αv integrins on mesenchymal cells regulate skeletal and cardiac muscle fibrosis

Abstract

Mesenchymal cells expressing platelet-derived growth factor receptor beta (PDGFRβ) are known to be important in fibrosis of organs such as the liver and kidney. Here we show that PDGFRβ+ cells contribute to skeletal muscle and cardiac fibrosis via a mechanism that depends on αv integrins. Mice in which αv integrin is depleted in PDGFRβ+ cells are protected from cardiotoxin and laceration-induced skeletal muscle fibrosis and angiotensin II-induced cardiac fibrosis. In addition, a small-molecule inhibitor of αv integrins attenuates fibrosis, even when pre-established, in both skeletal and cardiac muscle, and improves skeletal muscle function. αv integrin blockade also reduces TGFβ activation in primary human skeletal muscle and cardiac PDGFRβ+ cells, suggesting that αv integrin inhibitors may be effective for the treatment and prevention of a broad range of muscle fibroses.

Conflict of interest statement

P.G.R. and D.W.G. hold equity in Antegrin Therapeutics, LLC. The remaining authors declare no competing financial interests.

Figures

Fig. 1

Pdgfrb-Cre effectively targets recombination in quiescent PDGFRβ+ cells and activated myofibroblasts. a Immunofluorescence micrographs of skeletal muscle from mTmG;Pdgfrb-Cre mice co-stained with anti-PDGFRβ antibody. Scale bar 10 μm. b, c Quantification of GFP reporting and PDGFRβ antibody colocalisation in skeletal muscle (n = 7). d Flow cytometric analysis of PDGFRβ expression in GFP+ cells sorted from mTmG:Pdgfrb-Cre mouse skeletal muscle (n = 3). e Immunofluorescence micrographs of skeletal muscle sections harvested from mTmG;PDGFR β-Cre reporter mice 4, 8 and 21 days after control (PBS) or CTX intramuscular injection. Scale bars 30 μm. f Percentage field coverage of GFP+ cells at 4, 8, 21 and 60 days after control (PBS) or CTX intramuscular injection (n = 4). g Quantitation of GFP+ nuclei at 4, 8, 21 and 60 days after control (PBS) or CTX intramuscular injection (n = 4). h Gene expression profile of freshly sorted GFP+ cells from skeletal muscle at day 10 following control (PBS) or CTX intramuscular injection (n = 4). i Immunofluorescence staining of a typical GFP+ cell sorted from uninjured skeletal muscle of mTmG;Pdgfrb-Cre reporter mice and plated on tissue culture plastic for 5 days. Scale bar 50 μm. j qPCR analysis of the genes encoding PDGFRβ, αSMA, Col1A1 and TIMP1 in freshly sorted GFP+ cells from mTmG;Pdgfrb-Cre reporter mice and from GFP+ cells cultured for 7 and 14 days (n = 4). Data are mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (Student’s t-test)

Fig. 2

αv integrins on PDGFRβ+ cells regulate skeletal muscle fibrosis via TGFβ activation. a αv expression on PDGFRβ+ cells from control and Itgavflox/flox;Pdgfrb-Cre (αv Cre) mice using flow cytometry (n = 3). b αv Cre and control mice received intramuscular CTX with muscles harvested at day 21. c Representative images of picrosirius red stained sections 21 days after CTX. Upper panels: scale bar 25 μm, Lower panels: 400 μm. d Digital image analysis quantification of picrosirius red staining in control and αv Cre mice day 21 post CTX or control injection (n = 7). e qPCR of Col1a1 in control and αv Cre PDGFRβ+ cells culture-activated for 5 days (n = 7). f qPCR of TGFβ1 in control and αv Cre PDGFRβ+ cells culture-activated for 5 days (n = 7). g qPCR of Col1a1 in GFP+ cells from mTmG;Pdgfrb-Cre mice cultured with CWHM 12 and control for 5 days (n = 3). h TGFβ activation by control- or CWHM 12-treated GFP+ cells from mTmG;Pdgfrb-Cre mice (n = 6). TGFβ activation was assessed alone, with TGFβ–blocking antibody (Anti–TGFβ), and with recombinant human TGFβ1 (rhTGFβ). i Representative images of picrosirius red stained sections from mice treated with CWHM12 or control from the time of CTX. Mice received rhTGFβ1 (lower panels) to the region of injured muscle every 72 h up to 21 days. j Digital image analysis quantitation of picrosirius red stained sections from CWHM12 or control treated mice that received rhTGFβ1 or PBS to the region of injured muscle (n = 5). k Representative images of phospho-SMAD 3 stained sections from control or CWHM12-treated mice. Scale bar 30 μm. l Quantitation of phospho-SMAD 3 stained sections from mice treated with control or CWHM12 from the time of CTX (n = 10). m Force required for passive elongation of muscles 21 days following intramuscular CTX in mice treated with CWHM 12 or control (n = 9). n Laceration-induced skeletal muscle fibrosis model dosing regime. o Representative images of picrosirius red stained sections from control and CWHM 12 treated mice. Scale bar 25 μm. p Digital image analysis quantitation of picrosirius red staining in CWHM12 and control treated mice (n = 10). Data are expressed as mean ± SEM. *P < 0.05, **P < 0.01 (Student’s t-test)

Fig. 3

Selective depletion of αv integrins on PDGFRβ+ cells is protective in cardiac fibrosis. a Immunofluorescence micrographs of heart muscle from mTmG;Pdgfrb-Cre mice co-stained with anti-PDGFRβ (scale bar 5 μm), anti-CD31 (scale bar 5 μm) and anti-myosin antibodies (scale bar 30 μm). b, c Quantification of GFP reporting and PDGFRβ antibody staining colocalisation in cardiac muscle (n = 6). d Overview of the cardiac fibrosis model. Mice were treated with 200 ng kg−1 min−1 AngII or vehicle control for 14 days prior to harvest and analysis of tissues. e Immunofluorescence micrographs of heart sections from control or AngII-treated mTmG;Pdgfrb-Cre mice. Scale bars 30 μm. f Percentage field coverage of PDGFRβ+ cells at day 14 in vehicle control or AngII-treated mTmG;Pdgfrb-Cre mice (n = 3). g Quantitation of GFP+ nuclei at day 14 in vehicle control or AngII-treated mTmG;Pdgfrb-Cre mice (n = 4) h qPCR of Col1a1 in cardiac PDGFRβ+ cells sorted from AngII injured and control hearts (n = 5) i qPCR of Col1a1 in cardiac PDGFRβ+ cells activated in culture for 5 days (n = 3). j Blood pressure response of control and Itgavflox/flox;Pdgfrb-Cre (αv Cre) mice to AngII treatment or vehicle (n = 6). k Picrosirius red staining of cardiac tissue 14 days after commencement of AngII treatment in control and Itgavflox/flox;Pdgfrb-Cre mice. Scale bars 1 mm in whole heart sections, 70 μm for magnified fields. l Digital image analysis of collagen staining (n = 6 mice per group). Data are expressed as mean ± SEM. *P < 0.05, **P < 0.01 (Student’s t-test)

Fig. 4

Blockade of αv integrins by a small molecule (CWHM 12) attenuates established skeletal muscle and cardiac fibrosis, and αv integrins represent a tractable therapeutic target in human muscle fibrosis. a Dosing regime in the therapeutic skeletal muscle fibrosis model. Alzet osmotic minipumps containing CWHM 12 or CWHM 96 (control) were inserted ten days after intramuscular CTX injection. Tissues were harvested at day 21 following CTX injection. b Representative images of picrosirius red stained sections from control- and CWHM 12-treated mice. Scale bar 25 μm. c Digital image analysis quantification of collagen (picrosirius red staining) (n = 10). d Dosing regime in the therapeutic cardiac fibrosis model. Seven days following commencement of AngII treatment, Alzet osmotic minipumps containing CWHM 12 or CWHM 96 (control) were inserted. Tissues were harvested at day 14 after commencement of AngII treatment. e Representative images of picrosirius red stained sections from control- and CWHM 12-treated mice. Scale bars 1 mm in whole heart sections, 70 μm for magnified fields. f Digital image analysis quantification of collagen (picrosirius red staining) (n = 11). g, h Flow cytometric analysis of PDGFRβ and αv integrin expression on PDGFRβ+ cells from human skeletal muscle (g) and heart (h). i, j TGFβ activation by control- or CWHM 12-treated PDGRβ+ cells isolated from human skeletal muscle (i) and cardiac muscle (j) (n = 4). TGFβ activation was assessed alone, in the presence of TGFβ-blocking antibody (clone 1D11, 40 μg ml−1) (anti-TGFβ), and in the presence of recombinant human TGFβ1 (rhTGFβ) (300 pg ml−1). Data are expressed as mean ± SEM. *P < 0.05, **P < 0.01 (Student’s t-test)