Behind a paywall
Senescence has traditionally been studied largely on the basis of changes in gene expression, but senescent cells also undergo profound biochemical remodeling. Zhang and colleagues introduce RamanOmics, a multimodal approach that links Raman-based chemical imaging with spatial transcriptomics to reveal the hidden biochemical landscape of cellular senescence.
If a person has access please post a copy.
From;
Cellular senescence is typically defined by changes in gene expression and the secretion of inflammatory factors, yet locating and characterizing these cells within intact tissues remains a major challenge. A new multimodal analytical approach called RamanOmics integrates label-free Raman imaging with spatial transcriptomics to map both the biochemical and transcriptional landscapes of senescent cells. This technique revealed that while senescent cells exhibit tissue-specific gene expression, they share a common biochemical fingerprint characterized by enriched lipid-associated molecular structures. This framework provides a new multidimensional method to identify senescent cell states during aging and tissue repair.
For decades, researchers have relied on a fragmented collection of molecular clues to identify senescent cells. These clues include cell-cycle inhibitors like p21 and p16, positive SA-beta-gal staining, DNA damage accumulation, and the senescence-associated secretory phenotype. However, there is no universal marker that unambiguously identifies senescent cells in complex tissues. This difficulty arises because senescence is not controlled by a single switch but is a coordinated alteration across epigenetic, transcriptional, proteomic, and metabolic layers.
Zhang and colleagues introduce a solution called RamanOmics, designed to visualize this multidimensional identity directly within intact tissues. Current single-cell RNA sequencing methods capture transcriptional programs but fail to measure the dynamic metabolite profiles and biochemical composition of individual cells. Raman spectroscopy bridges this gap by detecting intrinsic molecular vibrations, capturing chemical information without the need for labels or extensive sample preparation. Because Raman signals alone cannot identify which genes a cell is expressing, the researchers combined it with STARmap spatial transcriptomics on the same tissue sections.
Applying this to aging mouse lung and skin, the researchers discovered both tissue-specific and shared features of senescence. In lung tissue, p21-positive senescent cells showed transcriptional programs linked to extracellular matrix remodeling and transforming growth factor-beta signaling, involving genes like Serpine1, Dab2, and Igfbp7. In skin tissue, the p21-positive cells showed a different program enriched for keratinization and barrier-related genes, such as Krt10, Lor, and Sbsn.
Despite these divergent gene expressions, a common biochemical pattern emerged. Specific Raman features corresponding to lipid-associated molecular structures were enriched in p21-positive cells across both tissue types. To capitalize on this, the team developed a machine-learning based multimodal barcode that integrates the Raman features with gene-expression patterns. This combined approach achieved better classification performance for senescent cells than using either modality alone. The researchers also applied RamanOmics to injured skin during wound healing, detecting transient p21-positive cells with coordinated transcriptional and lipid-associated biochemical changes, proving the utility of the tool beyond chronological aging.
The provided source material is a review of a diagnostic methodology rather than a clinical or therapeutic intervention. Consequently, there are no treatment protocols to implement, and statistical effect sizes cannot be calculated from the text. However, the conceptual shift offered by this research provides practical foresight for health optimization strategies.
The discovery that lipid-associated structural changes form a universal biochemical fingerprint for senescence strongly suggests that lipid remodeling is a core driver or consequence of cellular aging. For individuals monitoring biological age, this indicates that future diagnostic testing will likely move away from isolated single-protein assays toward comprehensive lipidomic and metabolic profiling. Practically, this reinforces the importance of interventions that regulate lipid metabolism and membrane health. Maintaining healthy lipid oxidation rates and cellular clearance mechanisms may be just as critical for managing senescent cell burden as targeting traditional genetic pathways.