Schmiedel et al. · Nature Immunology · 2026
Tissue-resident immune cells drive genetic risk in autoimmune and lung diseases
Schmiedel Benjamin J., Gonzalez-Colin Cristian, Fajardo-Rosas Vicente, Rocha Job, Simon Hayley, Ottensmeier Johannes, Ramirez-Bernabe Ignacio E., Cano April, De la Cruz Castillo Angel, Marquez-Gomez Elizabeth, Ha Brendan, Greenbaum Jason A., Chudley Lindsey, Cave Judith, Alzetani Aiman, Woo Edwin, Shackcloth Michael, Chee Serena J., Chandra Vivek, Kronenberg Mitchell, Peters Bjoern, Ottensmeier Christian H., Ganesan Anusha-Preethi, Vijayanand Pandurangan
The study
What was asked, and what was found
Schmiedel et al., Nature Immunology, 2026 asked which genes carry the inherited risk of autoimmune and lung disease inside the immune cells that live in the lung, rather than inside the blood cells that are easier to sample. The team built a single-cell transcriptomic dataset of 1,150,336 immune cells from surgically resected lung tissue of 120 living individuals, resolved it into 29 immune cell subsets, mapped expression quantitative trait loci in each subset, and colocalised those with genome-wide association signals from 32 immune-mediated diseases. That yielded 281 colocalised genes, and ZFP57 stood out as one of the top hits for sarcoidosis, with genotype-dependent expression across several lung immune cell types.
Genetics named the gene but said nothing about what it does, and as the authors note, the function of ZFP57 in immune cells had not been explored. To answer that they silenced it with AUM BioTech reagents: a pool of four AUMsilence sdASOs, each against ZFP57 messenger RNA, set against a non-targeting AUMscramble control. Both were used at 2.0 μM in primary human monocyte-derived macrophages grown from the CD14 positive monocytes of six healthy donors, present in the culture from day 0 and kept there for 10 days, with no transfection reagent named anywhere in the Methods. Knockdown was confirmed on day 4 by real-time PCR against the housekeeping gene YWHAZ, with a paired two-tailed t test P value of 0.0005 across the six donors.
On day 10 the macrophages were stimulated with lipopolysaccharide for 6 h and profiled by RNA sequencing. Silencing ZFP57 moved 425 transcripts up and 322 down, and the movement concentrated on the chemokines and cytokines that direct immune cell traffic: CXCL10 and CXCL11 at P = 0.0002, CXCL8 at P = 0.0011, CXCL5 at P = 0.0014, CCL7 at P = 0.0052, CSF2 at P = 0.0053, CCL18 at P = 0.0060, IL27 at P = 0.0061 and CCL2 at P = 0.0089. The authors read that as evidence that ZFP57 shapes what a macrophage does, and so may be one route by which a common variant raises the risk of sarcoidosis and other autoimmune diseases.
Key findings
- A pool of four AUMsilence sdASOs against ZFP57 messenger RNA lowered ZFP57 transcript in primary human monocyte-derived macrophages from six healthy donors, against a non-targeting AUMscramble control run at the same 2.0 μM, with a paired two-tailed t test P value of 0.0005.(Methods, Validation of gene function in immune cells; value in Supplementary Fig. 10b)
- Silencing ZFP57 with AUMsilence sdASOs moved 425 transcripts up and 322 down in lipopolysaccharide-stimulated macrophages, concentrated on chemokines and cytokines that direct immune cell traffic.(Results, Genes associated with systemic autoimmune diseases, page 1949)
- The silencing experiment was the functional test of a gene the authors reached by human genetics, ZFP57 being one of the top genes whose expression in lung immune cells colocalised with sarcoidosis risk signals.(Results, Genes associated with systemic autoimmune diseases, page 1949)
- Before this study nobody had asked what ZFP57 does in an immune cell, so the AUMsilence sdASO knockdown supplied the first functional read on that question.(Results, Genes associated with systemic autoimmune diseases, page 1949)
For research use only. Not for use in diagnostic or therapeutic procedures.