single-cell intermediate AI-generated ✓ machine-checked

Single-cell spatial transcriptomic analysis of human skin anatomy

P. Restrepo, A. Wilder, Aubrey Houser, Harkirat Singh Sandhu, Angie V. Ramirez, M. Grace Hren, Raman Gill, Abiha Kazmi, Larry Chen, Alexandra Nigro, Ichiro Imanishi, Deniz Demircioglu, Dan Hasson, Alan Soto, Stephanie McQuillan, Edgar Gonzalez-Kozlova, Rachel Brody, Benjamin Ungar, Maria Kasper, Ping Lu, Philip J. Torina, Jesse M. Lewin, Sacha Gnjatic, Sai Ma, Andrew L. Ji

Nature Genetics · Mar 23, 2026

AI-generated summary (Gemini), independently checked for faithfulness by a second model (Claude). Automated checking catches most errors, not all — verify anything important against the original.

TL;DR

Researchers created a comprehensive single-cell spatial map of human skin across 15 body sites, revealing how specific cellular neighborhoods maintain skin health and are disrupted in disease.

Problem / question

Despite being the body's largest organ and a major site of disease, the cellular and molecular organization of human skin across different anatomical sites remains poorly understood, limiting our ability to treat site-specific skin conditions.

Methods

The team used a spatial transcriptomics technique called MERFISH to map over 1.2 million cells from 114 healthy skin samples across 15 body sites. They integrated this with existing single-cell RNA sequencing data, used computational tools to identify cell neighborhoods and communication networks, and compared healthy tissue to diseased skin using public spatial data. They also validated key cell signaling pathways using lab-grown human skin samples.

Key findings

The study identified 45 distinct cell types organized into ten recurring multicellular neighborhoods. Cell composition varied predictably by body site, such as immune cells clustering in joint creases and specialized fibroblasts in the soles of the feet. They highlighted a specific blood-vessel-associated (perivascular) neighborhood that acts as an immune hub, maintained by tumor necrosis factor signaling. In diseased skin, this perivascular neighborhood consistently expanded and changed its gene expression, acting as a central site for disease-related inflammation.

Why it matters

Mapping the exact locations and interactions of skin cells provides a blueprint for engineering artificial skin, understanding aging, and developing targeted treatments for inflammatory skin diseases and cancers.

Limitations

The imaging technique could not detect rare immune cells or nerves, and the cell-segmentation software sometimes grouped cells incorrectly. The study also could not account for lifestyle factors like sun exposure, and some gene expression data had to be computationally predicted rather than directly measured.

Takeaway

Human skin relies on highly organized, site-specific cellular neighborhoods to function, and targeting the immune-rich perivascular neighborhood could offer new ways to treat various skin diseases.