Cellular senescence contributes to age-related diseases, and clearing it extends lifespan in mice. Partial reprogramming has already shown a rejuvenating effect, but it was unclear whether it acts directly on senescent cells. This work sits alongside studies of epigenetic rejuvenation and the links between aging and cell plasticity, yet it is the first to systematically test how exactly reprogramming alters the senescence phenotype.
Mouse embryonic fibroblasts were driven into senescence in three ways: irradiation, activation of the oncogene HrasV12, or treatment with nutlin-3. Then, over six days, the Yamanaka factors (Oct4, Sox2, Klf4, c-Myc) were switched on three times for 48 hours each via a doxycycline-inducible system, or a chemical cocktail of tranylcypromine and RepSox was added. The researchers assessed the transcriptome, secretome, mitochondria, and apoptotic sensitivity by BH3 profiling. In in vivo experiments, old 29-month-old mice received the cocktail, and their behavior, as well as aging markers in tissues and blood, were examined.
Both interventions shifted the transcriptome of senescent cells toward a younger state, but did not restore proliferation and did not reduce the activity of senescence-associated beta-galactosidase. At the same time, the expression and secretion of SASP inflammatory factors dropped sharply: interleukins 6 and 1b, Ccl2, Timp1 and others. The secretome of treated cells lost the ability to induce paracrine senescence, attract macrophages, and enhance reprogramming of neighboring cells. Mitochondria returned to normal: excess mass decreased, membrane potential and ATP production were restored, and reactive oxygen species levels fell. Apoptotic readiness, measured by cytochrome c retention, returned to the level of young cells. In old mice, the cocktail improved performance in tests of coordination, strength and memory, reduced the expression of pro-inflammatory genes in liver and kidneys, and lowered the level of SASP factors in serum, without reducing the number of senescent cells.
The work shows that partial reprogramming acts as a senomorphic: it weakens the harmful functions of senescent cells without killing them or reversing the division arrest. Unlike neighboring studies, where the effect was often assessed by epigenetic clocks or individual markers, this one systematically dissects the cellular mechanisms for the first time: mitochondria, SASP, apoptosis. The chemical cocktail proved more selective than genetic factors: it suppressed mesenchymal drift and TGF-beta signaling without erasing somatic identity.
Chemical partial reprogramming could become the basis for treating age-related diseases, if safety can be confirmed over longer periods and in other models. An open question remains: which components of the cocktail are responsible for the effect? Tranylcypromine inhibits monoamine oxidases on the outer mitochondrial membrane, which in itself may improve their function, while RepSox directly blocks the TGF-beta receptor. Separating these contributions will help create more precise drugs.
The results matter for gerontology, regenerative medicine and oncology, because they show how to control the secretory phenotype of senescent cells without the risk of tumor transformation.
It is worth testing the long-term effects of the cocktail in mice with age-related pathologies, and also finding out whether it acts on human senescent cells in vivo.
The work directly addresses the unresolved question of whether the harmful sides of cellular senescence can be separated from the protective division arrest. It also links aging to mitochondrial dysfunction and mesenchymal drift, offering concrete targets for intervention.
Senescent cells don't just passively grow decrepit: they actively release dozens of inflammatory molecules, and this "secretory phenotype" can accelerate the aging of neighboring healthy cells. In this study, the secretome of treated cells lost the ability to attract macrophages and induce senescence in neighbors, even though the cells themselves remained old by classical markers.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.