Back

Fibroblast-encoded inflammatory memory orchestrates recurrent skin inflammation via NNMT-dependent metabolic remodeling

Zhou, Y.; Li, K.; Xie, Y.; Li, Z.; Yang, Y.; Qiao, P.; Liu, W.; Bai, Y.; Xue, K.; Gudjonsson, J.; Wang, G.; Shao, S.

2026-04-30 immunology
10.64898/2026.04.27.721223 bioRxiv
Show abstract

Chronic skin inflammation frequently recurs at the same anatomical sites after therapy withdrawal, implying stromal cells may encode local inflammatory memory. Here, we identified nicotinamide N-methyltransferase (NNMT) as a central metabolic-epigenetic regulator of fibroblast inflammatory memory enabling psoriasis relapse. Single-cell and spatial transcriptomics revealed that dermal fibroblasts acquire a persistent senescence-associated secretory phenotype (SASP) during inflammation, maintaining pro-inflammatory niche in resolved skin that supports CD 8+CD 103+ tissue-resident memory T cell (Trm) differentiation. Multi-omics profiling demonstrated that NNMT depletes S-adenosylmethionine (SAM), reduces H 3K 27me3 deposition, and permits sustained AP-1 occupancy at SASP gene promoters. Fibroblast-specific NNMT ablation or pharmacologic inhibition suppressed SASP activity, limited Trm accumulation, and prevented both initiation and relapse of skin inflammation in mice. These findings establish NNMT as a stromal regulator linking fibroblast metabolism to durable epigenetic memory and propose its targeting to erase inflammatory memory and achieve long-term remission in psoriasis and related immune-mediated diseases. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=139 SRC="FIGDIR/small/721223v1_ufig1.gif" ALT="Figure 1"> View larger version (53K): org.highwire.dtl.DTLVardef@34b2f4org.highwire.dtl.DTLVardef@1d0bb89org.highwire.dtl.DTLVardef@c8277borg.highwire.dtl.DTLVardef@9825ac_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOGraphic Abstract:C_FLOATNO Proposed mechanism through which NNMT-driven fibroblast SASP facilitates the initiation and recurrence of psoriasis C_FIG

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.