependent motor deficits, striatal dopamine loss, and profound degeneration of nigral neurons, with Nurr1 expression remaining stable in the surviving VTA cells, suggesting an alternative regulatory route operates there. Meanwhile, PRC2 disruption causes pronounced identity disturbance in nigral neurons even when survival is preserved, experimentally dissociating identity destabilization from cell death. Kim cautions, however, that these genetic ablations are not disease models, and that nigral fragility co-varies with established bioenergetic and anatomical liabilities, leaving the causal interpretation open.

A central contribution of the essay is a practical framework for distinguishing reversible state remodeling from genuine identity-maintenance failure. A stressed neuron normally changes state: it induces heat shock genes, alters mitochondrial transcripts, or activates an unfolded protein response while remaining recognizably dopaminergic. Identity-maintenance failure is different, because the constraints that stabilize the subtype itself begin to erode. Kim proposes that classification should require coordinated disruption across at least three of four prespecified domains: coupling of identity transcription factors to their targets, coherence of identity regulons and terminal effector genes, accessibility of identity-associated enhancers, and repression of normally excluded programs. Chromatin data from single-cell ATAC-seq and multi-omic comparisons are expected to be the strongest observational discriminators, since generic stress can remodel stress-responsive chromatin without touching the elements that define neuronal subtype.

The framework also makes a distinctive prediction about recovery. If identity-factor activity, enhancer accessibility, effector expression, and repression of alternative programs reinforce one another, then a deep perturbation may cross a point at which regulatory loss becomes self-sustaining, producing hysteresis: the route back to the original state would differ from the route out, and removing the initiating stressor might no longer suffice. Demonstrating this would require longitudinal, lineage-resolved measurements showing that the same cells occupy different persistent states depending on their prior trajectory, while excluding selective death of the most damaged cells. Causality, Kim argues, will ultimately require graded, bidirectional interventions compared against burden-matched non-identity controls, so that any shift in vulnerability can be attributed to architectural disruption rather than the generic cost of perturbing an essential regulator.

The implications extend across diseases and cell types. In Alzheimer’s disease, the RORB-marked excitatory neurons of the entorhinal cortex, which are selectively depleted and susceptible to neurofibrillary inclusions, offer a concrete test of whether an identity network erodes before terminal decline. In ALS, viral re-expression of the embryonic motor neuron factors ISL1 and LHX3 reactivated a youthful expression program and reduced pathology in SOD1(G93A) mice, suggesting that redeployed lineage regulators can improve stress buffering, even though these factors are not detectably active in the adult neurons examined. The essay also highlights the tissue environment, including microglia, astrocytes, and aging-related chromatin decay, as factors that may either alter the stress burden or the stability of the regulatory machinery itself. Therapeutically, the message is subtle: protecting a cell from acute stress and preserving the regulatory program that lets it recover may overlap but are not identical, and if maintenance failure can become self-reinforcing, treatment timing becomes part of the mechanism. Whether preserving identity architecture truly shifts the boundary between reversible remodeling and irreversible dysfunction is the empirical question on which this provocative framework will stand or fall.

Subject of Research: Selective neuronal vulnerability and cellular subtype identity maintenance in neurodegenerative disease

Article Title: Why are some neurons spared in neurodegeneration? Cellular subtype identity maintenance as a candidate threshold-setting mechanism

Article References: Kim, D. W. (2026). Why are some neurons spared in neurodegeneration? Cellular subtype identity maintenance as a candidate threshold-setting mechanism. PLOS Biology, 24(10), e3004014. https://doi.org/10.1371/journal.pbio.3004014

Image Credits: AI Generated

DOI: 10.1371/journal.pbio.3004014

Keywords: neurodegeneration, selective vulnerability, Parkinson's disease, dopamine neurons, Nurr1, Pitx3, PRC2, gene regulation, chromatin, neuronal identity, Alzheimer's disease, ALS