Signaling Pathways Involved in Human Vascular Smooth Muscle Cell Proliferation and Matrix Metalloproteinase-2 Expression Induced by Leptin : Inhibitory Effect of Metformin

Recombinant human leptin was purchased from R&D Systems (Minneapolis, MN). Vitamin E and 2′,7′-dichlorodihydrofluorescein diacetate (DCF-DA) were obtained from Sigma (St Louis, MO). PBS was obtained from Invitrogen (Burlington, ON, Canada). Smooth muscle cell basal medium (SmBM) and smooth muscle cell growth medium (SmGM-2) were purchased from BioWhittaker (Walkersville, MD). Fetal bovine serum (FBS) and trysin/EDTA were obtained from Wisent (St. Bruno, PQ, Canada). Apocynin, a selective NAD(P)H oxidase inhibitor (20); diphenyleneiodonium (DPI) chloride, an inhibitor of flavoprotein-containing enzymes, including NAD(P)H oxidase (21); thenoyltrifluoroacetone (TTFA), a mitochondrial complex II inhibitor (22); GF10923X, a selective protein kinase C (PKC) inhibitor (23); PD98059, a selective mitogen-activated protein kinase (MAPK) kinase inhibitor (24); and BAY 11-7085, a selective nuclear factor-κB (NF-κB) inhibitor (25) were obtained from Calbiochem (La Jolla, CA). Metformin was supplied by Merck-Santé (Lyon, France).

HASMCs were obtained from BioWhittaker (Walkersville, MD). The cells were grown in SmGM-2 at 37°C in a 5% CO₂/95% air atmosphere. At confluence, cells were trypsinized and subcultured in 24- or 96-well culture plates or 100-mm tissue culture dishes depending on assay conditions. Cells were used in all experiments between passages 3 and 6. Leptin treatment was carried out in cells cultured in SmBM supplemented with 2% FBS and 1% (vol/vol) penicillin-streptomycin.

Subconfluent HASMCs were serum starved in SmBM supplemented with 2% FBS for 48 h, preincubated for 1 h with metformin or other appropriate agents, and then treated with leptin for an additional 72 h at 37°C. HASMC proliferation was assessed using the 3-(4,5-dimethylthiazol-2-yl)-5-diphenyltetrazolium bromide (MTT) assay (Promega, Madison, WI). This assay is a colorimetric method based on reduction of the tetrazolium salt, MTT, by actively growing cells to produce a blue formazan product (26).

Subconfluent HASMCs were pretreated or not pretreated for 1 h with metformin or other appropriate agents prior to exposure to leptin for a further 24-h period at 37°C, with the cell-permeable fluorogenic probe DCF-DA (20 μg/ml) added during the final 20-min incubation period. At the end of this incubation period, cells were washed and trypsinized. Intracellular reactive oxygen species (ROS) production was monitored by measuring fluorescence in an LS50B luminescence spectrophotometer (Perkin Elmer) using excitation and emission wavelengths of 498 and 522 nm, respectively.

Subconfluent HASMCs were pretreated or not pretreated for 1 h with metformin or other appropriate agents and then incubation was pursued in the presence of leptin for an additional 72 h at 37°C. The amount of MMP-2 secreted in the culture medium was measured using a double-sandwich enzyme-linked immunosorbent assay (ELISA) (Amersham Biosciences, Baie d’Urfé, QC, Canada) (27). The minimum detectable concentration of MMP-2 with this assay is 1.5 ng/ml. The intra- and interassay coefficients of variation of this assay were <5.3 and 8.3, respectively. Levels of MMP-2 in the supernatants were normalized to the levels of total cell proteins.

Confluent HASMCs were treated or not treated with leptin in 100-mm tissue culture dishes in the presence or absence of metformin for 30 min at 37°C. At the end of this incubation period, cells were washed three times with cold PBS and harvested. Cell pellets were suspended in 1 ml cold sample preparation buffer (50 mmol/l Tris-HCl, pH 7.5, 5 mmol/l EDTA, 10 mmol/l ethylene glycol-bis tetraacetic acid, 50 mmol/l 2-mercaptoethanol, 1 mmol/l phenylmethylsulfonyl fluoride, and 10 mmol/l benzamidine) and sonicated. Membrane and cytosol fractions were separated by centrifugation at 100,000g for 1 h at 4°C. Cytosolic and membrane PKC activities were measured using the MESACUP protein kinase assay kit (MBL, Naka-ku Nagoya, Japan) (28).

Confluent HASMCs were treated or not treated with leptin in 96-well plates in the presence or absence of metformin or other appropriate agents for 30 min at 37°C. At the end of this incubation period, cells were rapidly fixed with 4% formaldehyde. Total and phosphorylated extracellular signal–regulated kinase (ERK)1/2 activities were measured using a fast-activated cell-based ELISA (FACE) kit (MJS Biolynx, Brockville, ON, Canada) (29). Levels of total and phosphorylated ERK1/2 were normalized to cell number as determined by crystal violet cell staining.

Confluent HASMCs were treated or not treated with leptin in 100-mm tissue culture dishes in the presence or absence of metformin or other appropriate agents for 30 min at 37°C. At the end of this incubation period, cells were washed twice with cold PBS/phosphate inhibitor buffer (125 mmol/l sodium fluoride, 250 mmol/l β-glycerophosphate, 250 mmol/l paranitrophenyl phosphate, and 25 mmol/l sodium orthovanadate), scraped, and centrifuged. Cell pellets were resuspended in 1 ml ice-cold hypotonic buffer (20 mmol/l HEPES, pH 7.5, 5 mmol/l sodium fluoride, 10 μmol/l sodium molybdate, and 0.1 mmol/l EDTA), and incubation was pursued on ice for 15 min. At the end of this incubation period, 10% nonidet P-40 was added to the lysed cells. After centrifugation, nuclear pellets were resuspended in Complete Lysis Buffer (MJS Biolynx) and nuclear extracts were collected by centrifugation at 9,000g for 10 min. Aliquots of the supernatants were kept at −70°C, and protein concentration was determined. Levels of NF-κB activation in 3-μg nuclear extracts were determined using the TransAM NF-κB family transcription factor assay kit (MJS Biolynx) (30).

Total protein content was measured according to the Bradford method using a colorimetric assay (Bio-Rad, Mississauga, ON, Canada).

To exclude the possibility that experimental agents at maximal concentration used in the study may exert cytotoxic effects, cell viability was determined by trypan blue exclusion. It was consistently found to be ≥90% (data not shown).

Statistical analysis of the results was performed by one-way ANOVA followed by the Student-Newman-Keuls test. Differences were considered statistically significant at P < 0.05. Results are expressed as means ± SE.

Incubation of HASMCs with leptin (0–200 ng/ml) increased the proliferation of these cells in a concentration-dependent manner. Maximal effect was observed after 72 h at a concentration of 100 ng/ml leptin (Fig. 1A). Leptin also enhanced, in a dose-dependent manner, MMP-2 production (Fig. 1B) and intracellular accumulation of ROS (Fig. 1C) in HASMCs, with a maximal effect at 100 ng/ml. Pretreatment of HASMCs with vitamin E abolished the stimulatory effects of leptin on HASMC proliferation, MMP-2 expression, and ROS production (Fig. 1A–C).

To evaluate the role of NAD(P)H oxidase and mitochondria in leptin-induced intracellular ROS production, HASMCs were preincubated with the NAD(P)H oxidase inhibitors apocynin and DPI chloride or with the inhibitor of the mitochondrial electron transport, TTFA, prior exposure to leptin. As shown in Fig. 2, apocynin and DPI chloride prevented the effect of leptin on intracellular ROS production, whereas TTFA was ineffective, thus identifying NAD(P)H oxidase as the source of ROS in leptin-treated HASMC. Like TTFA, the mitochondrial superoxide dismutase mimetic, MnTBAP [Mn(III)tetrakis (4-benzoic acid) pophyrin chloride] failed to inhibit leptin-induced ROS production in HASMCs (ROS production [percent of control values]: control 100 ± 2, leptin 156 ± 11, P < 0.05 vs. control; MnTBAP + leptin 135 ± 6). To determine whether PKC is involved in the stimulation of ROS by leptin, PKC activation in leptin-treated HASMCs was assessed and the effect of PKC inhibition on leptin-induced ROS generation determined. As shown in Fig. 3, treatment of HASMCs with 100 ng/ml leptin for 30 min induced PKC activation, as assessed by the translocation of this kinase to the membrane fraction (Fig. 3A). Pretreatment of HASMCs with the specific classic PKC inhibitor GF10923X totally prevented leptin-induced ROS generation (Fig. 3B). Apocynin and GF10923X also inhibited the effect of leptin on HASMC proliferation (Fig. 4A) and MMP-2 expression (Fig. 4B).

Incubation of HASMCs for 30 min with 100 ng/ml leptin resulted in a marked increase in phophospecific ERK1/2 expression (Fig. 5A) and nuclear levels of NF-κB activation (Fig. 5B). Leptin-induced ERK1/2 activation was abolished by GF10923X, apocynin, and PD98059 (Fig. 5A). These compounds, as well as BAY 11-7085, also prevented the induction of NF-κB in response to leptin (Fig. 5B). Inhibition of the ERK1/2 and NF-κB signaling pathways totally abrogated the stimulatory effect of leptin on HASMC proliferation (Fig. 5C) and MMP-2 production (Fig. 5D).

Preincubation of HASMCs with metformin (1–5 μg/ml) inhibited the stimulatory effect of leptin (100 ng/ml) on HASMC proliferation (Fig. 6A) and MMP-2 expression (Fig. 6B) with maximal effect being observed at concentrations ranging from 2.5 to 5 μg/ml metformin (Fig. 6B). Metformin (5 μg/ml) also abolished leptin-induced intracellular accumulation of ROS in HASMCs (Fig. 7A) and inhibited PKC (Fig. 7B), ERK1/2 (Fig. 7C), and NF-κB (Fig. 7D) activation in these cells.

Metformin has multiple biological effects, among which vasculoprotective properties have recently been identified (18). In experimental models of atherosclerosis, administration of metformin reduces the development and extent of atherosclerotic lesion formation (31). In patients with diabetes, this drug reduces the risk for developing macrovascular complications (16). The biological mechanisms responsible for the antiatherosclerotic properties of metformin are poorly understood but may include a suppressive effect of this drug on early key cellular events involved in atherogenesis, such as arterial lipid accumulation and metabolism (32,33), leukocyte-endothelium interaction, foam cell formation (19), and smooth muscle cell proliferation (34,35). It is well established that proliferative growth and migration of vascular smooth muscle cells (VSMCs) contribute to the arterial neointimal thickening observed in atherosclerosis, and recent in vivo and in vitro studies have demonstrated that leptin promotes neointimal growth in mice (36), VSMC proliferation (7), and endothelial cell MMP expression (6), thus supporting a role for this hormone in vascular lesion growth and matrix remodeling. Evidence that plasma leptin concentrations are associated with coronary atherosclerosis in patients with type 2 diabetes (37) further supports the possibility that leptin signaling may represent a therapeutic target for the prevention of atherosclerotic disease in these subjects. In smooth muscle cell, ROS are key mediators of the proliferative effect of many growth factors. Evidence that leptin stimulates smooth muscle cell growth and induces oxidative stress both in vivo and in vitro (5,38–40) supports the possibility that ROS may act as second messengers in leptin-induced smooth muscle cell proliferation. In line with this possibility, our results demonstrate that leptin induces intracellular ROS production in HASMCs and that vitamin E inhibits the proliferative effect of leptin on these cells. Leptin-induced ROS generation in HASMCs may arise from the NAD(P)H oxidase. Indeed, this enzyme is the major source of ROS in VSMCs, and redox signaling through NAD(P)H oxidase induces smooth muscle cell proliferation (41,42). Our observations that inhibitors of NAD(P)H oxidase totally abolish leptin-induced ROS production and proliferation in HASMCs indicate that ROS generation in these cells arises from the NAD(P)H oxidase system and mediates the proliferative effect of leptin in HASMCs. Previous studies have demonstrated that metformin has antioxidant properties (18,43–45). Consistent with these observations, we found that metformin suppressed the intracellular accumulation of ROS in leptin-treated HASMCs. These results, together with our observations that metformin inhibits the effect of leptin on smooth muscle cell growth to a similar extent as vitamin E, suggest that the suppressive effect of metformin on leptin-induced HASMC proliferation may be related to its antioxidant effects. Accumulating data indicate that stimulation of ROS through PKC-dependent activation of NAD(P)H oxidase and subsequent ERK activation may be critical mechanisms responsible for smooth muscle cell proliferation (46–48). Previous studies have shown that leptin induces PKC and MAPK activation in vascular cells (5,49,50) and stimulates VSMC proliferation through the MAPK pathway (7). In agreement with these data, we found that leptin induces PKC and MAPK activation in HASMCs and that inhibition of these signaling events abrogates the proliferative effect of leptin on these cells. Our findings that inhibition of PKC suppresses ROS production in leptin-treated HASMCs and that apocynin blocks MAPK activation in these cells further support the notion that generation of ROS in leptin-stimulated HASMCs is PKC dependent and that leptin-induced MAPK activation involves NAD(P)H oxidase–generated ROS. Interestingly, our data demonstrate that metformin inhibits PKC and MAPK activation in leptin-treated HASMCs. On the basis of previous observations showing that inhibition of VSMCs by vitamin E correlates with PKC inhibition (51,52), it is tempting to speculate that metformin, by reducing PKC activity, may reduce ROS generation, thereby leading to decreased MAPK activation and smooth muscle cell growth. Because metformin activates AMP kinase (AMPK) and that activation of this kinase appears to reduce diacylglycerol synthesis (53–55), metformin’s inhibitory effect on PKC activation may involve AMPK activation, with consequent inhibition of diacylglycerol synthesis.

Proliferation of VSMCs is regulated by nuclear transcription factors including NF-κB. NF-κB activation is redox sensitive and is induced through PKC- and MAPK-dependent pathways (56,57). Previous data have proposed a role for ROS-dependent activation of NF-κB in leptin-induced activation of endothelial cells (38). The present study, which demonstrates that leptin induces HASMC proliferation through NF-κB activation, further stresses the crucial role of this transcription factor in the regulation of vascular cell function by leptin. In accordance with previous observations (38,56,57), we found that activation of NF-κB by leptin was linked to ROS generation and PKC/ERK1/2 activation. We further demonstrated that metformin inhibits leptin-induced NF-κB activation. Because AMPK appears to inhibit NF-κB activation in endothelial cells (58), metformin’s inhibitory effect on NF-κB activation in HASMCs may involve AMPK activation.

MMP expression is associated with smooth muscle cell proliferation and migration. MMP-2 is expressed abundantly in atherosclerotic lesions and plays an important role in increasing VSMC migration to the intima (59). Evidence that MMP-2 expression is upregulated by leptin in vascular cells (6) suggests that this effect may account, at least partly, for the neointimal growth–promoting property of leptin (36). Induction of MMP-2 is redox sensitive and occurs, in angiotensin-treated smooth muscle cell, in a NAD(P)H oxidase–dependent manner (60). In accordance with these data, the present study demonstrates that leptin induces, through a PKC-dependent NAD(P)H oxidase activation, MMP-2 expression in HASMCs. Furthermore, it demonstrates that this effect involves activation of the MAPK/NF-κB pathway. These findings are in line with previous observations showing a role for PKC, ERK1/2, and NF-κB in the regulation of MMP secretion (61,62). Our observation that metformin abrogates the induction of MMP-2 by leptin in HASMCs further supports the possibility that this drug may reduce the development of atherosclerotic lesions by regulating intimal thickening. Supporting this possibility, metformin treatment of cholesterol-fed rabbits has been shown to inhibit smooth muscle cell proliferation (35). In conclusion, this study demonstrates that metformin inhibits, by interfering with the PKC/MAPK/NF-κB pathways, leptin-induced HASMC proliferation and MMP-2 expression in vitro. These data suggest a new mechanism by which metformin may reduce cardiovascular risk in type 2 diabetic subjects.

This study was supported by Merck-Santé (Lyon, France).