Actin Cytoskeleton Dynamics Promotes Leptin-Induced Vascular Smooth Muscle Hypertrophy via RhoA/ROCK- and Phosphatidylinositol 3-Kinase/Protein Kinase B-Dependent Pathways

Leptin is a pleiotropic peptide that is a product of the ob gene and that is produced primarily by adipocytes (Zhang et al., 1994). Circulating leptin concentrations are proportional to the degree of adiposity, which is a risk factor for cardiovascular disease such as hypertension and atherosclerosis (Söderberg et al., 1999; Wallace et al., 2001) by remodeling vascular smooth muscle cells (VSMC). Whether leptin directly contributes to increased cardiovascular risk is not known, although the peptide has been shown to directly affect vascular function by decreasing arterial distensibility (Singhal et al., 2002) and increasing platelet aggregation and arterial thrombosis (Nakata et al., 1999; Schäfer et al., 2004) and by promoting neointimal growth of VSMC after injury in mice (Schafer et al., 2004). Moreover, leptin seems to contribute to vascular remodeling by inducing an increase in VSMC surface area and protein synthesis (Shin et al., 2005; Zeidan et al., 2005) and by inducing proliferation and migration of VSMC (Oda et al., 2001). It has been shown that leptin promotes cellular effects through activation of the leptin receptors (OBR). OBR stimulation in various tissues is associated with the activation of a plethora of cell signaling pathways, including Janus tyrosine kinase/signal transducer and activator of transcription (Baumann et al., 1996), mitogen-activated protein kinase (Zeidan et al., 2005), RhoA (Zeidan et al., 2006), protein kinase C, and phosphatidylinositol 3-kinase/protein kinase B (PI3K/Akt) (Martin-Romero and Sanchez-Margalet, 2001; Oda et al., 2001), as well as increased expression of angiotensinogen and preproendotheilin-1 (Zeidan et al., 2005).

Although, as indicated above, leptin affects a number of cell signaling pathways, at present, the mechanism(s) by which leptin induces vascular hypertrophy is not well understood. We were particularly interested in the role of the RhoA/ROCK because this system seems to play a critical role in leptin-induced cardiomyocyte hypertrophy (Zeidan et al., 2006), although its role in vascular tissue has not been studied. Accordingly, the present study was designed to first determine whether leptin-mediated VSMC hypertrophy occurs via the RhoA/ROCK pathway and to identify the downstream mediators, including actin dynamics and serum-response factor (SRF), that participate in this process. In addition, we determined whether the activation of PI3K/Akt pathway is required for leptin-induced VSMC hypertrophy and the nature of its involvement.

Materials and Methods

Preparation of Isolated Rat Portal Veins. Male Sprague-Dawley rats weighing 200 to 250 g (Charles River Canada, Montreal, QC, Canada) were killed by decapitation. Rat portal veins (RPVs) were dissected under sterile conditions and opened longitudinally. The RPVs were denuded and transferred to culture medium (Dulbecco's modified Eagle's medium and Ham's F-12 medium [1:1 (v/v)], containing 50 U/ml penicillin, and 50 μg/ml streptomycin) for 1 to 3 days as described previously (Zeidan et al., 2005) in the absence or presence of 3.1 nM leptin (Sigma-Aldrich, Oakville, ON, Canada), a concentration within the range of plasma concentrations found in obese individuals (Maffei et al., 1995). For some experiments, denuded RPVs were treated with an anti-leptin receptor antibody (OBR-Ab; Alpha Diagnostic International, Inc., San Antonio, TX) at 16.6 or 166 ng/ml added 1 h before leptin administration. The selective ROCK inhibitor Y-27632 at 10 μM (Sigma-Aldrich), actin depolymerization agent latrunculin B (Calbiochem, San Diego, CA) at 50 nM, or the PI3K inhibitor LY294002 at 0.5, 10, and 50 μM (Cell Signaling Technology Inc., Danvers, MA), when used, was also added 1 h before leptin administration.

Protein Synthesis. For analysis of protein synthesis, [³H]leucine, at activities of 0.2 and 1 μCi/ml, was added to the culture medium, and incorporation was determined as described previously (Zeidan et al., 2005).

Immunoblotting. Immunoblotting was performed as described previously (Zeidan et al., 2005). Protein extracts were analyzed by Western blotting with antibodies against the following proteins: RhoA, total-cofilin-2, phospho (p)-cofilin-2, serum-response factor (SRF; Upstate Biotechnology, Charlottesville, VA), LIM kinase (LIMK), p-LIMK1, Akt, p-Akt (Cell Signaling Technology Inc.), or actin (Cytoskeleton Inc., Denver, CO). Signal was detected with Luminol Reagent (Santa Cruz Biotechnology, Inc., Santa Cruz, CA).

RhoA Activity Assay. In the presence of leptin with OBR-Ab, RhoA activity was measured using the RhoA activation assay kit (Upstate Biotechnology) following the manufacturer's instructions and as described previously (Zeidan et al., 2006).

Measurement of F-Actin/G-Actin Ratio. F-Actin content compared with the amount of G-actin was determined using the F-/G-actin in vivo assay kit (Cytoskeleton Inc.) according to the manufacturer's directions and as described previously (Zeidan et al., 2006). In brief, upon exposure to various stimuli and/or inhibitors, the RPVs were homogenized in lysis buffer and F-actin stabilization buffer [50 mM PIPES, 50 mM NaCl, 5 mM MgCl₂, 5 mM EGTA, 5% (v/v) glycerol, 0.1% (v/v) Nonidet P-40, 0.1% (v/v) Triton X-100, 0.1% (v/v) Tween 20, 0.1% (v/v) 2-mercaptoethanol, and 0.001% (v/v) antifoam and a protease inhibitor cocktail] followed by centrifugation for 1 h at 100,000g to separate the F-actin from G-actin pool. Supernatants of the protein extracts were collected after centrifugation at 100,000g for 60 min at 30°C. The pellets were resuspended in ice-cold distilled H₂O plus 1 μM cytochalasin D, and then they were incubated on ice for 1 h to dissociate F-actin. The resuspended pellets were gently mixed every 15 min. Equal amounts of both the supernatant (G-actin) and the resuspended pellet (F-actin) were subjected to analysis of immunoblot with the use of an actin antibody (Cytoskeleton Inc.).

Data Analysis. Data were analyzed using one-way analysis of variance followed by a post hoc Student's t test. All values are presented as means ± S.E.

Results

Leptin Signaling Activates the RhoA/LIMK/Cofilin Pathway in Vascular Smooth Muscle. To investigate the effect of leptin on RhoA pathway activation, we first measured Rho-GTP binding at various time points after leptin administration. RhoA cycles between an inactive (GDP-bound) form and an active GTP-bound form; therefore, a GTP-pull-down assay was used to measure active RhoA. As shown in Fig. 1A, the GTP-pull-down assays showed that leptin significantly increased the active RhoA, which peaked (163 ± 20%) 5 min after leptin addition, an effect that was inhibited by an OBR-Ab in a concentration-dependent manner (Fig. 1B), demonstrating the importance of OBR receptor occupancy in RhoA activation. We next determined the phosphorylation of two downstream molecules in the RhoA pathway, namely, LIMK1 and cofilin-2. As shown in Fig. 1, C and E, leptin induced a time-dependent stimulation in both LIMK1 and cofilin-2 phosphorylation with peak values (160 ± 25 and 290 ± 25% increase in LIMK1 and cofilin-2 phosphorylation, respectively) attained 60 min after leptin addition and with phosphorylation declining to control levels after 24-h treatment. Both the increases in LIMK1 and cofilin-2 phosphorylation were significantly attenuated by OBR-Ab treatment (Fig. 1, D and F).

Leptin Modulates Actin Dynamics. To further investigate the effect of leptin on RhoA/ROCK downstream signaling, we determined actin cytoskeleton dynamics by measuring the amounts of F-actin and G-actin after culturing RPVs for 24 h with leptin. Fractionated cell extracts containing free G-actin and F-actin were analyzed using Western blotting. As shown in Fig. 2, leptin caused a 210 ± 17% increase in the F-/G-actin ratio in tissue treated with leptin (p < 0.05 versus control). The increased F-/G-actin ratio most probably reflects increased cofilin phosphorylation that results in cofilin inactivation and thus an inhibition in F-actin depolymerization. As shown in Fig. 2, the ability of leptin to increase the F-/G-actin ratio was prevented by the OBR-Ab as well as the ROCK inhibitor Y-27632.

Leptin-Induced VSMC Hypertrophy Is Dependent on the RhoA/ROCK Pathway and Modulation of Actin Cytoskeleton. We next examined whether the activation of RhoA was required for leptin-induced RPV hypertrophy. For this study, RPV strips were treated with leptin and cultured for 3 days, after which tissue wet weight and protein synthesis were determined. As shown in Fig. 3A, control tissues demonstrated a decrease in weight of approximately 10% after 3 days in culture, a phenomenon observed previously, although the precise mechanism underlying this effect has not been elucidated (Zeidan et al., 2000, 2003, 2005). However, in the presence of leptin, the reduction in tissue weight was abrogated, and vessel weight was increased by 11 ± 1% (Fig. 3A), which was associated with significantly increased protein synthesis (45 ± 7%; Fig. 3B). There were no significant differences in the dry weight/wet weight ratios in control and treated groups, indicating that increased tissue weights were not due to increased water retention. These results agree with our previous results showing that the hypertrophic effect of leptin is associated with increased in cell surface area (Zeidan et al., 2005). Both the increased tissue weight and leucine incorporation were prevented by the ROCK inhibitor Y-27632 as well as latrunculin B, an F-actin-disrupting agent. The latter findings suggest that intact ROCK activity and/or F-actin stabilization and preservation mediate leptin-induced hypertrophy in these vessels.

PI3K/Akt Mediates Leptin-Induced Change in Actin Dynamics via LIMK/Cofilin Phosphorylation. To investigate whether leptin can induce Akt phosphorylation, time course experiments were performed (0, 5, 15, 30, and 60 min) to determine the effects of leptin on Akt phosphorylation. As shown in Fig. 4A, leptin rapidly stimulated Akt phosphorylation, which peaked at 5 min (130 ± 30%) and returned to control values after 30 min of treatment. To verify whether leptin can induce Akt phosphorylation through PI3K activation, we pretreated RPV with different concentrations (0.5, 5, and 50 μM) of the PI3K inhibitor LY294002 (Vlahos et al., 1994) followed by leptin addition. As shown in Fig. 4B, increasing concentrations of LY294002 caused a gradual decrease in leptin-induced Akt phosphorylation levels with complete inhibition seen with 50 μM LY294002.

As noted previously, PI3K/Akt activity is associated with changes in actin dynamics (Qian et al., 2004). Accordingly, we postulated that leptin-induced modulation of RhoA/ROCK and the subsequent changes in actin dynamics could involve PI3K/Akt signaling. Therefore, we determined the effect of 50 μM LY294002 on leptin-induced RhoA activation (5-min postleptin addition) as well as LIMK1 and cofilin-2 phosphorylation (60-min postleptin addition). As shown in Fig. 5A LY294002 was without effect on leptin-induced RhoA activation, whereas it significantly but incompletely inhibited leptin-induced LIMK1 (Fig. 5B) and cofilin-2 (Fig. 5C) phosphorylation. Moreover, the leptin-induced increase in the F-/G-actin ratio (24-h postleptin addition) was similarly significantly blunted by LY294002 (Fig. 5D). Although not shown, it is important to note that the ability of LY294002 to inhibit activation of the RhoA/ROCK pathway was associated with diminished leptin-induced RPV hypertrophy as determined by both the percentage increase tissue wet weight (10 ± 3% control versus 2 ± 3% with LY294002; p < 0.05) as well as leucine incorporation (45 ± 15% versus 20 ± 10%; p < 0.05).

Leptin Induces SRF Overexpression in a RhoA/ROCK and PI3K/Akt-Dependent Manner. SRF transcriptional complex is an important regulator of numerous hypertrophic genes (Nelson et al., 2005), and previous reports have suggested that SRF is regulated by changes in actin dynamics (Sotiropoulos et al., 1999; Miralles et al., 2003) and that SRF protein concentration is increased during hypertrophy of skeletal muscle (Flück et al., 1999). We determined whether leptin-induced VSMC hypertrophy was associated with increase on SRF expression, and we assessed the possible contribution of the RhoA/ROCK and PI3K/Akt pathways. As shown in Fig. 6, leptin significantly increased the SRF content (136 ± 11%). This increase was significantly attenuated by the OBR-Ab as well as by the PI3K inhibitor LY294002 and the ROCK inhibitor Y-27632.

Discussion

Obesity, particularly visceral obesity, is a key risk factor for cardiovascular disease. The role of hyperleptinemia as a contributing factor to increased cardiovascular risk in obesity is not known with certainty. However, leptin exerts a number of effects, including increased VSMC proliferation (Oda et al., 2001), arterial distensibility (Singhal et al., 2002), platelet aggregation and arterial thrombosis (Nakata et al., 1999; Konstantinides et al., 2004), VSMC hypertrophy (Zeidan et al., 2005), and blood pressure (Dunbar et al., 1997), which makes it a possible candidate as a contributing factor to increased cardiovascular disease associated with obesity. In this report, we show for the first time a novel interrelationship between leptin-induced vascular hypertrophy and the RhoA/ROCK and PI3K/Akt pathway. This is based on several lines of evidence. First, we have shown that leptin administration increases phosphorylation of key components of the RhoA/ROCK pathways, including RhoA, LIMK1, and cofilin-2, resulting in a 2-fold increase in the F-/G-actin ratio. Moreover, pharmacological inhibition of ROCK and actin depolymerization prevents the hypertrophic response. We also show that leptin induces Akt phosphorylation and that Akt inhibition significantly attenuates leptin-induced downstream stimulation of the RhoA pathway, thereby showing a dependence of LIMK1/cofilin-2 activation on Akt activity. Finally, we also demonstrate in this study that leptin upregulates protein expression of SRF through a RhoA/ROCK- and PI3K/Akt-dependent pathway. Taken together, these results suggest that leptin-induced hypertrophy of VSMC is mediated by changes in actin dynamics and SRF expression, which is regulated by RhoA activation and partially by the PI3K/Akt pathway. On a cellular level, hypertrophy is characterized by morphological changes, including an increase in cell size and the reorganization of actin cytoskeleton through RhoA/ROCK-dependent pathway (for review, see Brown et al., 2006). Recent studies support a potential role of RhoA in modulating signal transduction pathways regulating actin cytoskeletal dynamics by phosphorylation (inhibition) of cofilin (Zeidan et al., 2006). Cofilin is an actin-binding protein that critically controls actin filament dynamics and reorganization by severing and depolymerizing actin filaments (Bamburg and Wiggan, 2002). The activation of ROCK leads to LIMK1 phosphorylation, which then phosphorylates cofilin-2, thereby inhibiting its activity (Arber et al., 1998), and to decreased the G-actin pool in the cytosol. Therefore, the established role of the RhoA/ROCK pathway as a key regulator of actin dynamics (Mack et al., 2001) probably reflects its ability to modulate actin dynamics through a cofilin-dependent mechanism. In addition to evidence based on the ability of leptin to activate the RhoA/ROCK pathway, the role of this pathway in mediating the hypertrophic effect of leptin is further reinforced by the ability of the ROCK inhibitor Y-27632 to abrogate leptin-induced vascular hypertrophy as determined by measurements of wet weight and [³H]leucine incorporation. Moreover, Y-27632 was also able to prevent the leptin-induced increase in actin polymerization as evidenced by significantly attenuating the increase in the F-/G-actin ratio, whereas increasing F-actin depolymerization with latrunculin B prevented leptin-induced hypertrophy. Taken together, the results offer strong evidence that the RhoA/ROCK pathway contributes to leptin-induced hypertrophy through a mechanism involving increased actin polymerization. How this occurs requires further investigation, although our study suggests the involvement of SRF, an important transcriptional factor that has been shown to regulate the expression of various genes involved in the hypertrophic phenotype (Nelson et al., 2005). Moreover, it has been reported that cardiac-specific overexpression of SRF transgene in mice resulted in cardiac hypertrophy, indicating that SRF is an important downstream regulator of the hypertrophic program (Zhang et al., 2001). In the present study, we reported that the leptin-induced increase in SRF protein is RhoA/ROCK- and PI3K/Akt-dependent. Miralles et al. (2003) have shown previously that RhoA/actin signaling regulated the subcellular localization of the SRF coactivator megakaryocytic acute leukemia. Megakaryocytic acute leukemia is normally sequestered by G-actin in the cytoplasm of serum-starved fibroblasts, but it accumulates in the nucleus following serum stimulation. As such, it is interesting to postulate a model in which leptin-induced decrease in free G-actin resulting in SRF activation and translocation may contribute to VSMC hypertrophy, although there is no direct evidence for this at present.

A large and growing body of evidence suggests that the PI3K/Akt pathway can regulate many cellular responses, including proliferation, cytoskeletal rearrangements, angiogenesis, and cell migration (Brazil et al., 2002). The PI3K/Akt pathway can be activated by growth factors, cytokines, and insulin (Cantley, 2002) and activation of this pathway leads to VSMC hypertrophy (Higaki and Shimokado, 1999; Hixon et al., 2000). Moreover, angiotensin II induces cardiomyocyte hypertrophy in a PI3K/Akt-dependent manner (Hingtgen et al., 2006), indicating the important role of PI3K/Akt signaling in hypertrophy. Furthermore, it was found that activity of the PI3K/Akt pathway is required for actin filament remodeling, an important hypertrophic (Morales-Ruiz et al., 2000) and cell migration factor (Gerthoffer, 2007), in numerous cell signaling pathways. It has been shown that leptin-induced VSMC migration and proliferation can be attenuated by LY294002 (Oda et al., 2001), suggesting an important role of PI3K signaling in actin cytoskeleton dynamics. However, the role of these signaling pathways in mediating leptin-induced hypertrophy of VSMC has not been determined. To gain additional insight into the molecular mechanisms that lead to leptin-induced VSMC hypertrophy, we analyzed the involvement of the PI3K/Akt pathway and its interaction with RhoA activation. Our study shows an OBR dependence for leptin activation of both pathways. In addition, we show that PI3K/Akt probably contributes to leptin-induced hypertrophy via the RhoA/ROCK pathways, because both the hypertrophy and activation of the RhoA pathway were significantly attenuated by a PI3K inhibitor. However, although PI3K inhibition completely prevented leptin-induced hypertrophy at a concentration that abrogated Akt phosphorylation, only a partial inhibition of RhoA, LIMK, and cofilin phosphorylation or F-/G-actin up-regulation was observed. This suggests that although PI3K/Akt only partially regulates the RhoA/ROCK pathway and actin polymerization, it seems to be essential for the leptin-induced hypertrophic response.

In conclusion, we have demonstrated in this study that leptin induces hypertrophy in RPVs through a mechanism involving activation of the RhoA cascade and that it is dependent on actin polymerization. An important role for the PI3K/Akt pathway was also demonstrated. Overall, these findings provide evidence for a novel pathway involved in leptin-induced vascular hypertrophy that may be a valuable tool in terms of developing therapeutic approaches aimed at mitigating the potential deleterious effects of leptin on vascular remodeling. Future investigations are necessary to more precisely delineate this pathway especially in terms of identifying precisely how RhoA signaling affects SRF expression as well as to identify specific targets downstream from SRF up-regulation.

Limitations of Study. Although the use of isolated vascular tissue may not necessarily reflect the in vivo setting, this approach mitigates any potential role of circulating leptin on the vascular responsiveness to exogenous peptide. In addition, importantly, this precludes the potential central effects of leptin that can affect vascular function via the sympathetic nervous system. Results obtained using venous tissues may not reflect effects on other vascular beds. In terms of vascular pathology, our results could have direct relevance to portal hypertension. However, this preparation has been widely used to study VSMC physiology and pharmacology (for review, see Sutter, 1990), and it has been well characterized in a number of recent publications (Zeidan et al., 2000, 2003, 2005; Albinsson et al., 2004). Because portal veins branch into capillaries (sinusoids) and because of their myogenic vasomotion, RPVs have been used for the development of vasoactive drugs as an analog of small precapillary resistance vessels (Ljung, 1990). Moreover, the longitudinally oriented musculature in the RPV make it an ideal blood vessel for investigating the effect of mechanical stress on muscle hypertrophy by stretching it by weight loading rather than perfusion. Previously, we have studied the effect of stretch on cultured RPVs, which causes increases in contractility, protein synthesis, and cell cross-sectional area relative to unstretched, control veins (Zeidan et al., 2000). These data resemble the pattern seen with elevated venous pressure in vivo over a similar time (Malmqvist and Arner, 1988), suggesting that chronic exposure to leptin, especially under elevated conditions, may alter the structure and function of VSMC in vivo.