Protease-activated receptor-1 contributes to cardiac remodeling and hypertrophy

Abstract

Background: Protease-activated receptor-1 (PAR-1) is the high-affinity receptor for the coagulation protease thrombin. It is expressed by a variety of cell types in the heart, including cardiomyocytes and cardiac fibroblasts. We have shown that tissue factor (TF) and thrombin contribute to infarct size after cardiac ischemia-reperfusion (I/R) injury. Moreover, in vitro studies have shown that PAR-1 signaling induces hypertrophy of cardiomyocytes and proliferation of cardiac fibroblasts. The purpose of the present study was to investigate the role of PAR-1 in infarction, cardiac remodeling, and hypertrophy after I/R injury. In addition, we analyzed the effect of overexpression of PAR-1 on cardiomyocytes.

Methods and results: We found that PAR-1 deficiency reduced dilation of the left ventricle and reduced impairment of left ventricular function 2 weeks after I/R injury. Activation of ERK1/2 was increased in injured PAR-1(-/-) mice compared with wild-type mice; however, PAR-1 deficiency did not affect infarct size. Cardiomyocyte-specific overexpression of PAR-1 in mice induced eccentric hypertrophy (increased left ventricular dimension and normal left ventricular wall thickness) and dilated cardiomyopathy. Deletion of the TF gene in cardiomyocytes reduced the eccentric hypertrophy in mice overexpressing PAR-1.

Conclusions: Our results demonstrate that PAR-1 contributes to cardiac remodeling and hypertrophy. Moreover, overexpression of PAR-1 on cardiomyocytes induced eccentric hypertrophy. Inhibition of PAR-1 after myocardial infarction may represent a novel therapy to reduce hypertrophy and heart failure in humans.

Figure 1

Role of PAR-1 in infarct size after cardiac I/R injury. Determination of area at risk (AAR) and infarct size in mice subjected to 30 minutes of ischemia and 2 hours of reperfusion. I/R-injured WT mice were treated with either saline (n=8; open bars) or hirudin (n=5; stippled bars). Infarct size was measured in PAR-1^−/− mice (n=14; solid bars) and WT littermates (n=10; open bars). *P<0.01.

Figure 2

Role of PAR-1 in cardiac remodeling after cardiac I/R injury. A, Representative cross sections of hearts from WT and PAR-1^−/− mice subjected to 45 minutes of ischemia and 2 weeks of reperfusion. Sections were stained with hematoxylin and eosin. The normal myocardium is pink, whereas the infarct is white. Infarct size was quantified in WT (n=10) and PAR-1^−/− (n=10) hearts. B, Cross-sectional area of cardiomyocytes in the hearts of WT and PAR-1^−/− mice after I/R injury. Six mice were analyzed for each group, and 30 to 40 cardiomyocytes were measured per mouse. C, Levels of phosphorylated ERK1/2 in hearts of WT and PAR-1^−/− mice before and after I/R injury. Two or 3 mice are shown per group. Data are mean±SD. D, Echocardiographic analysis of LV diameter (LVD) at diastole (d) and systole (s) and LV fractional shortening (FS) before and after myocardial infarction induced by 45 minutes of ischemia and 2 weeks of reperfusion. Results from WT (n=12; open bars) and PAR-1^−/− (n=12; solid bars) mice are shown.

Figure 3

Characterization of mice overexpressing PAR-1 in cardiomyocytes. A, PAR-1 mRNA expression was analyzed by Northern blotting in various tissues from transgenic (αMHC-PAR-1; line 18) and nontransgenic (WT) mice. Ten micrograms of total RNA was analyzed in each lane. PAR-1 mRNA expressed by the transgene is smaller than PAR-1 mRNA from the endogenous PAR-1 gene because it lacks most of the 3′ untranslated region. The larger bands may represent partially spliced transcripts. B, Comparison of PAR-1 mRNA expression in hearts from transgenic lines 18 and 43. GAPDH was used as a loading control. C, PAR-1 protein expression in αMHC-PAR-1 (line 18) mice and WT littermates was analyzed by Western blotting. Two exposures (shorter on top and longer below) are included to show the relative levels of PAR-1 in the 2 groups of mice. Four mice from each group are shown. D, PAR-1 expression in the hearts of WT, αMHC-PAR-1, and PAR-1^−/− mice was analyzed by immunohistochemistry. Original magnification ×1000.

Figure 4

Phenotype of mice overexpressing PAR-1 in cardiomyocytes. A, Western blot analysis of the levels of phosphorylated ERK1/2 in the hearts of WT and αMHC-PAR-1 mice at 6 months of age. Three mice are shown per group. Data are shown as mean±SD. B, Northern blot analysis of ANF and BNP mRNA expression in the LV of 6-month-old WT and αMHC-PAR-1 mice (5 mice per group). Levels of ANF and BNP mRNA normalized to GAPDH are shown (right). C, The heart weight:body weight (HW:BW) ratio was determined in 4-, 6-, and 10-month-old WT mice and αMHC-PAR-1 mice (5 to 10 mice per group). D, Representative hearts from 10-month-old WT and αMHC-PAR-1 mice. E, Representative cross sections of hearts from 6-month-old WT and αMHC-PAR-1 mice cut at the level of the papillary muscle (left) and quantitation of the lumenal area of the LV and thickness of the LV free wall in WT (n=6) and αMHC-PAR-1 (n=6) mice (right). F, Cross-sectional area of cardiomyocytes in hearts of WT and αMHC-PAR-1 (line 18) mice. Six mice were analyzed for each group, and 30 to 40 cardiomyocytes were measured per mouse. In panels with bars, results from WT mice (open bars) and αMHC-PAR-1 mice (solid bars) are shown.

Figure 5

Echocardiographic analysis of heart size and function in αMHC-PAR-1 mice. A, Representative M-mode echocardiographic images of an αMHC-PAR-1 mouse and a WT littermate control at 10 months of age. B, LV diameter (LVD), anterior and posterior LV wall thickness, and LV function (percentage of fractional shortening, %FS) were measured at diastole (d) and systole (s) by echocardiography in 10-month-old mice. WT (open bars) and αMHC-PAR-1 (solid bars) mice are shown (5 mice per group).

Figure 6

Deletion of the TF gene in cardiomyocytes reduces cardiac hypertrophy in αMHC-PAR1 mice. A, Levels of phosphorylated ERK1/2 were analyzed by Western blotting. Three mice were analyzed per group. Data are shown as mean±SD. B, Heart weight:body weight (HW:BW) ratio. C, Levels of ANF mRNA expression in the different groups of mice were normalized to levels of GAPDH mRNA. The Northern blot above shows 3 representative mice from each group. D and E, Measurement of lumenal area of the LV and thickness of LV free wall. Representative cross sections of hearts cut at the level of the papillary muscle from αMHC-PAR1/TF^flox/flox mice with or without the αMHC-Cre transgene are shown. Quantitation is shown below. n indicates number of mice in each group. Data are shown for 4-month-old male mice on a mixed genetic background (75% C57Bl/6 and 25% Sv129). αMHC-PAR1 mice on this genetic background developed hypertrophy faster than αMHC-PAR1 mice on a C57Bl/6 genetic background.