Plasminogen Activator Inhibitor-1 Protects Mice Against Cardiac Fibrosis by Inhibiting Urokinase-type Plasminogen Activator-mediated Plasminogen Activation

Abstract

Plasminogen activator inhibitor-1 (PAI-1) is known to protect mice against cardiac fibrosis. It has been speculated that PAI-1 may regulate cardiac fibrosis by inactivating urokinase-type plasminogen activator (uPA) and ultimately plasmin (Pm) generation. However, the in vivo role of PAI-1 in inactivating uPA and limiting the generation of Pm during cardiac fibrosis remains to be established. The objective of this study was to determine if the cardioprotective effect of PAI-1 is mediated through its ability to directly regulate urokinase -mediated activation of plasminogen (Pg). An Angiotensin II (AngII)-aldosterone (Ald) infusion mouse model of hypertension was utilised in this study. Four weeks after AngII-Ald infusion, PAI-1-deficient (PAI-1^-/-) mice developed severe cardiac fibrosis. However, a marked reduction in cardiac fibrosis was observed in PAI-1^-/-/uPA^-/- double knockout mice that was associated with reduced inflammation, lower expression levels of TGF-β and proteases associated with tissue remodeling, and diminished Smad2 signaling. Moreover, total ablation of cardiac fibrosis was observed in PAI-1^-/- mice that express inactive plasmin (Pm) but normal levels of zymogen Pg (PAI-1^-/-/Pg^S743A/S743A). Our findings indicate that PAI-1 protects mice from hypertension-induced cardiac fibrosis by inhibiting the generation of active Pm.

Figure 1

PAI-1^−/−/uPA^−/− mice display reduced AngII-Ald-induced cardiac fibrosis. (a) Representative photomicrographs of Masson trichrome (collagen) stain (magnification: 200x). Extensive interstitial collagen deposition (blue staining) is indicated by black arrows. (b) Quantitative data showing the levels of collagen deposition in mouse hearts (n = 4–7 mice per group). (c) qRT-PCR analysis of cardiac Col1a1 gene expression (n = 4 mice per group).

Figure 2

PAI-1^−/−/uPA^−/− mice show decreased spontaneous bleeding and vascular permeability in AngII-Ald-infused mice. (a) Representative photomicrographs of heart sections from WT, PAI-1^−/−, uPA^−/−, and PAI-1^−/−/uPA^−/− mice stained with Prussian blue for hemosiderin analysis at 4 wk of AngII-Ald infusion (magnification: 200x). (b) Histogram shows the quantification of areas of hemosiderin deposition in cardiac tissue (n = 4–7 mice per group). (c) Vascular permeability in mouse hearts at 1 wk of AngII-Ald infusion was detected by an Evans Blue vascular leakage assay (n = 3–6 mice per group).

Figure 3

Inflammatory cytokine expression is attenuated in AngII-Ald-infused PAI-1^−/−/uPA^−/− mice. qRT-PCR (a) and Western blot (b) analysis of ICAM-1 expression in heart tissues at 4 wk of AngII-Ald infusion. In Western blot, densitometric analysis of relative ICAM-1 expression after normalisation with GAPDH is shown in the lower panel. Western blot images are cropped for concise presentation. All uncut western blot images are provided in Supplementary Figure S6. qRT-PCR (c) and ELISA assays (d) of KC expression in heart tissues at 4 wk of AngII-Ald infusion were performed as described.

Figure 4

Leukocyte infiltration is altered in AngII-Ald-infused PAI-1^−/−/uPA^−/− mice. (a) Representative photomicrographs from cardiac sections of AngII-Ald-infused (4 wk) mice immunostained with anti-CD45 and anti-Mac-3 antibody. CD45 positive leukocytes and Mac-3 positive macrophages are identified in brown. Magnification: 200x. (b and c) Quantification of CD45 and Mac-3 positive staining of hearts (n = 4–7 mice per group).

Figure 5

Cardiac ECM remodeling and Smad2 signaling are altered in AngII-Ald-infused PAI-1^−/−/uPA^−/− mice. qRT-PCR analysis of mRNA transcripts of TGF-β2 (a), MMP-2 (d), and uPA (f), proteins associated with cardiac tissue remodeling and fibrosis, at 4 wk of AngII-Ald infusion (n = 4 mice per group). ELISA assays of total (b) and active TGF-β2 (c), MMP-2 (e), and active uPA (g) were performed as described. Immunohistochemistry (i) for pSmad2 in AngII-Ald-induced mice hearts and the quantitative data (h) showing the percent of cells with pSmad2⁺ nuclei (n = 3–4 mice per group).

Figure 6

Loss of Pm activity results in reduced cardiac fibrosis in AngII-Ald infused mice. qRT-PCR (a) of transcripts of Col1a1, ICAM-1, TGF-β2, and MMP-2, Western blot analysis (b) of ICAM-1 expression, ELISA assays (c) of total and active TGF-β2 and MMP-2, and histological analyses (d) of Masson trichrome (collagen), hemosiderin deposition, and leukocyte infiltration (CD45 and Mac-3) at 4 wk of AngII-Ald infusion were performed as described (n = 3–4 mice per group). Western blot images are cropped for concise presentation. The uncut images are provided in Supplementary Figure S6.

Figure 7

Schematic diagram showing functional roles of PAI-1 and uPA during AngII-Ald-induced cardiac fibrosis. AngII-Ald infusion in mice induces the expression of PAI-1 and TGF-β in the heart. Inhibition of TGF-β activation by PAI-1 inhibiting tPA/uPA generation of Pm suppresses TGF-β signaling in hearts. Binding of active TGF-β to its receptors (TGFβR) facilitates the recruitment of Smad2/3 to TGFβR leading to the phosphorylation of Smad2/3. Phosphorylated Smad2/3 forms a complex with Smad4 and translocates to the nucleus where it regulates transcription of target genes involved in cardiac tissue remodeling and fibrosis. Increased MMP activity in hearts further enhances the activation of TGF-β.