Research Signal – Source date 2021-02-01
A cell that stops dividing sounds harmless. Senescent cells remain alive and can release signals that alter the cells around them, but arrested division and continued signaling alone do not establish senescence. A review published on February 1, 2021, examined how this stress-associated cell state might contribute to neurodegenerative diseases and brain tumors.
The interest is not simply that senescent cells can be found in diseased tissue. The more consequential question is whether they help drive disease, and whether removing them could change what happens next. The evidence summarized in the review comes from human tissue observations, cell studies, mouse models and early clinical investigation. Those forms of evidence answer different questions, so the treatment idea remains plausible rather than established for patients.
A cell that stops dividing is not necessarily silent
Cellular senescence is a survival response to damaging stress. Radiation, chemotherapy, repeated replication and oncogenic signaling can all push a cell into a stable arrest of the cell cycle while the cell remains alive and metabolically active. This differs from quiescence, in which a cell pauses proliferation but can return to the cell cycle when it receives the right growth signals.
Senescent cells also change their surroundings through the senescence-associated secretory phenotype, usually shortened to SASP. The SASP is not one substance. It is a changing collection of inflammatory mediators, growth factors, extracellular-matrix proteins and other biologically active molecules. These secretions can influence neighboring cells, immune activity, tissue structure and the behavior of cells that have not themselves become senescent.
That activity can be useful when senescence is temporary and the immune system clears the affected cells. Senescent cells take part in processes such as development, tissue repair and immune modulation. The concern arises when they persist. A continuing SASP can turn a short-lived protective response into a source of chronic inflammation or altered tissue behavior.
Why the brain makes senescence harder to identify
The brain adds an important complication. Neurons normally do not divide, so a lack of cell division cannot by itself identify a senescent neuron. Researchers instead look for a combination of features, such as accumulated DNA damage, cell-cycle inhibitor activity and senescence-associated secretions. Even those features require interpretation because no single marker is present in every senescent cell or exclusive to that state.
The review describes senescence-associated features in several nervous-system cell types, including neural stem cells, neurons, astrocytes, oligodendrocytes and microglia. These cells perform different jobs, and their responses to stress are not interchangeable. Their possible contribution to neurodegeneration could include persistent inflammation, reduced regenerative capacity and deterioration in the blood–brain barrier and microvasculature.
Human Alzheimer’s disease samples and experimental models provide observations consistent with this possibility. For example, the review reports increased expression of cell-cycle regulators such as p16 and p53 in post-mortem Alzheimer’s tissue compared with age-matched controls. Such findings show an association with disease tissue. They do not, on their own, show that senescent cells are causing cognitive decline.
What plaque-associated support cells reveal in Alzheimer’s models
A stronger test comes from removing a defined cell population in a disease model. One study discussed in the review identified senescent oligodendrocyte precursor cells, or OPCs, in human Alzheimer’s tissue and in an Alzheimer’s mouse model. OPCs are a subset of the brain’s glial support cells.
In both species, the senescent OPCs were associated with amyloid plaques. In the mouse model, the evidence linked amyloid fibrils with the senescence state, and the cells showed several features used to identify senescence, including markers of cell-cycle arrest and increased CDKN2A expression. Researchers then used the senolytic combination dasatinib and quercetin to selectively remove these cells in the mice.
The reported mouse results were notable because clearance was accompanied by lower IL-6 levels, reduced amyloid burden and improvement in the cognitive defects measured in that model. This intervention makes senescent OPCs look more like contributors to the disease process than passive bystanders in that experimental setting. It still does not establish improved cognition, slowed neurodegeneration or restoration of brain function in people.
The review also notes that a phase II Alzheimer’s trial using this senolytic combination had been initiated by the time of publication. That fact marks a move toward human testing, not a demonstrated clinical benefit. Whether senescent-cell clearance can help people with established disease, and whether it can preserve or restore neurological function, remained open questions in the review’s 2021 perspective.
Tumors expose two opposite roles
Brain tumors show why senescence cannot be treated as a universally harmful state. In one setting, senescence-associated signaling appears linked to growth restraint. In another, senescent cells may influence neighboring cells in ways that support tumor formation or recurrence.
Pilocytic astrocytoma is a grade I brain tumor that the review identifies as the most prevalent pediatric low-grade glioma. The review describes near-universal activation of the MAPK pathway in these tumors. In primary samples, that oncogenic signaling was associated with features of oncogene-induced senescence, including beta-galactosidase activity and increased p16 expression in many samples. SASP factors such as IL-1β and IL-6 were also elevated.
In this example, the secretory program was associated with maintaining growth arrest, and SASP expression was associated with a favorable prognosis. The review reports that suppressing the SASP with dexamethasone was followed by regrowth of senescent cells. Senolytic compounds could ablate senescent pilocytic astrocytoma cells in experimental work, but the findings do not support clearing every senescent cell from every brain tumor.
Adamantinomatous craniopharyngioma presents a different pattern. These tumors contain distinctive clusters of cells with senescence features, including cell-cycle arrest, DNA damage, an enlarged lysosomal compartment and activation of inflammatory signaling. Human tumor analyses found that the clusters activate a SASP containing inflammatory mediators and growth factors. Their position within tumor tissue suggested that their secretions could help remodel the surrounding epithelium and promote proliferation or invasion in nearby cells.
Mouse lineage-tracing experiments added a further distinction. The senescent clusters arose from SOX2-positive pituitary stem cells carrying oncogenic beta-catenin, while the main tumor tissue came from a different cell lineage. In the model described by the review, the senescent cells could therefore influence tumor formation without becoming the bulk of the tumor themselves. Reducing the senescence-associated response lowered the tumor-inducing potential of the clusters in mice.
These examples do not conflict once senescence is understood as a cell state with effects that depend on its setting, timing and neighboring cells. In pilocytic astrocytoma, senescence-associated activity was linked to maintained growth arrest. In craniopharyngioma, a senescent cell population appeared capable of driving changes in another population. The therapeutic implication is selective intervention, not blanket suppression.
The glioblastoma treatment sequence
Glioblastoma raises a more specific treatment problem. The review describes glioblastoma cell-line experiments in which temozolomide or radiation induced senescence. In the temozolomide experiments, this response involved DNA-damage signaling and p21 expression, followed by NF-κB activation and production of IL-6 and IL-8. In mice implanted with glioblastoma cell lines, temozolomide treatment also produced a senescence response, evidenced by p21 expression and NF-κB activation.
The proposed sequence is important. First, an existing treatment pushes some tumor cells into senescence. Next, a senolytic could be used to remove those surviving cells before they escape the arrested state. In the cell-line experiments summarized by the review, navitoclax, also known as ABT-263, selectively killed glioblastoma cells that had been driven into senescence by treatment or irradiation.
The reason for pursuing this sequence is the possibility that therapy-surviving senescent cells could later re-enter the cell cycle or create a microenvironment that favors tumor regrowth. Removing them might therefore make treatment more durable and reduce recurrence. That is a therapeutic rationale, not an observed reduction in recurrence or an improvement in survival for people with glioblastoma.
The target is a cell state, not a single marker
Before senolytics can be used safely, researchers need to know which cells should be removed and which should be left alone. This is difficult because senescent cells are heterogeneous. Their molecular programs vary with cell type, the stress that induced senescence and how long the state has lasted.
The marker p16 illustrates the problem. p16 expression can accompany senescence, but the review notes that it can also be found in non-senescent cells such as macrophages. Some senescent cells do not express it. A p16-positive cell is therefore not automatically a senescent cell, just as a p16-negative cell is not automatically free of senescence.
The SASP has similar limitations. Its components and intensity can vary substantially between cell populations. Some secreted factors have several biological roles, and overlapping functions among SASP components can make therapeutic targeting difficult. A drug that removes a particular marker-defined population, or suppresses one secreted molecule, may miss relevant senescent cells or affect healthy cells with overlapping features.
The review discusses several broad strategies, including killing senescent cells, modifying their secretions and strengthening immune clearance. Each approach faces the same underlying problem: the useful or harmful consequence of senescence depends on which cells are involved and what those cells are doing. A treatment that helps in one disease model could be ineffective or counterproductive in another.
From experimental change to patient benefit
The evidence therefore forms a chain rather than a single result. Human tissue findings can show that senescence-associated features occur in diseased brains or tumors. Cell experiments can show that a stressor induces a senescence-like state and that a compound can selectively kill treated cells. Mouse interventions can test whether changing a selected cell population alters pathology or behavior in that model.
Those steps are valuable, but they are not interchangeable with patient outcomes. A mouse cognitive measure is not the same as restored human memory. A reduction in amyloid burden in a model is not proof that neurodegeneration has been slowed in patients. Selective killing of treated glioblastoma cell lines does not establish fewer recurrences or longer survival in people.
There is also a practical safety question. Senescence can participate in tissue maintenance and repair, and a senolytic may not recognize only the disease-driving cells. The contrasting tumor examples show why suppressing all senescence-associated signaling would be too broad a goal. The relevant treatment may need to distinguish cell types, disease contexts and stages of disease, while preserving senescence that is serving a protective role.
By the review’s publication date, the field had a plausible biological explanation and encouraging preclinical experiments, but not an established brain-disease therapy. The next boundary was not simply finding a stronger senolytic. It was demonstrating that the intended cells can be identified and altered safely, then showing that the change produces meaningful benefit in humans.
What targeted cell clearance could change for patients
If researchers can safely identify and remove senescent cells that actively drive neurodegeneration, targeted clearance could help preserve brain function by interrupting a disease-driving cellular process before further damage accumulates. Preservation would be a distinct goal from recovering function already lost, which remains an unresolved possibility.
In brain tumors, the same principle could make treatment more durable by removing cells that survive chemotherapy or radiation in a senescent state and might later support relapse. That possibility depends on selective targeting, appropriate timing and evidence that the strategy benefits patients rather than only changing a laboratory model. The review supports these as important therapeutic goals, not as outcomes already established in people.
Key Sources
- Primary source: Cell senescence in neuropathology: A focus on neurodegeneration and tumours.
DOI or Journal Page · Review or Meta-Analysis · 2021-02-01
Keep exploring