Researchers at MIT have created a noninvasive way to detect senescent cells, the so-called "zombie cells" that accumulate as we age and contribute to diseases like cancer, tissue degeneration, and inflammation. By combining Raman microscopy with AI-powered analysis of gene expression data, the team identified unique biochemical "barcodes" that can quickly identify these problematic cells in living tissue.

What Are Senescent Cells and Why Do They Matter?

Senescent cells are cells that have stopped dividing but refuse to die. As we age, our immune system becomes less efficient at clearing them out, allowing them to accumulate in tissues throughout the body. This buildup can trigger inflammation, weaken muscles, cause sagging skin, and contribute to chronic conditions like osteoarthritis and type 2 diabetes.

The challenge for researchers has been identifying these cells without destroying them in the process. Traditional methods rely on detecting proteins like p16 and p21, which mark senescent cells, but these detection techniques require killing the cells to analyze them. The MIT team wanted a better approach.

How Does the New Detection Method Work?

The breakthrough combines two complementary technologies. Raman microscopy uses near-infrared or visible light to reveal the chemical composition of cells without harming them, while spatial RNA sequencing shows which genes are active within tissue. By analyzing both the biochemical fingerprint and gene expression patterns simultaneously, the researchers created a comprehensive picture of what makes a senescent cell unique.

The team tested their approach on skin and lung tissue from young mice (2 months old) and older mice (26 months old). One of the most striking findings was a dramatic increase in lipid synthesis and accumulation in older cells across both tissue types. In senescent skin cells, they also observed changes in pathways related to muscle contraction and collagen remodeling, while aged lung tissue showed increased immune activation and inflammation.

"Our idea was to look at many different features to characterize senescence. That's why we wanted to combine both single-cell gene expression and Raman microscopy, so that we can characterize the senescence from two complementary views," explained Jian Shu, an assistant professor at Massachusetts General Hospital and Harvard Medical School.

Jian Shu, Assistant Professor at Massachusetts General Hospital and Harvard Medical School

How to Identify Senescent Cells Using AI-Powered Barcodes

• Raman Spectroscopy Analysis: The system shines light on tissue samples and analyzes the scattered light to identify specific chemical bonds linked to lipids, proteins, and other molecules characteristic of senescent cells.
• Gene Expression Mapping: Spatial RNA sequencing reveals which genes are active in specific locations within the tissue, providing a molecular signature of senescence.
• Barcode Creation: By combining the most informative Raman peaks with key gene signatures, researchers created a unique barcode that can identify senescent cells quickly and without bias.

Using these barcodes, researchers can now focus on just a few specific Raman bands that emerged as most informative, potentially enabling faster diagnostics. Currently, analyzing a tissue sample about one square millimeter in size takes approximately 30 hours, but the team is developing a faster imaging system to accelerate the process.

What Could This Mean for Patients?

The long-term vision is ambitious. One of the senior researchers on the project imagines a future where doctors could use an endoscope equipped with this technology to look inside a patient's body and identify cellular senescence in real time. This could help diagnose age-related disorders earlier and guide treatment decisions.

"You can imagine that one day we may develop an endoscope that can look inside your body and identify cellular senescence," noted Jeon Woong Kang, an MIT research scientist and senior author of the study.

Jeon Woong Kang, MIT Research Scientist

The research is part of a National Institutes of Health initiative called the Cellular Senescence Network, which aims to develop therapies that could combat the tissue-damaging effects of senescent cells. The work was published today in Nature Aging and was funded by the National Institutes of Health and Massachusetts General Hospital.

While the current study focused on mouse tissue, the MIT team is now working to adapt the method for human tissue. If successful, this AI-powered approach could transform how doctors diagnose and eventually treat the cellular aging process itself, opening new possibilities for extending healthy lifespan and preventing age-related diseases.