Raman barcodes identify senescent cells in aged mouse tissue

Researchers at the Massachusetts Institute of Technology have combined nondestructive Raman microscopy with spatial RNA sequencing to develop molecular “barcodes” for identifying senescent cells in aged mouse tissue. The study examined skin and lung from two-month-old and 26-month-old mice and found that selected combinations of Raman spectral bands corresponded with gene-expression signatures associated with senescence. The peer-reviewed study was published in Nature Aging.

Cellular senescence is a state in which cells stop dividing without dying. It can be triggered by DNA damage that leads to an irreversible halt in the cell cycle. Senescent cells then undergo substantial changes in shape, metabolism, lipid production and gene activity.

As people age, the immune system can become less efficient at removing these cells, allowing them to accumulate. The source research links this accumulation with age-related disorders such as cancer, tissue degeneration and inflammatory diseases, while also noting that senescence participates in embryonic development and tissue regeneration. Identifying the cells accurately is therefore important for understanding both normal physiology and disease.

Researchers have previously identified senescence-associated proteins, including p16 and p21, which are involved in arresting the cell cycle. The disadvantage of existing approaches is that detecting these proteins destroys the cells being analyzed. The MIT team sought a nondestructive alternative based on Raman microscopy, a technique that reveals the chemical composition of cells or tissues by illuminating them with visible or near-infrared light.

Chemical bonds produce characteristic peaks in the resulting Raman spectrum, allowing the method to detect features associated with lipids, proteins and other molecules without consuming the sample. The team paired Raman microscopy with spatial RNA sequencing, which reveals gene activity and where active genes are located within tissue. By examining the same cells with both methods at single-cell resolution, the researchers could relate Raman features to gene-expression signatures, cellular positions and other biochemical information.

Senior author Jian Shu described the rationale as viewing senescence from two complementary perspectives rather than relying on a single type of molecular measurement. The resulting data contained a much broader characterization of senescent cells than either technique could provide independently. From the paired measurements, the investigators identified combinations of Raman peaks that correlated with senescence.

The researchers then combined the most informative Raman features with the most informative gene signatures to create a barcode for recognizing senescent cells. This approach focuses on a small number of spectral bands rather than requiring a lengthy display of the full spectrum, and the researchers described it as a more unbiased way of identifying the cells. The barcode is a pattern calibrated against gene-expression data from the two mouse tissues studied, rather than a universal signature established across the body.

One of the clearest changes detected in both organs was an increase in lipid synthesis and lipid accumulation in cells from aged tissue. The measurements showed a difference between young and old samples, but the researchers said they do not yet know how the additional lipids affect cellular physiology. The results also differed between tissues.

Senescent skin cells showed significant effects on pathways associated with muscle contraction and with remodeling collagen and the extracellular matrix. In aged lung tissue, the researchers found greater activity of genes involved in immune activation and inflammation. These findings document distinct molecular features in the two tissues as well as a shared lipid-related change.

Preserving the sample is one of the method’s main technical advantages. Raman microscopy can reveal chemical features while leaving cells available for complementary analyses, whereas RNA sequencing consumes the cells as it measures them. The approach would still require substantially faster imaging before it could support practical diagnostic uses.

At present, the team’s Raman system takes about 30 hours to analyze a tissue sample approximately one square millimeter in size. The researchers are working on a higher-speed version intended to identify the barcodes in larger samples, a necessary step for moving from a research demonstration toward clinical application. “You can imagine that one day we may develop an endoscope that can look inside your body and identify cellular senescence,” said Jeon Woong Kang, an MIT research scientist and a senior author of the study.

The work forms part of the National Institutes of Health’s Cellular Senescence Network, an initiative pursuing a comprehensive understanding of senescence in normal physiological and pathological settings. Senior author Peter So said the broader aim is not to treat every senescent cell as harmful, because the process also has important roles in normal biology. Instead, the network seeks to identify the cells and understand the functions that their molecular changes support or disrupt.

Several limitations bound the current findings. The experiments used mouse skin and lung tissue, and the researchers are now working to adapt the method for human tissue. The comparison between two-month-old and 26-month-old animals identified age-associated differences, but it did not follow the same cells from youth into old age.

The physiological significance of the increase in lipids also remains unresolved. In addition, the study did not demonstrate an endoscope, test diagnostic performance in people, or show that detecting senescent cells could improve clinical outcomes. The results therefore establish a research-stage method for identifying senescence-associated patterns in the tissues examined, not a validated human diagnostic.

Despite those constraints, the work combines nondestructive spectroscopy with spatially resolved transcriptomics in a way that could help researchers map senescent cells while retaining samples for further molecular study. Its central result is that carefully selected combinations of Raman bands can encode a senescence-associated pattern when calibrated against single-cell gene-expression data. The team is now pursuing two concrete developments: adapting the barcode for human tissue and increasing imaging speed so that it can be applied to larger areas.

Whether those advances will eventually support in vivo detection or clinical diagnosis will depend on further testing beyond the aged mouse skin and lung samples analyzed in this study.

Subject of Research: Raman and spatial RNA profiling of cellular senescence in aged mouse tissue

Article Title: MIT researchers unmask “zombie cells” in aging tissue with an AI-powered barcode

Article References: MIT researchers unmask “zombie cells” in aging tissue with an AI-powered barcode. Original publication (Bibliographic details unavailable.)

Image Credits: AI Generated

DOI: Not provided

Keywords: cellular senescence, Raman microscopy, spatial RNA sequencing, Raman barcodes, aging, senescent cells, lipid accumulation, mouse tissue, Nature Aging, Massachusetts Institute of Technology, researchers, unmask