Poised cell circuits in human skin are activated in disease

Gary Reynolds, Péter Végh, James Fletcher, Elizabeth Poyner, Emily Stephenson, Issac Goh, Rachel A. Botting, Ni Huang, Bayanne Olabi, Anna Dubois, David Dixon, Kile Green, Daniel Maunder, Justin Engelbert, Mirjana Efremova, Krzysztof Polański, Laura Jardine, Claire Jones, Thomas Ness, Dave Horsfall, Jim McGrath, Christopher D. Carey, Dorin-Mirel Popescu, Simone Webb, Xiaonong Wang, Ben Sayer, Jong‐Eun Park, Victor Augusti Negri, Daria Belokhvostova, Magnus Lynch, David A. McDonald, Andrew Filby, Tzachi Hagai, Kerstin B. Meyer, Akhtar Husain, Jonathan Coxhead, Roser Vento‐Tormo, Sam Behjati, Steven Lisgo, Alexandra–Chloé Villani, Jaume Bacardit, Philip H. Jones, Edel A. O’Toole, Graham S. Ogg, Neil Rajan, Nick J. Reynolds, Sarah A. Teichmann, Fiona M. Watt, Muzlifah HaniffabioRxiv 2020 · DOI

TL;DR

This study presents a comprehensive single-cell atlas of human skin, revealing intricate cellular circuits that are dynamically reconfigured in inflammatory diseases like atopic dermatitis and psoriasis. By profiling nearly half a million cells, the researchers uncovered novel cell states and interactions that are crucial for understanding skin health and disease.

What Problem Does This Paper Address?

The human skin is a complex barrier, and understanding its cellular composition and how it changes in disease is critical for developing effective treatments. While previous studies have looked at specific cell types or limited regions of the skin, a comprehensive, high-resolution map of the entire skin ecosystem, from development through adulthood and into disease states, has been lacking. This paper aims to fill that gap by dissecting the cellular landscape of healthy and diseased human skin at an unprecedented scale, focusing on the dynamic interplay between various cell types.

Methods

The researchers employed a multi-pronged approach:

  • Single-cell RNA sequencing (scRNA-seq): This was the primary technique, used to profile the gene expression of individual cells. They utilized both droplet-based (10x Genomics) and plate-based (Smart-seq2) methods to capture a broad range of cell types and gene expression levels.
  • Mass Cytometry (CyTOF): This technique was used for protein-level validation of cell types identified through scRNA-seq, providing a complementary view of cellular populations.
  • Computational Analysis: A suite of sophisticated bioinformatics tools was used for data integration, dimensionality reduction (UMAP), clustering (Leiden), differential gene expression analysis, trajectory inference (PAGA), and cell-cell communication analysis (CellPhoneDB).
  • Immunofluorescence Staining: This was used to visualize the spatial distribution of specific cell markers within the skin tissue.

Datasets

The study utilized the following datasets:

  • Healthy adult skin: Samples from mammoplasty surgery.
  • Developing skin: Human embryonic and fetal skin samples (7-10 post-conception weeks).
  • Diseased skin: Biopsies from patients with atopic dermatitis (AD) and psoriasis.
  • Publicly available datasets: Used for comparative analysis, particularly for murine splenic and dermal DCs.

Key Insights and Ideas

  • Dynamic Skin Development: The skin's cellular composition shifts dramatically from embryonic development to adulthood, with a predominance of innate immune cells in developing skin and a greater presence of T cells and migratory dendritic cells in adult skin.
  • Dual Keratinocyte Differentiation: Keratinocytes, the main skin cells, follow two distinct differentiation pathways, one related to lamellar body production (crucial for skin barrier function) and another associated with inflammatory responses.
  • Specialized Endothelial Cells: The study identified a specific type of vascular endothelial cell (VE3) that appears specialized for leukocyte adhesion and trafficking, potentially playing a key role in immune cell recruitment to the skin.
  • Disease-Specific T Cell Circuits: In AD and psoriasis, distinct populations of T cells, characterized by the expression of specific cytokines (IL13/IL22 in AD, IL17A/F in psoriasis), are expanded and localized to specific skin compartments, suggesting disease-specific immune responses.
  • Inflammatory Macrophages: The research identified IL23A-expressing macrophages, termed "inflammatory macrophages," which are implicated in the inflammatory processes of skin diseases.

Key Results

  • ~500,000 single cells profiled: This massive dataset allowed for a highly granular dissection of skin cell types.
  • Identification of 34 distinct cell states: The analysis revealed a rich heterogeneity within the skin's cellular ecosystem.
  • Disease-specific T cell populations: Clonal expansion of IL13/IL22-expressing T cells in AD and IL17A/F-expressing T cells in psoriasis was observed.
  • Keratinocyte differentiation pathways: Two distinct trajectories were identified, with implications for skin barrier function and inflammatory responses.
  • Enriched inflammatory circuits in disease: The study highlighted activated cellular circuits involving vascular endothelial cells, immune cells, and keratinocytes in AD and psoriasis.

Limitations and Future Work

  • Snapshot in time: While the study captures different disease states, it represents a snapshot. Longitudinal studies would be valuable to track the dynamic evolution of these cellular circuits over time.
  • In vitro vs. in vivo: Some findings, particularly regarding cell-cell interactions, are based on computational predictions and would benefit from further validation through in vitro and in vivo experiments.
  • Microbiome influence: The paper acknowledges the potential influence of the microbiome on skin immunity but does not directly investigate it. Future work could explore this interaction.
  • Therapeutic targets: The identification of specific cell types and pathways involved in disease opens avenues for developing targeted therapies, which is a natural next step for future research.

Summary updated

Abstract

Abstract The human skin confers biophysical and immunological protection through a complex cellular network that is established early in development. We profiled ~500,000 single cells using RNA-sequencing from healthy adult and developing skin, and skin from patients with atopic dermatitis and psoriasis. Our findings reveal a predominance of innate lymphoid cells and macrophages in developing skin in contrast to T cells and migratory dendritic cells in adult skin. We demonstrate dual keratinocyte differentiation trajectories and activated cellular circuits comprising vascular endothelial cells mediating immune cell trafficking, disease-specific clonally expanded IL13/IL22 and IL17A/F -expressing lymphocytes, epidermal IL23 -expressing dendritic cells and inflammatory keratinocytes in disease. Our findings provide key insights into the dynamic cellular landscape of human skin in health and disease. One Sentence Summary Single cell atlas of human skin reveals cell circuits which are quantitatively and qualitatively reconfigured in inflammatory skin disease.
BibTeX
@misc{reynolds2020poised,
  author = {Gary Reynolds and Péter Végh and James Fletcher and Elizabeth Poyner and Emily Stephenson and Issac Goh and Rachel A. Botting and Ni Huang and Bayanne Olabi and Anna Dubois and David Dixon and Kile Green and Daniel Maunder and Justin Engelbert and Mirjana Efremova and Krzysztof Polański and Laura Jardine and Claire Jones and Thomas Ness and Dave Horsfall and Jim McGrath and Christopher D. Carey and Dorin-Mirel Popescu and Simone Webb and Xiaonong Wang and Ben Sayer and Jong‐Eun Park and Victor Augusti Negri and Daria Belokhvostova and Magnus Lynch and David A. McDonald and Andrew Filby and Tzachi Hagai and Kerstin B. Meyer and Akhtar Husain and Jonathan Coxhead and Roser Vento‐Tormo and Sam Behjati and Steven Lisgo and Alexandra–Chloé Villani and Jaume Bacardit and Philip H. Jones and Edel A. O’Toole and Graham S. Ogg and Neil Rajan and Nick J. Reynolds and Sarah A. Teichmann and Fiona M. Watt and Muzlifah Haniffa},
  title = {{Poised cell circuits in human skin are activated in disease}},
  year = {2020},
  doi = {10.1101/2020.11.05.369363},
  howpublished = {bioRxiv},
}
Figure