Common genetic variants identified by genome-wide association studies (GWAS) account for a substantial portion of heritable risk for infectious, autoimmune and immune-mediated diseases. Translating these associations into cellular mechanisms and therapeutic targets requires mapping how disease-risk variants influence gene expression in the relevant cell types. Prior eQTL studies have largely focused on blood immune cells or on bulk tissues, which can miss regulatory effects that are cell type-specific, particularly those operating in tissue-resident immune populations at barrier sites such as the lung.
The authors reasoned that interrogating gene regulation in purified, tissue-derived immune cell populations would improve resolution for linking genetic risk to functional outcomes. The human lung is a particularly suitable tissue for this purpose because (1) unaffected surgical lung tissue can be obtained from living individuals undergoing resection for early-stage cancer; (2) lung immunity is central to respiratory infection; (3) many immune-mediated diseases target the lung (for example, asthma, COPD and sarcoidosis); (4) lung involvement occurs in systemic autoimmune diseases such as rheumatoid arthritis and systemic lupus; and (5) tissue-resident immune cells across organs share a core cis-regulatory landscape and transcriptional program, allowing insights to be extrapolated beyond the lung.
To enable high-resolution mapping, the study generated a large-scale single-cell transcriptomic dataset comprising 1,150,336 immune cells isolated from surgically resected lung tissue of 120 living individuals (a mean of 9,255 cells per individual). The experimental design included fluorescence-activated cell sorting (FACS) enrichment of the five major immune cell populations from each lung sample to ensure broad representation of tissue immune cell types across individuals.
Unbiased clustering of the FACS-enriched single-cell transcriptomes identified 29 transcriptionally distinct immune cell subsets among CD45+ tissue-resident immune cells in the lung. These subsets encompassed diverse T cell, B cell, natural killer (NK) cell, dendritic cell and myeloid populations, including tissue-resident-memory T (TRM) and B resident-memory (BRM) phenotypes described in tissue immunity studies. The relative proportions of these 29 subsets varied substantially between individuals, and the dataset was filtered and quality-controlled to support downstream eQTL analysis.
Using the single-cell transcriptional atlas as the basis, the researchers performed expression quantitative trait locus (eQTL) mapping across the 29 defined lung immune cell subsets. This single-cell eQTL strategy was designed to detect genetic effects on gene expression that are restricted to, or enriched in, tissue-resident immune cell types—effects that are difficult to detect in blood-focused or bulk-tissue eQTL datasets.
By profiling purified populations from lung tissue, the approach addresses problems inherent to prior studies: the underrepresentation of tissue-resident immune populations in blood and the loss of cell-type resolution in bulk tissue analyses. The authors emphasize that genetic regulation of gene expression can be highly cell type-specific, making targeted eQTL studies in disease-relevant tissue-resident cell types necessary to map variant-to-function relationships accurately.
The eQTLs identified in each lung immune cell subset were subjected to colocalization analyses with GWAS signals from a range of traits and diseases affecting the lung, as well as infectious and autoimmune diseases. These colocalization analyses aim to determine whether the same genetic variants underlie both changes in gene expression in a specific cell type and disease association signals detected by GWAS.
Results of the colocalization analyses implicated disease-associated variants and their target genes as functioning in a single or a restricted group of immune cell types within the lung. This pattern supports the view that many genetic risk signals act through discrete tissue-resident immune populations rather than broadly across immune cells in blood or whole tissues.
Among the colocalizations, several genes showed significant overlap between lung immune cell eQTLs and GWAS signals from multiple systemic and organ-restricted autoimmune diseases. The authors highlight ZFP57 as an example of a gene with significant colocalization across autoimmune disease signals and lung immune cell eQTLs, suggesting that shared genetic mechanisms mediated by tissue-resident immune cells contribute to risk across distinct autoimmune conditions.
The observation that disease-associated variants from a wide range of autoimmune diseases impact gene expression in lung tissue-resident immune cell types underscores the relevance of tissue immunity in systemic and organ-specific autoimmunity and provides candidate target genes for further functional follow-up.
The study provides a resource linking GWAS risk variants to gene expression changes in tissue-resident immune cells of the human lung. By generating a large single-cell atlas from FACS-enriched lung immune populations and performing eQTL and colocalization analyses across 29 cell subsets, the authors demonstrate that many disease-associated variants exert their effects in a limited set of tissue immune cell types. Several genes implicated by colocalization, including ZFP57, point to shared genetic mechanisms across autoimmune diseases.
The authors make the data accessible as a community resource (referenced at ), facilitating further studies to validate causal genes and cell types and to explore translational opportunities for diagnostics, prognostics and therapeutic targeting. Details of specific statistical thresholds, full colocalization results, and downstream functional validation experiments were reported in the article supplementary materials and the public dataset.