TNF-α represses fibroblast to myofibroblast transition through the histone methyltransferase Setdb2
Tyler M. Bauer, Kevin D. Mangum, Samuel D. Buckley, James Shadiow, Amrita D. Joshi, Christopher O. Audu, Jadie Y. Moon, Lindsey D. Hughes, Rachel Bogel, Lam C. Tsoi, Qinmennge Li, He Zhang, Steven Kunkel, Johann E. Gudjonsson, Frank M. Davis, Katherine A. Gallagher
Tyler M. Bauer, Kevin D. Mangum, Samuel D. Buckley, James Shadiow, Amrita D. Joshi, Christopher O. Audu, Jadie Y. Moon, Lindsey D. Hughes, Rachel Bogel, Lam C. Tsoi, Qinmennge Li, He Zhang, Steven Kunkel, Johann E. Gudjonsson, Frank M. Davis, Katherine A. Gallagher
View: Text | PDF
Research Article Dermatology Immunology

TNF-α represses fibroblast to myofibroblast transition through the histone methyltransferase Setdb2

Abstract

Fibroblast to myofibroblast transition is a critical event required for effective tissue repair. In pathologic wound repair processes, such as type 2 diabetes (T2D), fibroblast to myofibroblast transition is impaired. The exact factors that control this transition in wounds are unclear. Here, using human tissue and murine transgenic models, we show that the histone methyltransferase SETDB2 is elevated in diabetic wound fibroblasts and TNF-α represses fibroblast to myofibroblast transition via Setdb2. We identified that TNF-α increases Setdb2 in fibroblasts via a JAK1,3/STAT3 signaling pathway, where pharmacologic or genetic manipulation of this pathway altered Setdb2 in fibroblasts. We also found that fibroblasts treated with pro-inflammatory macrophage supernatants displayed increased Setdb2 and downregulated myofibroblast genes; inhibition of the TNF-α receptor reduced the upregulation of Setdb2. In diabetes, we showed that TNF-α signaling was increased in wound fibroblasts, which functions to increase Setdb2 expression and represses fibroblast to myofibroblast transition. Fibroblast-specific knockdown of SETDB2 and therapeutic inhibition of JAK1,3/STAT3 improved diabetic wound repair, where wound fibroblasts expressed increased myofibroblast genes. This study is the first to our knowledge to identify an epigenetic mechanism for reduced fibroblast to myofibroblast transition in diabetic wounds. Therapeutic targeting of the TNF-α/STAT3/SETDB2 axis in wound fibroblasts may improve diabetic wound healing.

Authors

Tyler M. Bauer, Kevin D. Mangum, Samuel D. Buckley, James Shadiow, Amrita D. Joshi, Christopher O. Audu, Jadie Y. Moon, Lindsey D. Hughes, Rachel Bogel, Lam C. Tsoi, Qinmennge Li, He Zhang, Steven Kunkel, Johann E. Gudjonsson, Frank M. Davis, Katherine A. Gallagher

×

Figure 1

TNF-α represses myofibroblast gene expression and increases Setdb2 expression in wound fibroblasts.

Options: View larger image (or click on image) Download as PowerPoint
TNF-α represses myofibroblast gene expression and increases Setdb2 expre...
(A) Flow gating on wound fibroblasts that were isolated from wounds 7 days after injury in C57BL/6 mice utilizing magnetic separation for Tie2–, Ter119–, CD45–, CD31–, and EPCAM– cells. Cells were plated and passaged 3 times prior to flow cytometry (n = 5 mice). (B) Acta2, Tagln, Myl9, and Cald1 gene expression in wound fibroblasts isolated 7 days after wounding following TNF-α stimulation for 6 hours (25 ng/μL) (n = 4 mice/group, pooled and run in triplicate). (C) Unbiased epigenetic array comparing expression of chromatin modifying enzymes in dermal fibroblasts following 6 hours of TNF-α (25 ng/μL) stimulation with control. Chromatin modifying enzymes with less than 1.5- to 0.5-fold regulation were excluded. Fold regulation was normalized to arithmetic mean of 5 housekeeping genes (n = 4 mice/group, run in singlicate). (D) Setdb2 expression in wound fibroblasts 7 days after wounding following TNF-α (25 ng/μL) stimulation for 6 hours (n = 4 mice/group, run in triplicate). (E) Cluster analysis uniform manifold approximation and projection (UMAP) of single-cell RNA sequencing from human T2D and non-T2D wounds showed 10 unique cell clusters (representative). Dot plots detailing ACTA2, TAGLN, CALD1, and MYL9 gene expression between human wound fibroblasts expressing high amounts of Setdb2 compared with those expressing low levels of Setdb2 (n = 10). (F) Cluster analysis UMAP of single-cell RNA sequencing from human T2D and non-T2D wounds showed 8 unique fibroblast cell clusters (representative) (n = 10). (G) Dot plots detailing SETDB2, ACTA2, TAGLN, CALD1, and MYL9 gene expression between human wound fibroblast subtypes (n = 10). *P < 0.05, ***P < 0.001, ****P < 0.0001. Data are presented as the mean ± SEM. Data were first analyzed for normal distribution, and if data passed the normality test, 2-tailed Student’s t test was used. Representative figures are displayed for panel B and D, which were repeated 3 times independently.