The ral exchange factor rgl2 promotes cardiomyocyte survival and inhibits cardiac fibrosis

Abstract

Cardiomyocytes compensate to acute cardiac stress by increasing in size and contractile function. However, prolonged stress leads to a decompensated response characterized by cardiomyocyte death, tissue fibrosis and loss of cardiac function. Identifying approaches to inhibit this transition to a decompensated response may reveal important targets for treating heart failure. The Ral guanine nucleotide disassociation (RalGDS) proteins are Ras-interacting proteins that are upregulated by hypertrophic stimuli. The Ral guanine nucleotide dissociation stimulator-like 2 (Rgl2) is a member of the RalGDS family that modulates expression of hypertrophic genes in cardiomyocytes. However, the pathophysiologic consequence of increased Rgl2 expression in cardiomyoctyes remains unclear. To evaluate the effect of increasing Rgl2 activity in the heart, transgenic mice with cardiac-targeted over-expression of Rgl2 were generated. Although Ral activation was increased, there were no apparent morphologic or histological differences between the hearts of Rgl2 transgenic and nontransgenic mice indicating that increased Rgl2 expression had no effect on basal cardiac phenotype. To determine if Rgl2 modulates the cardiac response to stress, mice were infused with the ß-adrenergic receptor agonist, isoproterenol. Isoproterenol infusion increased heart mass in both Rgl2 transgenic and nontransgenic mice. However, unlike nontransgenic mice, Rgl2 transgenic mice showed no morphologic evidence of cardiomyocyte damage or increased cardiac fibrosis following isoproterenol infusion. Increased Rgl2 expression in cultured cardiomyocytes stimulated Ral activation and inhibited staurosporine-induced apoptosis via increased activation of PI3-kinase. Activation of the PI3-kinase signaling pathway was confirmed in hearts isolated from Rgl2 transgenic mice. Increased expression and function of Rgl2 in cardiomyocytes promotes activation of the PI3-kinase signaling cascade and protects from carciomyocyte death and pathologic cardiac fibrosis. Taken further, these results suggest that Rgl2 upregulation in hypertrophic hearts may be a protetive mechanism, and that Rgl2 may be a novel therapeutic target in treating heart disease.

Figure 1. Transgenic expression of Rgl2 in mice.

A) Representative pictures of hearts from nontransgenic (NTg) and transgenic mice with cardiac specific overexpression of Rgl2 (Rgl2-Tg). Ruler gradations are 1 mm. Graph represents the mean ± SEM from at least 4 different aged-matched hearts. * denotes significant difference from NTg. B) Tissue homogenates from hearts of NTg and Rgl2-Tg mice were probed with an antibodies to the hemagglutinen (HA) epitope and Rgl2. Shown are a representative blot from a single experiment and the mean ± SEM (n = 3) of Rgl2 expression normalized to tubulin. * denotes significant difference from NTg. C) Photomicrographs of histological sections of ventricular tissue demonstrating Rgl2 expression in myofibrils of Rgl2-Tg. The sections depicted in the left panels were incubated without primary antibody and the right sections were incubated with a polyclonal antibody to the HA epitope (200×magnification).

Figure 2. Expression of Rgl2 activates Ral.

Heart homogenates from Rgl2-Tg and NTg littermate mice were prepared and the amount of GTP-bound Ral (active) assessed by immunoblotting. Total Ral was assessed on separate immunoblots. Total and GTP-bound Ral were quantified and a representative image from a single experiment and the means ± SEM from at least 6 individual mice are shown. * denotes significant difference from NTg.

Figure 3. Effect of Rgl2 on isoproterenol induced cardiac hypertrophy.

Rgl2-Tg and NTg littermates were continuously infused with isoproterenol (30 mg/kg/day; ISO) or vehicle (ascorbic acid; AA) for 14 days via mini-osmotic pumps. The effect of isoproterenol infusion on cardiac hypertrophy (A,B) and pathology (C–E) were determined. A) Heart weight (HW) and body weight (BW) were determined and are expressed as the heart weight (HW) to body weight (BW) ratio. B) Cardiomyocyte cross-sectional area was quantified by morphometric analysis of individual cells and expressed relative to untreated NTG mice. C) Pathological changes in hearts of Rgl2-Tg and NTg mice following infusion were assessed using a modified elastic trichrome stain to visualize collagen (green). Shown are representative sections from two different hearts from NTg and Rgl2-Tg mice at 200×magnification. Representative areas of fibrosis are indicated by black arrows, and myocyte vacuolization by yellow arrows. Fibrotic area (D) and extent of vacuolization (E) were quantified by morphometric analysis. Results are expressed as the mean ± SEM obtained from at least 9 different mice. F) Apoptotic index was determined by western blotting for Bax and Bcl-xl in left ventricular homogenates and expressed as the Bax/Bcl-xl ratio. A representative blot of lysates from 3 mice of each treatment and graph of mean ± SEM from 9 mice are shown. * denotes significant difference from AA infused NTg mice using Dunnett’s multiple comparison test.

Figure 4. Adenoviral-mediated expression of Rgl2 in cardiomyocytes.

NRVMs (A and C) or HL-1 cardiomyocytes (B and D) were infected with 90 ifu/cell of AdNull or AdRgl2. After infection, cell lysates were prepared and analyzed for Rgl2 expression (A and B) via immunoblotting using Rgl2 antibodies, and function (C and D) by assessing Ral activation. Rgl2 expression was normalized to cyclophilin A (cycloA) or tubulin. Ral activation was determined as described in Figure 2 and are presented as the ratio of values obtained in AdRgl2- to AdNull-infected cells. Shown are representative immunoblots of a single experiment and the means ± SEM of at least 4 separate experiments. * denotes significant difference from AdNull infected cells.

Figure 5. Effect of Rgl2 on apoptotic pathways in cardiomyocytes.

AdNull- and AdRgl2-infected NRVMs (A) and HL-1s (B and C) were treated with staurosporine (Staur, 1 µM) for 0, 2, and 4 hr. Caspase-3 (A and B) and PARP (C) activation were determined by immunoblot of cell lysates using antibodies to activated caspase-3, procaspase-3, and PARP. Active (cleaved) caspase-3 and PARP densities were normalized to total enzyme and are expressed relative to the respective control value (0 hr Staur). Shown are representative immunoblots from a single experiment and the means ± SEM of at least 3 separate experiments. * denotes significant difference compared to AdNull infected cells.

Figure 6. Effect of Rgl2 on the PI3K/Akt signaling pathway in cardiomyocytes.

Cell lysates were collected from AdNull or AdRgl2 infected NRVMs (A, C) or HL-1 cardiomyocytes (B) and immunoblotted with antibodies for pAkt, total Akt, or tubulin. (A and B) Rgl2-induced Akt phosphorylation was quantified and normalized to AdNull infected cells. (C) AdNull and AdRgl2 infected NRVMs were treated with wortmannin (Wort) for 30 min prior to preparing cell lysates. Shown are representative immunoblots from a single experiment and mean ± SEM of at least 3 separate experiments. * denotes significant difference from AdNull infected cells.

Figure 7. Effect of Rgl2 on staurosporine-induced caspase-3 and PARP cleavage.

AdNull- and AdRgl2-infected NRVM (A) and HL-1 cardiomyocytes (B and C) were pretreated with vehicle or 200 nM wortmannin (Wort) for 30 min followed by exposure to 1 µM staurosporine (Staur) for 0, 2, and 4 hr. Caspase-3 (A and B) and PARP (C) cleavage were analyzed as described in Figure 5. Results are presented normalized to the respective baseline value (0 hr Staur). Values are expressed as mean ± SEM of at least 3 separate experiments. * denotes significant difference from AdNull infected cells.

Figure 8. Effect of Rgl2 on the PI3K/Akt signaling pathway in transgenic mice.

Heart homogenates prepared from Rgl2-Tg and NTg littermates were prepared and Akt phosphorylation quantified by immunoblotting with a phosphospecific Akt antibody and normalizing to total Akt. Shown are a representative immunoblot and the mean ± SEM of 12 different homogenates. * denotes significant difference from NTg.