Aging is an inevitable process that affects every cell in the body, leading to decreased function, increased disease risk, and eventual decline. Scientists have long hoped to reverse this decline, inspired by results like exposing old animals to young blood (parabiosis) or using powerful gene cocktails (Yamanaka factors) that can rewind cells to a stem-like state. However, despite their promise, the lack of a cure or even a therapy to reduce the negative symptoms of aging suggests that the continued pursuit of new approaches is still needed. In a new study published in PNAS, BARI member Hao Li, with help from fellow member Saul Villeda and their labs set out to create a precise approach to identify new transcription factors that can rejuvenate aging cells. By making it easier for researchers to discover these factors, this work paves the way for targeted therapies that could treat aging at its molecular roots, ultimately extending healthy lifespans.
Building the discovery platform
To tackle this challenge, Li and his team engineered the Transcriptional Rejuvenation Discovery Platform (TRDP), an integrated pipeline that combines bioinformatics analysis with experimental validation. They first examined gene expression profiles (transcriptomes) from young and aged human fibroblasts, skin cells that recapitulate many features of natural aging after extensive culturing, such as reduced proliferation, senescence, and inflammatory signaling. Using computational tools, the team reconstructed regulatory networks of transcription factors and their downstream targets, identifying candidates likely to reverse age-associated transcriptional patterns.
The core of the platform relies on Perturb-seq, a high-throughput method that integrates CRISPR-based gene perturbations (activation or repression) with single-cell RNA sequencing. This approach enabled simultaneous testing of hundreds of transcription factors across thousands of individual cells, allowing the identification of perturbation-induced changes in gene expression. Perturb-seq data were analyzed to detect convergent effects (shared transcriptional signatures among effective interventions) and to prioritize factors that modulated aging hallmarks without inducing dedifferentiation. This modular, single-factor strategy ensured that cells retained their fibroblast identify, while regaining functional youthfulness, offering a more controlled alternative to multi-factor reprogramming methods.
Identifying key rejuvenators
The test-case screen identified four high-priority transcription factors: overexpression of E2F3 or EZH2, and repression of STAT3 or ZFX. These interventions produced robust rejuvenation signatures, with gene expression profiles in treated aged fibroblasts showing a strong negative correlation with aging-associated patterns and positive alignment with youthful states observed in early-passage cells or rejuvenation models such as heterochronic parabiosis. Cross-referencing with independent datasets, from aged human skin to various mouse tissues, confirmed that these factors engaged conserved anti-aging transcriptional programs across contexts and species.
A notable feature was the convergence of effects: despite targeting different regulators, the interventions influenced overlapping downstream pathways, including those governing cell-cycle progression, epigenetic regulation, and intercellular signaling. This convergence points to a core set of molecular mechanisms underlying rejuvenation. Crucially, none of the selected factors induced upregulation of oncogenes or other cancer-associated signatures, addressing a major safety concern associated with broader reprogramming approaches and supporting their potential as therapeutic candidates.
Demonstrating rejuvenation across models
Functional validation in cultured human fibroblasts showed that these transcription factor perturbations effectively reversed multiple hallmarks of cellular aging. Treated cells displayed increased proliferation, as indicated by elevated KI67 expression and accelerated growth rates. Several senescence and pro-inflammatory factors were significantly downregulated. Proteostatic capacity improved, with enhanced proteasome activity to clear misfolded proteins, while lysosomal function normalized to reduce age-related waste accumulation. Mitochondrial performance recovered as well, with restored membrane potential and upregulated genes involved in oxidative phosphorylation. Compared to partial reprogramming with Yamanaka factors, these single-factor interventions achieved rejuvenation with greater specificity, preserving fibroblast identify and avoiding stem-cell-like dedifferentiation. Microscopy, staining, and quantitative assays provided clear visual and numerical evidence of these restorative effects.
To assess whether these findings could translate to living organisms, the Li and Villeda teams tested EZH2 overexpression in the livers of aged mice, a tissue susceptible to steatosis, fibrosis, and metabolic dysfunction with age. Using adeno-associated virus (AAV) vectors to deliver EZH2, they observed substantial transcriptional rejuvenation after just three weeks: thousands of age-deregulated genes reverted toward youthful expression patterns. Further, there was less lipid accumulation in livers, diminished fibrosis, and glucose tolerance tests demonstrating improved metabolic control. Importantly, there was no evidence of oncogenic gene profiles in rejuvenated mice. The consistency of effects between human cells in culture and mouse tissue in vivo underscores the robustness of the identified mechanisms and the platform’s potential to uncover broadly applicable rejuvenation strategies.
Why this breakthrough matters
This study represents a significant advance in geroscience by providing a scalable, precise framework for discovering transcription factors that reverse aging phenotypes. The TRDP accelerates identification of therapeutic targets, potentially extending to other tissues such as brain, muscle, or heart, where age-related decline remains a major unmet need. By revealing conserved rejuvenation pathways, this work brings the field closer to translating molecular insights into therapies that extend the number of years lived in good health.
What’s next for the Li lab?
Looking ahead, the Li lab plans to extend the TRDP framework to additional model systems, including other cell types such as neurons and disease-relevant contexts. The team will also evaluate additional candidate factors identified in their screen across multiple mouse tissues, assessing their effects on healthspan and lifespan. Ultimately, the goal is to build integrative systems-biology models to better understand the mechanisms of aging and rejuvenation across human tissues, and to translate these insights into therapeutic strategies.