β3 integrin in cardiac fibroblast is critical for extracellular matrix accumulation during pressure overload hypertrophy in mouse

Abstract

The adhesion receptor β3 integrin regulates diverse cellular functions in various tissues. As β3 integrin has been implicated in extracellular matrix (ECM) remodeling, we sought to explore the role of β3 integrin in cardiac fibrosis by using wild type (WT) and β3 integrin null (β3-/-) mice for in vivo pressure overload (PO) and in vitro primary cardiac fibroblast phenotypic studies. Compared to WT mice, β3-/- mice upon pressure overload hypertrophy for 4 wk by transverse aortic constriction (TAC) showed a substantially reduced accumulation of interstitial fibronectin and collagen. Moreover, pressure overloaded LV from β3-/- mice exhibited reduced levels of both fibroblast proliferation and fibroblast-specific protein-1 (FSP1) expression in early time points of PO. To test if the observed impairment of ECM accumulation in β3-/- mice was due to compromised cardiac fibroblast function, we analyzed primary cardiac fibroblasts from WT and β3-/- mice for adhesion to ECM proteins, cell spreading, proliferation, and migration in response to platelet derived growth factor-BB (PDGF, a growth factor known to promote fibrosis) stimulation. Our results showed that β3-/- cardiac fibroblasts exhibited a significant reduction in cell-matrix adhesion, cell spreading, proliferation and migration. In addition, the activation of PDGF receptor associated tyrosine kinase and non-receptor tyrosine kinase Pyk2, upon PDGF stimulation were impaired in β3-/- cells. Adenoviral expression of a dominant negative form of Pyk2 (Y402F) resulted in reduced accumulation of fibronectin. These results indicate that β3 integrin-mediated Pyk2 signaling in cardiac fibroblasts plays a critical role in PO-induced cardiac fibrosis.

Figure 1. Reduction of pressure overload induced fibronectin and collagen accumulation in β3−/− mice.

(A) WT and β3−/− mice underwent pressure overload by TAC for 72 h, 1 wk and 4 wk and LV sections were stained for fibronectin using anti-fibronectin antibody (green). LV samples from Sham operated mice served as controls. Nuclei are shown in blue (DAPI). Results were confirmed in two additional mice samples. Scale bar, 10 µm. (B) Triton-X-100 soluble fractions from the pressure-overloaded LV tissue from WT and β3−/− mice for indicated durations as shown in (A) were processed for Western blotting with fibronectin (FN) antibody. Normalized protein loading is shown with GAPDH immunoblot. This experiment was confirmed with a minimum three sets of samples. (C) LV tissue sections from WT and β3−/− mice with and without TAC for 4 wk were stained for collagen with picrosirius red stain. (D) Collagen volume fraction was calculated in photomicrographs using Sigma Scan Pro-5. Bar shows 10 µm. The numbers of mice were: WT control (n = 7); WT TAC (n = 6); β3−/− Control (n = 6); β3−/− TAC (n = 6). Individual and grouped data are presented in the graph. * p<0.05 vs. WT control; @ p<0.05 vs. WT 72 h TAC.

Figure 2. Loss of pressure overload-induced proliferation of cardiac fibroblasts in β3−/− mice.

(A) LV tissue sections from WT and β3−/− Sham (Cont) and 72 h TAC were stained for vimentin (green), Ki67 (red), and nuclei (blue). The merged image of all three stains is also shown. Scale bar, 20 µm. (B) Quantification of the number of Ki67 positive cells in 3–5 sections from three independent mice is shown in the graph. * p<0.05 vs. WT control; @ p<0.05 vs. WT 72 h TAC.

Figure 3. Loss of fibroblast specific protein-1 (FSP-1) expression in pressure overloaded myocardium of β3−/− mice.

(A) WT and β3−/− mice underwent Sham (Cont) or TAC pressure overload for 72 h, or for 4 wk. Tissue sections were stained for FSP-1 (green) and vinculin (red) with specific antibodies. (Nuclei stained with DAPI, blue). The merged images of all three stains are shown. Scale bar, 10 µm. (B) Western blot analysis for Triton X-100 soluble LV tissue samples from mice groups as shown in (A) above were analyzed for the presence of FSP1 (top) and GAPDH (bottom) with specific antibodies.

Figure 4. Phenotypic characteristics of isolated cardiac fibroblasts from β3−/− mice.

(A) WT and β3−/− cardiac fibroblasts were subjected to ³H-thymidine incorporation assay as described in Materials and Methods. Fold-induction in PDGF stimulated ³H-thymidine incorporation was calculated between WT control and WT+PDGF, and β3−/− control and β3−/− + PDGF. The difference in fold induction of cell proliferation between WT and β3−/− obtained from 5 independent experiments is presented as average ± SEM. @ p<0.005 vs. WT. (B) For comparing the contribution of β3 and β1 integrins using function blocking antibodies, WT cardiac fibroblasts were plated on glass coverslips in DMEM containing 10% FBS. After overnight incubation, the cells were incubated with indicated function blocking or control antibodies (50 µg/mL) for 36 hours. After the incubation period, the cells were fixed and stained for Ki67 and nucleus. The number of Ki67 positive nuclei was counted from 10 random fields of confocal images under 20X objective and the average ± SEM presented as a graph. # p<0.05 vs. control; @ p<0.05 vs. control and β1 Ab. (C) For the cell migration assay, WT and β3−/− cells were plated in 96 well format Oris TM plates (Platypus Technology) and allowed to grow overnight. Stoppers were then removed and the cells were stimulated to migrate by adding PDGF (10 ng/mL). After 28 h, the cells were fixed with 4% paraformaldehyde and stained for actin (phalloidin-Alexa Fluor 568; red) and then the entire well was imaged by fluorescent microscopy at 10X. (D) Quantification from the migration assay is depicted in the graph. *p<0.05 vs. WT control; @ p<0.05 vs. WT-PDGF. (E) For Transwell cell invasion assay, WT and β3−/− cardiac fibroblasts were seeded onto Transwell inserts in 0.1% FBS medium in the presence of mitomycin C (5 µg/ml). Cells were incubated for 16 h at 37°C in the presence of PDGF (10 ng/ml) at the bottom chamber and fixed with 1% paraformaldehyde and stained with DAPI. The number of nuclei on the bottom of the Transwell inserts was counted using a 10X objective. At least 10 random fields were recorded to count the total number of nuclei in each case and expressed as a mean (± SEM). @ p<0.05 vs. WT. (F) For the adhesion experiments, WT and β3−/− cardiac fibroblasts cultured in complete medium were washed and trypsinized. The suspended cells in complete medium were re-adhered on fibronectin-coated plates for 10 min, 30 min and 2 h. The cells were then fixed and stained for actin (red) and vinculin (green). Lower magnification photomicrograph is shown in (Top panel) and the higher magnification photomicrograph is shown in (Bottom panel). Scale bar, 10 µm. (G) For cell spreading measurements, cardiac fibroblasts from WT and β3−/− mice were trypsinized, plated on fibronectin-coated plates for 10 min, 30 min and 2 h. After washing away the non-adherent cells, the adhered cells were fixed and stained for actin (Phalloidin-Alexa Fluor 568) and vinculin. The cells were imaged using Olympus (IX71). Cells that showed vinculin staining that reached out of the actin ring/cell body were considered spread cells. At least 400 cells were measured in each case to calculate the % of spreading in each experiment. * p<0.05 vs. WT 10 min; ** p<0.05 vs. β3−/−10 min; @ p<0.05 vs. WT 10 min. (H) To analyze altered actin cytoskeletal changes in PDGF-induced cardiac fibroblasts, cells from WT and β3−/− grown on coverslips were serum starved for 16 h and then stimulated with PDGF (10 ng/mL) for indicated durations and fixed with 1% paraformaldehyde. The cells were then stained for actin (Grey) and nucleus (DAPI, blue) and imaged under confocal microscopy using 60X oil objective. Scale bar, 10 µm.

Figure 5. Loss of PDGF-induced activation of tyrosine kinases in isolated primary cardiac fibroblasts from β3−/− mice.

(A) Equal numbers of WT and β3−/− cardiac fibroblasts were allowed to adhere for 2 h in 35 mm culture dishes. Cells were stimulated with PDGF (10 ng/mL) for the indicated durations following overnight serum starvation. Triton-X-100 soluble cell extracts were analyzed by Western blotting using indicated phospho-specific antibodies. Individual total proteins are shown at the bottom. Blots are representative of data from three independent experiments. (B) The ratio of phosphorylated to total proteins in the case of PDGFR and Pyk2 are shown as average ± SEM. # p<0.05 vs. control; @ p<0.05 vs. WT 5 min. (C) Cell lysates from CFb as shown in (A) above were immunoblotted with indicated phospho-specific MAPK pathway members. Individual total proteins are shown at the bottom. Blots are representative of data from three independent experiments. (D) WT and β3−/− cardiac fibroblasts were serum starved for 16 h and then treated with PDGF (10 ng/mL) for indicated durations and fixed with 1% paraformaldehyde. The cells were then stained with phospho-PDGFR (α-Y849/β-Y857) (Red) and phosphor-Pyk2 (Y402) (Grey) antibodies and imaged under confocal microscopy using 60X oil objective. Scale bar, 10 µm. (E) WT cardiac fibroblasts were plated on glass coverslips in DMEM containing 10% FBS. After overnight incubation, the cells were either uninfected or infected with β-gal, Y402F Pyk2 (402 Pyk2) or β3-integrin (β3 Intg) at an MOI of 50. After 36 h of infection, the cells were fixed and stained for fibronectin (Green) and actin (Grey) and subjected to confocal microscopy as shown under (D). Representative images were from at least three independent experiments. Scale bar, 20 µm.