High Glucose Stimulates Synthesis of Fibronectin via a Novel Protein Kinase C, Rap1b, and B-Raf Signaling Pathway*

The molecular mechanism(s) by which high glucose induces fibronectin expression via G-protein activation in the kidney are largely unknown. This investigation describes the effect of high glucose (HG) on a small GTP-binding protein, Rap1b, expression and activation, and the relevance of protein kinase C (PKC) and Raf pathways in fibronectin synthesis in cultured renal glomerular mesangial cells (MCs). In vivo experiments revealed a dose-dependent increase in Rap1b expression in glomeruli of diabetic rat kidneys. Similarly, in vitro exposure of MCs to HG led to an up-regulation of Rap1b with concomitant increase in fibronectin (FN) mRNA and protein expression. The up-regulation of Rap1b mRNA was mitigated by the PKC inhibitors, calphostin C, and bisindolymaleimide, while also reducing HG- induced FN expression in non-transfected MCs. Overexpression of Rap1b by transfection with pcDNA 3.1/Rap1b in MCs resulted in the stimulation of FN synthesis; however, the PKC inhibitors had no significant effect in reducing FN expression in Rap1b-transfected MCs. Transfection of Rap1b mutants S17N (Ser → Asn) or T61R (Thr → Arg) in MCs inhibited the HG-induced increased FN synthesis. B-Raf and Raf-1 expression was investigated to assess whether Rap1b effects are mediated via the Raf pathway. B-Raf, and not Raf-1, expression was increased in MCs transfected with Rap1b. HG also caused activation of Rap1b, which was largely unaffected by anti-platelet-derived growth factor (PDGF) antibodies. HG-induced activation of Rap1b was specific, since Rap2b activation and expression of Rap2a and Rap2b were unaffected by HG. These findings indicate that hyperglycemia and HG cause an activation and up-regulation of Rap1b in renal glomeruli and in cultured MCs, which then stimulates FN synthesis. This effect appears to be PKC-dependent and PDGF-independent, but involves B-Raf, suggesting a novel PKC-Rap1b-B-Raf pathway responsible for HG-induced increased mesangial matrix synthesis, a hallmark of diabetic nephropathy.

From the Departments of ‡Pathology and §Medicine, Northwestern University Medical School, Chicago, Illinois 60611 The molecular mechanism(s) by which high glucose induces fibronectin expression via G-protein activation in the kidney are largely unknown. This investigation describes the effect of high glucose (HG) on a small GTP-binding protein, Rap1b, expression and activation, and the relevance of protein kinase C (PKC) and Raf pathways in fibronectin synthesis in cultured renal glomerular mesangial cells (MCs). In vivo experiments revealed a dose-dependent increase in Rap1b expression in glomeruli of diabetic rat kidneys. Similarly, in vitro exposure of MCs to HG led to an up-regulation of Rap1b with concomitant increase in fibronectin (FN) mRNA and protein expression. The up-regulation of Rap1b mRNA was mitigated by the PKC inhibitors, calphostin C, and bisindolymaleimide, while also reducing HGinduced FN expression in non-transfected MCs. Overexpression of Rap1b by transfection with pcDNA 3.1/Rap1b in MCs resulted in the stimulation of FN synthesis; however, the PKC inhibitors had no significant effect in reducing FN expression in Rap1b-transfected MCs. Transfection of Rap1b mutants S17N (Ser 3 Asn) or T61R (Thr 3 Arg) in MCs inhibited the HG-induced increased FN synthesis. B-Raf and Raf-1 expression was investigated to assess whether Rap1b effects are mediated via the Raf pathway. B-Raf, and not Raf-1, expression was increased in MCs transfected with Rap1b. HG also caused activation of Rap1b, which was largely unaffected by antiplatelet-derived growth factor (PDGF) antibodies. HGinduced activation of Rap1b was specific, since Rap2b activation and expression of Rap2a and Rap2b were unaffected by HG. These findings indicate that hyperglycemia and HG cause an activation and up-regulation of Rap1b in renal glomeruli and in cultured MCs, which then stimulates FN synthesis. This effect appears to be PKCdependent and PDGF-independent, but involves B-Raf, suggesting a novel PKC-Rap1b-B-Raf pathway responsible for HG-induced increased mesangial matrix synthesis, a hallmark of diabetic nephropathy.
Diabetic nephropathy is a common complication of both type-I and -II diabetes mellitus, and it is characterized by excessive accumulation of extracellular matrix (ECM) 1 pro-teins in the renal glomerulus (1). The major ECM proteins include various types of collagens, laminin, fibronectin, and proteoglycans, and they are an integral part of the capillary basement membrane and mesangial matrix, the latter being situated in the intercapillary region of the glomerulus (2). Several animal and human studies have revealed an altered synthesis or expression of various mesangial ECM proteins, especially of fibronectin, type-I, -III, and -IV collagens, and proteoglycans in diabetic nephropathy (1,3). Similarly, in vitro cell culture studies indicate that high glucose induces an increased synthesis of various ECM proteins expressed in glomerular mesangial and epithelial cells (4). The initial events involved in high glucose induction include activation of specific diacylglycerol (DAG)-sensitive protein kinase C (PKC) and mitogen-activated protein kinase (MAPK) signaling pathways (5)(6)(7), which are apparently regulated by transforming growth factor (TGF-␤) and platelet-derived growth factor (PDGF), advanced glycosylation end products (AGEs), and reactive oxygen species (ROS) (8 -11). In addition, the Smad family of proteins, regulated by TGF-␤, also play a critical role in modulating the effecter events (12). The downstream effects include increased expression of proto-oncogenes, like c-fos and c-jun, and activation of transcription factors such as AP1, a heterodimer of c-fos and c-jun, and NFB (13), which then conceivably can affect a number of target genes, e.g. collagen and fibronectin. The mechanism(s) that are known for the high glucose-induced increased expression of fibronectin include the involvement of the PKC pathway modulated by ROS, TGF-␤, and nitric oxide (NO) (14). In addition, it is known that interaction of AGEs with its receptors (RAGE) leads to an activation of the PKC followed by that of transcription factors, which then affect the target genes and cause increased expression of fibronectin (9,15).
Fibronectin, a large glycoprotein consisting of two similar polypeptide chains, is a key component of the mesangial matrix (16). It may exist in a soluble dimeric form or as oligomers of fibronectin or a highly insoluble fibrillar form in the extracellular matrix. The latter form has been shown to modulate various biological processes such as cell adhesion, migration, and differentiation (17). Fibronectin matrix assembly is a highly ordered stepwise process, and many domains along its monomer have been shown to be important for its formation and ultimate proper incorporation into the mesangial ECM (16,17). Conceivably, excessive production and a disordered incorporation or assembly of fibronectin or of other matrix proteins, secondary to the phenotypic change in the mesangial cell or due to glycoxidative-or ROS-induced damage to the proteins, is expected to result in an abnormal accumulation of ECM in diabetic nephropathy with the evolution of Kimmelstiel-Wilson lesions that progress to glomerulosclerosis and end stage renal disease with kidney failure (1)(2)(3). The information regarding the mechanism(s) of excessive synthesis of ECM proteins is available in the literature; further insights into the mechanism(s) involved in the evolution of diabetic glomerulosclerosis still remains to be explored, which may be helpful in developing future therapies in the amelioration of this disease process.
Previously, by suppressive subtractive hybridization procedures a small GTP-binding protein, Rap1b, was found to be up-regulated in the kidneys of diabetic newborn mouse (18). Similarly, up-regulation of Rap1b was seen in diabetic rats as well (18). The up-regulation was observed in several other organs, but Rap1b expression in the kidney was increased in parallel to the alterations in the blood levels of glucose, suggesting that glucose by itself, rather than any cytokine, e.g. insulin-like growth factor, may be responsible for the increased Rap1b expression. Indeed, an increased Rap1b expression was seen in the embryonic metanephric explants exposed to high glucose ambience (18). Based on these observations the present study was undertaken to determine the effect of hyperglycemia on the expression of Rap1b in the glomerular compartment of kidneys of diabetic rats as well as to assess the effects of high glucose on the expression and activation of Rap1b in cultured rat glomerular mesangial cells (MCs) and the underlying mechanisms involved in the stimulation of fibronectin synthesis.

EXPERIMENTAL PROCEDURES
Animals and Induction of Diabetes-A diabetic state was induced in 4-week-old Spargue-Dawley rats (Harlan Company) by a single intraperitoneal injection of streptozotocin (100 mg/kg of body weight; Sigma) in a citrate buffer, pH 4.6 (18). A week later the rats with blood glucose level Ͼ200 mg/dl were regarded as having a diabetic state, and their kidneys were processed for isolation of glomeruli, Rap1b protein, and mRNA expression and immunohistochemical studies. Control rats were injected with citrate buffer only.
Immunohistochemical Staining-The kidneys from both diabetic and control rats were snap-frozen in liquid nitrogen chilled isopentane and embedded in OCT compound (Sakura, Torrance, CA). Four-micrometerthick cryostat sections were prepared. The sections were overlaid with primary monoclonal antibody directed against Rap1b at a recommended dilution of 1:50 (Santa Cruz Biotechnology, Inc.) The sections were then stained with avidin-biotin peroxidase by following the instructions provided by the vendor (Ultra Streptavidin Detection System, Signet Laboratories Inc.).
Preparation of Isolated Glomeruli-The glomeruli were isolated from kidneys of diabetic rats with different blood glucose level 5 days following the injection of streptozotocin by the differential sieving method with minor modifications (19). Briefly, cortices were dissected, and then they were then gently minced with a razor blade and pressed through 106-m pore size steel mesh screen with a spatula. The paste was collected from the bottom of the screen and was mixed with 5 volumes of sterile phosphate-buffered saline and then filtered through a 180-m pore size mesh, and finally the glomeruli were collected on the surface of a third mesh with a 75-m pore size. The purity of the glomerular preparation was examined by light microscopy.
Cell Culture Studies-Glomerular MCs were isolated from kidneys of Sprague-Dawley rats (Harlan Laboratories), and they were used to establish primary cell cultures as described previously (20). The cells were grown in Dulbecco's modified Eagle's medium/Ham's F-12 medium (Sigma) supplemented with 10% heat-inactivated fetal calf serum, 100 units/ml penicillin and streptomycin, and 0.3 unit/ml insulin. They were maintained in 75-cm 2 flasks in a humidified atmosphere of 5% CO 2 and 95% air at 37°C, and at ϳ80% confluence they were trypsinized, propagated, and utilized between passages 5 and 7. At a given passage, the cell cultures were replaced with fresh serum-free Opti-MEM medium (Invitrogen). After 48 h, the medium was supplemented with varying concentrations of D-glucose (10 -30 mM), and the quiescent mesangial cell cultures were maintained for 24 h. The Lglucose-(30 mM) and 5 mM D-glucose-treated cells served as controls.
Northern Blot Analysis-Total RNAs were prepared from both the isolated glomeruli and MCs by the acid guanidinium isothiocyanatephenol-chloroform extraction method (21). About 10 g of RNA from various experiments was subjected to 1.5% agarose gel electrophoresis containing 2.2 M formaldehyde and capillary-transferred to the Hybond N ϩ nylon membrane (Amersham Biosciences). After cross-linking of RNA to the membrane, prehybridization and hybridization of different blots was carried out with various [ 32 P]dCTP-labeled (1 ϫ 10 6 cpm/ml) cDNA probes (18,22). The cDNA probes for Rap1a, Rap2a, and Rap2b were generated from plasmid constructs provided by Martina Schmidt of Institut fur Pharmakologie, Universittsklinikum Essen, Germany (23). The Rap1b cDNA probe was generated as described previously in our laboratory (18). The fibronectin (type-III) 590-bp cDNA probe was generated by PCR using a sense (5Ј-CCC CGC CCT GGT GTC ACG GAG GCC-3Ј) and an antisense (5Ј-GGC ACT GAC GAA GAG CCC TTA CAG-3Ј) primers and single-stranded cDNA prepared from rat kidney mRNA. Following the preparation of autoradiograms, the membrane blots were stripped and re-hybridized with radiolabeled ␤-actin cDNA probe. The integrity of RNA was monitored by visualization of the intact 18 and 28 S bands on the membrane blots stained with 0.05% methylene blue.
Immunoprecipitation Studies-For glomerular immunoprecipitation, 12 diabetic rats with different blood glucose levels and 4 control normoglycemic rats received an intraperitoneal injection of [ 35 S]methionine (1 mCi/kg of body weight). After 24 h, the radiolabeled glomeruli were isolated and briefly sonicated to release the cells. The glomerular cells were collected by centrifugation at 1,000 ϫ g and processed for immunoprecipitation as described below. Similarly, control and glucose-treated MCs in culture were labeled with [ 35 S]methionine (10 Ci/ml) in a methionine-free Dulbecco's modified Eagle's medium at 37°C. After 12 h, the cells were washed with the culture medium and lysed in an immunoprecipitation buffer (IP buffer: 20 mM Tris-HCl, pH 7.4, containing 100 mM NaCl, 1 mM CaCl 2 , 1 mM MgCl 2 , 10 mM benzamidine HCl, 10 mM ⑀-amino--caproic acid, 2 mM phenylmethylsulfonyl fluoride (PMSF), and 1% Triton X-100) by vigorous shaking at 4°C for 2 h. The lysates were centrifuged at 12,000 ϫ g for 30 min at 4°C, and the protein concentration in the supernatant was determined by Bradford assay (Bio-Rad) and adjusted to 1 mg/ml. The supernatants (500 l) from control and high glucose-treated cell cultures were mixed with monoclonal anti-Rap1b antibody (Santa Cruz Biotechnology, Inc.) with gentle agitation at 4°C for 2 h, followed by addition of 50 l of 50% protein A-Sepharose 4B TM (Amersham Biosciences) and incubation extended for another 1 h at 4°C. The protein A-Sepharose TM beads were then washed with IP buffer. 25 l of 2ϫ SDS-sample buffer was added to the washed beads, boiled for 5 min, and subjected to 15% SDS-PAGE. The gels were then dried and autoradiograms prepared (18,22).
Western Blot Analysis-Protein expression of fibronectin, Rap1b, B-raf, and Raf-1 were assessed by Western blot analysis (18). For fibronectin, the culture medium was collected, and the protein concentration was adjusted to 1 mg/ml. Equal amounts (ϳ25 g) protein (control versus high glucose) were loaded into the gel wells and subjected to 5% SDS-PAGE under reducing conditions. The gel proteins were then electroblotted onto a nitrocellulose membrane. The membrane was then immersed in a blocking solution containing 5% nonfat milk in TBS-T (0.1 M Tris-HCl, pH 7.5, 0.15 M NaCl, and 0.1% Tween 20) for 60 min, followed by successive incubations with the fibronectin antibody (Invitrogen) and anti-rabbit IgG conjugated with horseradish peroxidase (Amersham Biosciences) for 1 h each at 22°C with an intermediate wash with TBS-T. Following the final wash the autoradiograms were developed using an enhanced chemiluminescence (ECL) Western blot kit (Amersham Biosciences).
For Rap1b protein expression, MCs were lysed in a lysis buffer (10 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 2 mM NaF, 1 mM PMSF, 1 g/ml leupeptin, and 2 g/ml aprotinin). The lysate was centrifuged at 12,000 ϫ g for 30 min at 4°C, and the protein concentration in the supernatant was adjusted to 1 mg/ml. The lysates from various cell cultures were subjected to 15% SDS-PAGE and transferred to nitrocellulose membranes. After blocking, the membranes were successively incubated with anti-Rap1b and anti-mouse IgG conjugated with horseradish peroxidase (Amersham Biosciences) and autoradiograms developed using the ECL system.
For the determination of B-Raf and Raf-1 protein expression, the MCs in culture flasks were immersed in 1 ml of TME lysis buffer (10 mM Tris-HCl, pH 7.5, 5 mM MgCl 2 , 1 mM EDTA, 25 mM NaF, 100 M Na 3 VO 4 , and 1 mM dithiothreitol). After freezing and thawing on ice for 30 min, the cells were scrapped from the flasks lysed by using a Dounce homogenizer for 5 min at 4°C. The homogenates were centrifuged at 1,000 ϫ g for 10 min. The supernatant was saved and centrifuged at 100,000 ϫ g for 90 min at 4°C. The pellet (membrane proteins) was solubilized with 1% Triton X-100 and protein concentration adjusted to 1 mg/ml. Equal amounts of protein (20 g) from various experiments were loaded onto the gel wells and subjected to 10% SDS-PAGE, followed by Western blot analyses using anti-B-Raf and anti-Raf-1 antibodies (Santa Cruz Biotechnology).
Generation of Rap1b Eukaryotic Expression Construct and Stable Transfectants-A full-length Rap1b cDNA in PCR II vector (18) and sense (5Ј-GGGGGGGGATCCACCATGGTCATGCGTGAGTACAAGC-T-3Ј) and antisense (5Ј-GGGGGGCTCGAGTCACTTGTCATCGTCGT-CCTT GTAGTCAAGCAGCTGACA-3Ј) primers were used to generate the expression construct by PCR. The GC clamps (GGGGGG), BamHI site (GGATCC) and Kozak's sequence (ACCATGG) were introduced into the sense primer, while the antisense primer included XhoI site (CTCGAG) and FLAG epitope (N-Asp-Tyr-Lys-Asp-Asp-Asp-Asp-Lys-C). The restriction sites are in bold, the Kozak's sequence is italicized, and the FLAG epitope is underscored. The generated PCR product was digested with BamHI and XhoI, cloned into pCR II vector (Invitrogen), gel-purified, and then subcloned into BamHI-and XhoI-digested pcDNA3.1 and designated as pcDNA3.1/Rap1b. They were then transfected into MCs using LIPOFECTAMINE TM 2000 reagent, and stable transfectants were selected by growing cells in 800 g/ml G418. The surviving clone of cells were then propagated in the presence of relatively low concentration of G418 (100 g/ml) and used for further studies. To assess the Rap1b expression in the cells, RT-PCR, immunoprecipitation, and Western blot analysis were performed (18,22).
PKC Inhibitor Studies-The MCs cells were transfected with pcDNA3.1/Rap1b as described above. Both transfected and non-transfected MCs were maintained in a fresh serum-free Opti-MEM containing 5 or 30 mM D-glucose in the medium. Various PKC inhibitors, such as calphostin C and bisindolymaleimide, were individually added into the culture medium at a concentration range of 10 -100 nM and 5-20 M, respectively. The MCs were treated for 12 h. After confirming the viability of MCs by trypan blue exclusion assay, both media and the cells were collected as described above. The MCs were processed for Rap1b and fibronectin mRNA expression by Northern blot analysis, while the medium for the fibronectin protein expression was processed by Western blot analysis. In addition, Rap1b protein expression in MCs labeled with [ 35 S]methionine was assessed by immunoprecipitation followed by SDS-PAGE autoradiography.
Rap1b Mutangenesis Studies-A full-length Rap1b cDNA was cloned into pcDNA3.1 vector lacking FLAG epitope and was generated as described above. Using this pcDNA3.1/Rap1b plasmid two mutants were generated by employing a QuikChange Site-Directed Mutagenesis kit (Stratagene) following vendor's instructions. In the first Rap1b mutant, serine 17 was substituted to asparagine and designated as Rap1b/S17N. The second mutation included substitution of threonine 61 to arginine, and the mutant was designated as Rap1b/T61R. For generation of Rap1b/N17 and Rap1b/R61, the following respective primers were used: 5Ј-GGAGGTGTTGGGAAGAATGCTCTGACTGTACA-G-3Ј and 5Ј-CCTGGATACTGCAGGAAGGGAGCAGTTTACAGCCATG-3Ј. After confirming the nucleotide sequence, the mutants were transfected into MCs and maintained overnight in an environment of 5% CO 2 in a Lab-Tek chamber slides system (Nunc). The D-glucose concentration in the medium was adjusted to either 5 mM (low glucose) or 30 mM (high glucose), and cultures were maintained for an additional 24 h. Cells were then rinsed with PBS and fixed with 3.7% paraformaldehyde in PBS, followed by incubation in 0.1 M glycine in PBS for 15 min. They were permeabilized with 0.2% Triton X-100 (w/v) for 10 min. After another PBS wash, the cells were treated with blocking solution containing 1% BSA, 0.1% Tween 20, and 5 mM MgCl 2 for 20 min. They were then successively incubated with anti-fibronectin and fluorescein isothiocyanate-conjugated anti-rabbit IgG and examined with UV light microscope. The medium from MC cultures was also collected to determine the fibronectin protein expression. In addition, MCs from the above experiments were harvested and processed for B-Raf and Raf-1 protein expression as described above.
Rap1b and Rap2b Activation Assays-A pGEX-RalGDS (RBD) plasmid encoding glutathione S-transferase (GST) fusion protein containing the RBD of RalGDS proteins was used for the activation assays, and this plasmid was a generous gift from Dr. Johannes L. Bos of the Utrecht University, the Netherlands (24 -27). After transformation, the plasmid was propagated using Escherichia coli (DH5␣) host, and recombinant protein production was induced with isopropyl-1-thio-␤-D-galactopyranoside as previously described (18,22). Following induction for 3 h, the bacteria were collected and lysed by sonication in a bacterial lysis buffer (50 mM Tris, 200 mM NaCl, 2 mM MgCl 2, 1 mM PMSF, 10% glycerol, 1% Nonidet P-40, 2 g/ml aprotinin, 1 g/ml leupeptin, 10 g/ml soybean trypsin inhibitor). After sedimenting the disrupted cells by centrifugation, the supernatant was passed through an affinity glutathione-Sepharose TM column and eluted with the buffer containing 50 mM Tris-HCl, pH 8.0, 10 mM glutathione, 0.25 mM dithiothreitol, and protease inhibitors. The eluted GST-Ral fusion protein was subjected to 12.5% SDS-PAGE to confirm the purity of the protein, following which it was dialyzed against 50 mM Tris-HCl, 100 mM NaCl, and 5% glycerol and stored at Ϫ70°C in 10% glycerol. The recombinant fusion protein was immobilized on glutathione-Sepharose 4B TM beads (Pharmacia Biosciences) to be used for the activation assays (24 -27). To assess Rap1b or Rap2b activation, the D-glucose with varying concentrations (5-40 mM), was added to the medium and MC cultures maintained for 5-60 min at 37°C, followed which the cells were harvested and disrupted with the lysis RIPA buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% SDS, 0.5% deoxycholate, 1% Nonidet P-40, 1 mM Na 3 V0 4 , 1 mM PMSF, 10 g/ml leupeptin, and 1 g/ml aprotinin) and the lysate centrifuged for 10 min at 10,000 ϫ g. The resulting supernatants were incubated with freshly prepared glutathione-Sepharose 4B TM beads coupled to recombinant protein for 1 h to allow the association of activated Rap1b or Rap2b with the effector-GST fusion protein. The beads were washed three times with the RIPA buffer, suspended in sample buffer, and subjected to 12.5% SDS-PAGE. The gel proteins were electroblotted onto nylon membranes and processed for Western blot analysis using anti-Rap1b and -Rap2b antibodies and the ECL system. The Rap1b activation was also assessed following inclusion of anti-PDGF neutralizing antibody (1-10 g/ml) in the MC culture medium. Similarly, Rap1b activation was assessed following transfection of MCs with wild type versus mutant Rap1b as described above.

Rap1b Expression in Renal Tissues and Glomeruli-Immu-
nohistochemical studies revealed a mild Rap1b expression in cortical and juxtamedullary glomeruli as well as in the inner medullary collecting tubules of the kidneys of normoglycemic rats (Fig. 1B, arrowheads). The Rap1b expression notably increased in kidneys of hyperglycemic rats, particularly in the glomerular compartment (Fig. 1A, arrowheads). To confirm the Rap1b expression in glomerular compartment of the kidney, glomeruli from hyperglycemic rats with different blood glucose levels, ranging between 200 and 400 mg/dl, were isolated by the differential sieving method and processed for Northern blot analysis and immunoprecipitation studies. By the sieving method the preparation was found to be made up of almost 100% glomeruli (Fig. 1C). Northern blot analysis of glomeruli of normoglycemic rat with blood glucose level of 120 mg/dl revealed a faint ϳ2.3-kb Rap1b mRNA transcript (Fig. 1D). In hyperglycemic rats, an increase in the Rap1b mRNA expression, proportional to the blood glucose levels (200 -400 mg/dl), was observed (Fig. 1D). The ␤-actin expression was unaffected by the blood glucose levels (Fig. 1F). Similarly, immunoprecipitation studies revealed a dose-dependent increase in the intensity of the ϳ21-kDa band, corresponding to the size of Rap1b protein, in SDS-PAGE autoradiograms (Fig. 1G), suggesting that both protein and mRNA expression increase in proportion to the rise in blood glucose levels in diabetic rats.
Effect of HG on Rap1b and Fibronectin Expression in Mouse MCs-Under basal conditions with 5 mM glucose concentration in the medium, a single faint ϳ2.3-kb Rap1b mRNA transcript was detected by Northern blot analysis ( Fig. 2A). Exposure of MCs to HG (10 -30 mM) induced a dose-dependent increase in the expression of Rap1b, with a maximal response of ϳ20-fold increase at 30 mM. The ␤-actin expression was unaffected (Fig.  2B). An increase in the protein expression of de novo synthesized Rap1b was also seen with 30 mM glucose in the culture medium by immunoprecipitation methods (Fig. 2D). To assess the specificity of HG response, MCs were exposed to 30 mM L-glucose. No increase in the expression was noted, and the intensity of the ϳ21-kDa Rap1b band was similar to that of the cells treated with 5 mM D-glucose (Fig. 2D).
In a protocol similar to the assessment of Rap1b, the effect of HG on fibronectin mRNA was also determined. At 5 mM, a faint ϳ8-kb fibronectin mRNA transcript was seen (Fig. 2E). Exposure of MCs to HG for 24 h caused a significant dose-dependent increase in the fibronectin mRNA expression, which paralleled the increase in Rap1b mRNA expression observed under similar experimental conditions (Fig. 2, A and E). MCs exposed to HG also resulted in a significant increase in fibronectin protein expression as determined by Western blot analysis (Fig. 2H). Incubation of MCs with 30 mM L-glucose had no stimulating effect on fibronectin protein expression, and the intensity of the ϳ220-kDa fibronectin band was similar to that seen in the control cells treated with 5 mM D-glucose (Fig. 2H).
Effect of Calphostin C and Bisindolylmaleide on Glucoseinduced Rap1b Expression in MCs-MCs were exposed to 5-30 mM D-glucose in the absence or presence of various concentrations of PKC inhibitors, calphostin C and bisindolylmaleide (Fig. 3). The treated cells were then processed for the isolation of total RNA and radiolabeled cellular proteins for the Northern blot analysis and immunoprecipitation procedures, respectively. An increase in the Rap1b mRNA and protein expression was observed in HG ambience (Fig. 3, A and D, lane 2 versus  lane 1). Both the PKC inhibitors, calphostin C and bisindolylmaleide, caused a dose-dependent diminution in the Rap1b expression in cells treated with HG (Fig. 3, A and D, lanes 2-5). The Rap1b expression was almost undetectable at high concentrations of PKC inhibitors, i.e. 100 nM calphostin C and 20 M bisindolylmaleide (Fig. 3, A and D, lane 5). The expression of ␤-actin remained unchanged in MCs treated either with calphostin C or bisindolylmaleide in HG ambience (Fig. 3, C and  G, lanes 1-5).
Role of PKC in HG-induced Fibronectin Synthesis in MCs Overexpressing Rap1b-First a mammalian Rap1b cDNA expression construct was synthesized by cloning the wild type Rap1b cDNA into pcDNA 3.1 and designated as pcDNA3.1/Rap1b. The MCs were then transfected with pcDNA3.1/Rap1b, and stable trans- fectants were selected by using antibiotic G418. These stable transfectants (Rap1b(VTϩ)) exhibited a 30-fold increase in Rap1b mRNA expression compared with control cells transfected with empty vector (Vector(VTϪ)), as assessed by RT-PCR analysis (Fig.  4A). The authenticity of stable transfectants was confirmed by immunoprecipitation and Western blot analyses, and both revealed a high expression of de novo synthesized (Fig. 4B) and native Rap1b proteins (Fig. 4C).
Since activation of PKC by HG has been shown to mediate the stimulation of fibronectin synthesis both in vitro and in vivo models of diabetes (28 -30), the role of PKC in HG-induced fibronectin expression in control MCs as well as MCs transfected with Rap1b was investigated. Both control MCs and Rap1b-transfected cells were exposed to 5-30 mM glucose for 24 h in the absence or presence of a PKC inhibitor calphostin C (10 -100 nM) or bisindolylmaleide (5-20 M), and fibronectin mRNA expression in the cells and protein expression in the culture medium were assessed by Northern and Western blot analyses, respectively. Similar to the previous observations (Fig. 2, E and H), HG induced a significant increase in both fibronectin mRNA and protein expression in control non-transfected MCs (Fig. 5, lane 2 versus lane 1). In control MCs, both the PKC inhibitors, calphostin C and bisindolylmalede, inhibited the HG-induced stimulation of fibronectin mRNA and protein expression in a dose-dependent manner (Fig. 5, lanes 6 -8). By contrast, both the PKC inhibitors had no significant effect in reducing HG-induced increase fibronectin mRNA and protein levels in cells overexpressing Rap1b (Fig. 5, lanes 3-5). The ␤-actin expression in MCs was unaffected by treatments with HG and PKC inhibitors (Fig. 5, C and G). These results suggest that PKC-dependent mechanism(s), potentially involving Rap1b, mediate HG-induced stimulation in fibronectin synthesis in cultured MCs, and also Rap1b may be downstream of PKC in the induction of fibronectin expression.
Effect of Rap1b Mutation in HG-induced Fibronectin Synthesis-MCs were transfected either with pcDNA3.1 empty vector alone or pcDNA3.1/Rap1b (WT) or mutant (Rap1b/S17N, Ser 3 Asn) or (Rap1b/T61R, Thr 3 Arg) in the presence of low and high glucose in the culture medium. Then cellular expression of fibronectin and that secreted in the medium were determined by immunofluorescence and Western blotting procedures, respectively. Minimal cellular expression of fibronectin was ob-

FIG. 3. Effect of PKC inhibtors on Rap1b expression in MCs exposed to HG ambience (30 mM).
Rap1b mRNA is significantly increased in the MCs exposed to HG ambience compared with cells treated with 5 mM glucose (A and E, lane 2 versus lane 1). Both the PKC inhibitors, calphostin C and bisindolylmaleide, reduce the Rap1b mRNA expression in a dose-dependent manner in MCs exposed to HG ambience (A and E, lanes [3][4][5]. The ␤-actin expression in MCs treated with in HG or PKC inhibitors (C and G) is unaffected. B and F indicate the intactness of the RNA and their loading of equal amounts of total RNA in various lanes. D and H represent the expression of [ 35 S]methionine-labeled and immunprecipitated Rap1b protein in MCs exposed to HG ambience and treated with PKC inhibitors. In parallel to mRNA expression, the protein expression increases in MCs exposed to HG ambience and decreases in a dose-dependent manner with the treatment of PKC inhibitors (D and H).  (A, D, E, and H, lanes 1 and 2). While in MCs overexpressing Rap1b, the PKC inhibitors did not significantly reduce the FN mRNA or protein expression (A, D, E, and H, lanes 3-5). The expression of ␤-actin is unaffected by high glucose treatment in both non-transfected or transfected MCs (C and G). B and F indicate the intactness of the RNAs and their loading of equal amounts of total RNA in various lanes. served at low glucose (5 mM) concentration in the medium (Fig.  6A). At high glucose (30 mM) concentration, its expression markedly increased, and it was seen diffusely distributed in the cytoplasm (Fig. 6, B versus A). A remarkable reduction in the HG-induced cellular fibronectin was observed in cells transfected with mutant Rap1b/S17N or Rap1b/T61R (Fig. 6, C and  D versus B). No distinguishable differences between the effects of these mutants on the reduction of fibronectin expression were observed. Transfection with pcDNA3.1/Rap1b (WT) resulted in a marked increase in expression of cellular fibronectin compared with the MCs exposed to HG alone (Fig. 6, F versus  B). However, transfection with pcDNA3.1 empty vector induced minimal increase in the fibronectin expression compared with the cells incubated with HG alone (Fig. 6, E versus B). Similar results were observed for the fibronectin secreted in the medium (Fig. 6G). An increased fibronectin expression was observed with the HG (Fig. 6G, lane 1 versus lane 2). A reduction in the expression was seen with transfection of the mutants (Fig. 6G, lanes 3 and 4 versus lane 2). A mild increase in the expression was observed with transfection of vector alone (E-VE) (Fig. 6G, lane 5), while a remarkable increase in the fibronectin protein expression in culture media of the MCs was seen with the transfection of wild type pcDNA3.1/Rap1b (WT) (Fig. 6G, lane 6 versus lane 2).

Effect of Rap1b Transfection on B-Raf and Raf-1 Expression-To delineate which one of the Raf pathways is involved in
Rap1b-mediated responses, expression of B-Raf and Raf-1 in MCs exposed to HG was determined. The cells were transfected either with empty pcDNA3.1 vector (E-VE), or wild type pcDNA3.1/Rap1b (WT), or with Rap1b mutants Rap1b/S17N (S17N) and Rap1b/T61R (T61R) and exposed to HG, followed by the assessment of B-Raf and Raf-1 expression by Western blot analysis (Fig. 7). In B-Raf experiments, two bands of ϳ94 and ϳ68 kDa were observed in cells transfected with empty vector (E-VE) (Fig. 7A, lane 1). The ϳ68-kDa band was faint, but was detectable. This band disappeared in cells transfected with S17N or T61R Rap1b mutants, and the upper ϳ94 kDa band was unaffected (Fig. 7A, lanes 2 and 3). The transfection of MCs with WT resulted in an increase in the intensity of both the bands (Fig. 7A, lane 4). However, there was a marked increase in the intensity of the ϳ68-kDa band, suggesting that the B-Raf pathway is affected by Rap1b-mediated responses in the presence of HG (Fig. 7A, lane 4). In contrast to the results of B-Raf, the expression of Raf-1 was unaffected. In cells transfected with empty vector (E-VE), a ϳ74-kDa band was observed (Fig.  7B, lane 1), and its intensity was unaffected by transfection of the mutants S17N or T61R (Fig. 7B, lanes 2 and 3). Similarly, no increase in the intensity of this band was observed with the transfection of WT Rap1b. Instead, a mild decrease in its intensity was observed (Fig. 7B, lane 4), suggesting that, most likely, the Raf-1 pathway is not involved and rather the Rap1b may have a negative effect on the expression of Raf-1 in the HG milieu.
Effect of High Glucose on Rap1b Activation-Rap1b activation was performed by following the method of Bos and his colleagues (23)(24)(25)(26)(27). The assay is based on the differential affinity of the RBD of RalGDS for Rap1GTP versus Rap1GDP, and thus only the activated form of Rap1bGTP will bind to RBD containing fusion protein, which then can be analyzed by SDS-PAGE and Western blotting procedures. The MCs exposed to 5-40 mM glucose resulted in a marked increase in Rap1b-GTP in a dose-dependent manner, with a ϳ7-fold stimulation at 30 mM glucose (Fig. 8A, panel a). Next, the time course of activation of Rap1b by HG was determined, and the MCs were exposed to 30 mM glucose for 0 -60 min. The time course analyses revealed a significant activation of Rap1b within 5 min of incubation with HG, which increased over a period of 60 min (Fig. 8A, panel b). Maximal ϳ25-fold stimulation in Rap1b was observed at 30 min of incubation.
Since PDGF is produced by cells in response to HG (31) and has been shown to cause an activation of Rap1b in other cell types (32), the role of PDGF in HG-induced stimulation of Rap1b was investigated. The MCs were incubated with HG (30 mM) for 30 min in the absence or presence of PDGF neutralizing antibody (1-10 g/ml), and Rap1b activation was assessed. The addition of PDGF antibody in the incubating medium had no effect in reducing Rap1b activation induced by HG irrespective of the antibody concentration used (Fig. 8A, panel c). No cytotoxicity of anti-PDGF antibody was noted as assessed by trypan blue exclusion test. These results suggest that the HGinduced activation of Rap1b is, most likely, independent of the effects of PDGF.
Specificity of Rap1b Activation by HG-To assess the specificity of Rap1b activation by HG, the cells were transfected with wild type Rap1b (WT) or mutants (S17N and T61R) or by the empty vector (E-VE). Compared with non-transfected (NT) MCs, the cells transfected with wild type Rap1b (WT) had a Whereas, transfection with Rap1b/S17N or Rap1b/T61R mutants markedly diminishes the FN expression (C and D, and G, S17N and T61R, lanes 3 and 4). significant increase in the Rap1b activation after 30 min treatment with HG (Fig. 8A, panel d, lane 4 versus lane 1). By contrast, the cells transfected with Rap1b mutants (S17N and T61R) had no significant effect on the activation of Rap1b under HG conditions, rather activity was markedly diminished (Fig. 8A, panel d, lanes 2 and 3). The cells transfected with empty vector (E-VE) showed a small increase in the activity of Rap1b (Fig. 8A, panel d, lane 5), which was reminiscent of the increase in its protein expression (Fig. 6G, lane 5).
To further define the specificity of HG response to Rap1b, the activation of Rap2b and the mRNA expression of Rap1a, Rap2a, and Rap2b were assessed in MCs incubated with 5-30 mM glucose. No significant activation of Rap2b with a transcript size of ϳ4.8 kb (33) was observed in MCs exposed to HG up to a concentration of 40 mM glucose (Fig. 8B). Exposure to HG (30 mM) also had no significant effect on the mRNA expression of Rap2a and Rap2b when compared with their expression in the MCs exposed to 5 mM glucose (Fig. 9, B and C). However, Rap1a, which exhibits ϳ80% homology to Rap1b, showed a significant increase in the mRNA levels in MCs treated with HG in a dose-dependent manner similar to Rap1b (Fig. 9A). Although there is a high degree of homology between the Rap1a and Rap1b, their transcript sizes substantially differ, i.e. ϳ2.3 and ϳ1.6 kb, respectively (34). The ␤-actin levels were unaffected by HG treatment (Fig. 9, D-F). DISCUSSION The clinical course of diabetic nephropathy is characterized by hyperfiltration, microalbuminuria progressing to overt proteinuria, and azotemia culminating into end stage renal disease. The pathologic changes that may correlate with the functional abnormalities include thickening of the glomerular basement membrane and increased deposition of mesangial matrix. During the late stages, changes in the mesangial matrix become markedly accentuated with the formation of intercapillary mesangial nodules in the renal glomerulus. These nodular lesions are believed to be the result of hyperplasia followed by hypertrophy of MCs accompanied with certain phenotypic alterations and accumulation of excessive ECM pro-teins (1-4). Since among the various cell types of the glomerulus the MCs seems to be predominantly affected by the high glucose ambience; thus in the present investigation, like in previous studies, it was utilized as the model culture system to study the mechanism(s) involved in the HG-induced activation of Rap1b and synthesis of fibronectin, an ECM protein. The utilization of in vitro MCs culture system was considered suitable for the present investigation, since our initial studies suggested that Rap1b is expressed in vivo in the glomerular compartment of the kidney, and its up-regulation is modulated by the hyperglycemic state in diabetic rats (Fig. 1). Further support for the use of in vitro model system was derived from the fact that Rap1b was found to be expressed in the cultured MCs (Fig. 2).
The Rap1b is a Ras-related (Ras-proximate) GTP-binding protein, and it belongs to a superfamily consisting many members that regulate cell proliferation, differentiation, intracellular vesicular trafficking, cytoskeletal rearrangement, cell cycle events, and glucose transport (35)(36)(37). In addition, they can modulate the expression of ECM proteins, e.g. oncogenic transformation of fibroblasts by v-Src and v-Ras is shown to be associated with down-regulation of fibronectin (38). Among the various family members, the amino acid sequence of Rap1b and Ras are homologous in their putative effector domain that includes a stretch of 32-40 amino acid residues (39). But Rap1 has been shown to antagonize the Ras functions, such as in the Ras-induced transformation of NIH 3T3 cells (40) and activation of c-Raf-1 protein kinase-dependent MAP kinase cascade in Rat-1 cells (41). The findings that there is a concomitant increased expression of Rap1b and fibronectin under high glucose ambience (Fig. 2) would support the concept of antagonistic actions of Rap1b and Ras GTPases, since overexpression of the Ras results in down-regulation of fibronectin (see above). This effect of D-glucose seems to be specific, since the cells exposed to L-glucose did not alter the expression of either of Rap1b or fibronectin (Fig. 2). The increased expression or activation of Rap1b can occur by a wide variety of stimuli, including various growth factors, phospholipase C, and second mes-  1, arrows). The band intensity of ϳ68 kDa, however, is markedly increased. This ϳ68-kDa band disappears in cells transfected with S17N or T61R Rap1b mutants, while the intensity of the upper ϳ94-kDa band is unaffected (A, lanes 2 and 3). In contrast to the results of B-Raf, the expression of the Raf-1 ϳ74-kDa band is not increased with the transfection of WT Rap1b compared with cells transfected with the empty vector (E-VE), instead, a small decrease in its intensity is observed (B, lane 4 versus lane 1, arrow). The intensity of the Raf-1 band is unaffected by transfection of the mutant S17N or T61R (B, lanes 2 and 3).
sengers, which include Ca 2ϩ , cAMP, and DAG, some of which conceivably can affect the ECM synthesis via PKC pathway (35)(36)(37). Nevertheless, to assess whether Rap1b can directly affect the expression of ECM protein like fibronectin, a Rap1b expression construct was prepared, and a stable mesangial cell line overexpressing Rap1b was generated (Fig. 4). The fact that MC transfectant overexpressing Rap1b exhibited an up-regulation of fibronectin (Fig. 5 versus Fig. 2), in contrast to previously reported effect of Ras on fibronectin expression, further strengthens the notion that their effects are antagonistic. Using the Rap1b transfectants the studies were initiated to assess whether or not the increased fibronectin expression by high glucose was mediated via PKC pathway. The role of PKC in HG-induced increased expression of ECM proteins is known, and it is believed that PKC pathway in MCs can be activated by a wide array of glucose byproducts, including hexosamine-6phosphate, diacylglycerol (DAG) and AGEs, as well other endogenous biological compounds, such as angiotensin II and ROS (6 -11, 42-44). Since these endogenous products are derived from various cellular sources, it would indicate that the involvement of PKC pathway may not be restricted to MCs, and indeed HG-induced PKC-mediated increased expression of fibronectin has been reported in endothelial cells as well as peritoneal mesothelial cells (45,46). The results obtained in this investigation also point toward the fact that HG mediates an increased expression of fibronectin via the PKC-dependent pathway, but also involving the small GTPase protein, i.e. Rap1b. The fact that PKC inhibitors reduce Rap1b mRNA and protein expression in MCs exposed to HG support such a con-tention (Fig. 3). Similar results have been reported in other systems as well, where PKC inhibitors were shown to inhibit the thrombin induced second phase Rap1b activation in blood platelets (47). The blood platelet experiments also indicated that Rap1b activation is downstream of PKC signaling pathway. To attest that Rap1b is involved downstream of the PKC pathway, transfection experiments were carried out. The treatment of MC Rap1b transfectants with PKC inhibitors (calphostin C and bisindolylmaleide) did not result in any significant change in the fibronectin expression, suggesting that the effects of Rap1b are downstream of the PKC effect (Fig. 5). To further confirm the contribution of Rap1b in HG-induced PKCdependent effects, mutagenesis experiments were carried out.
The mutagenesis experiments have been performed in the past to assess the function of both Ras and Rap1 with conflicting results. The disparity in the findings may be related to the fact that their effects may be cell-specific, and these GTPases do not bind to the same guanine nucleotide exchange factors (GEFs). For instance, the RasN17 mutant tightly binds to Ras-GEFs, while RapN17 has very little affinity to associate with Rap1-GEF (48). Moreover, the GEF for Ras is SOS, while that for Rap1 is C3G, which may account for the differential effects of Ras versus Rap1. Although C3G is the GEF for Rap1, the Rap1A(S17N) does not act as a dominant negative mutant on the C3G-mediated activation of Rap1 in vivo (48). Similarly, cell specificity is exhibited in the activation experiments where RapN17 has been shown to inhibit the activation of B-Raf by PKA in COS-7 cells, while in others, e.g. RK13 or PC12, activation could not be elucidated (49,50). Nevertheless, several studies were able to demonstrate the effects of dominant negative of Rap1N17 mutant. The mutant could block the ability of nerve growth factor to stimulate the sustained phase of ERK activation in PC12 cells (51). The Rap1N17 mutant has also been shown to interfere in the endogenous Rap1 and thus alters the Dictylostelium discoideum morphology by affecting the cytoskeleton; the effect on the latter could conceivably affect the phagocytosis as well (52). In another study, this mutant was shown to affect the Rap1 adaptor protein CrkL, as a result it inhibited the ␤ 1 -integrin-fibronectin-mediated adhesion of hematopoietic cells (53). In view of the above convincing dominant negative effects in several studies, a mouse Rap1b/ S17N mutant was constructed, and its effect on the glucosemediated increased expression of fibronectin was assessed (Fig.  6). In line with the above observations, a markedly decreased expression of fibronectin was observed in mesangial cells. Another mutant, Rap1b/T61R, was also constructed, since residue Thr 61 plays a key role in transformation suppressor activity in NIH 3T3 cells transfected with K-ras. Interestingly, this mutant also had a negative effect on the fibronectin, which supports the notion that Ras and Rap1b have antagonistic actions as overexpression of v-Ras or v-Raf leads to down-regulation of fibronectin (38). Taken together, these results suggest that both Ser 17 and Thr 61 are sites that significantly modulate the functions of Rap1b. Since there is a certain degree of overlap in Rap1b and Ras-mediated signaling events following their activation, the status of B-Raf and Raf-1 following glucose-mediated induction of Rap1b expression was determined utilizing Rap1b mutants.
There are three isoforms of Raf kinase in mammals, and they include Raf-1, A-Raf, and B-Raf, and they are non-redundant proteins serving distinct functions, since each yields a different phenotype in knock-out mice (54). At present, all three isoforms share Ras as an upstream activator and mitogen-activated protein kinase/extracellular signal-regulated kinase kinase (MEK) as a commonly accepted downstream substrate. However, since Raf-1 is ubiquitously expressed and A-Raf and B-Raf have restricted distribution, it is likely that the latter two serve specialized functions in specific locations in the mammalian system (55,56). Most of the extracellular signals, e.g. PDGF and EGF, which activate Rap1 also induce Raf-1 activation. The activation of the latter is related to the binding of Rap1 with the cysteine-rich and Ras binding domains of Raf-1, and such domain-specific interactions are believed to competitively inhibit Ras functions (57). On the other hand Rap1 exhibits an additive effect on the Ki-Ras-stimulated B-Raf activity and downstream MAPK cascade (41), suggesting that these two small G proteins, i.e. Ras and Rap1, have differing actions with the involvement of different Raf isoforms. Intriguingly, in other external stimuli, such as nerve growth factor, which affects both Ras and Rap1, they divergently modulate Raf-1 and Raf-B activities so that the early activation of ERKs is Ras-Raf-1-dependent, while sustained activation is dependent upon Rap1-B-Raf cascade (58,59). The activation of ERKs may selectively involve Rap1-B-Raf pathway as exemplified by experiments with depolarization-mediated calcium influx in PC12 cells (60). Along these lines, it seems that in the present studies with MCs the HG selectively induced the expression of B-Raf, while Raf-1 was unaffected (Fig. 7). In fact a mild negative effect on the Raf-1 expression was observed upon transfection with wild type Rap1b. The notion that HG mediates B-Raf expression was strengthened by the mutational analysis where transfection with S17N and T61R Ra1b mutants resulted in the disappearance of the ϳ68-kDa band in Rap1b autoradiograms. The dominant negative effect of the S17N mutant on the B-Raf activation has also been reported in PC12 cells subjected EGF stimulation (50). The above discussion suggests that small G-proteins are activated in non-renal cells in response to a number of growth factors, thus the next critical question that needs to be addressed is whether glucose can directly or indirectly, e.g. via growth factors, induce the activation of Rap1b (Fig. 8).
The activation of small GTPases has been the subject of many recent reviews (36,37,51,58,61) and following discussion briefly summarizes certain features relevant to the activation of Rap1. A wide range of stimuli in different cell types has been reported to activate Rap1. The stimulus may be thrombin as in platelets, in lymphocyte by B-cell antigen receptor activation, and platelet-activating factor in neutrophils. In mesenchymal fibroblasts, Rap1 activation has been shown to occur by growth factors, such as EGF and PDGF. The Rap1 activation is usually rapid, and it is largely abolished by inhibitors of phospholipase C, while sparing the activation by cAMP. The second messenger generated by phospholipase C activation includes Ca 2ϩ and DAG. Both exhibit a certain degree of cell specificity, i.e. Ca 2ϩ is involved in thrombin-induced Rap1 ac- tivation in platelets, while DAG mediates B-cell antigen receptor-induced activation. Thus, these three distinct second messengers, Ca 2ϩ , DAG, and cAMP, which activate Rap1, have been clearly identified, and they play a pivotal role in various signal transduction processes. Beside the stimuli described above, glucose could conceivably induce activation of Rap1b. Among the various second messengers, DAG may be responsible for its activation in MCs exposed to high glucose, since DAG is de novo synthesized in hyperglycemic state and, in part, also by the hydrolysis of phosphatidylcholine (62). DAG is known to activate PKC, which is followed by increased TGF-␤-mediated ECM production (1,3,4,6,8,15). This idea seems to be plausible, since fibronectin, an ECM protein, expression was reduced by PKC inhibitors in non-transfected MCs (Fig. 5). However, in addition to PKC the possibility that protein kinase A (PKA) may be involved should be considered as well, since PKA has been described to increase the transcription of type-IV collagen, another ECM protein, in MCs induced with HG (63). This transcriptional increase may be related to the phosphorylation of cAMP-responsive element (CRE)-binding protein (CREB) (64). Thus, in this scenario of complex biological signaling system (65), it is conceivable that in addition to PKC, PKA may also be involved, which could explain the increase in fibronectin expression, since this ECM protein contains CRE in its promoter region, and cAMP is known to activate Rap1. Both Rap1a and Rap1b have been shown to be phosphorylated by cAMP-dependent PKA in intact cells and cell-free systems (66). The PKA-dependent activation of B-Raf in PC12 cells has been described (58); however, the present findings suggest that transfection of MCs with Rap1b, a G-protein activated downstream of PKC, also leads to the activation of B-Raf (Fig. 7). Here, it would be of interest to determine whether PKC and cAMP-dependent PKA can exert a synergistic effect activating Rap1b, and this will certainly be the subject of future investigations. Nevertheless, this cAMP-dependent pathway may be potentially involved in the HG-mediated intracellular events, since, in contrast to Ras in the plasma membrane, the Rap1 is localized to the perinuclear Golgi region and a protein activator known as Epac, exchange protein activated by cAMP, is also co-distributed in the perinuclear region with Rap1b as well (67,68).
Like Rap1b expression its activation was dose-dependent (Fig. 8A, panel a) and rapid, i.e. occurring within minutes (Ͻ5 min) of exposure to HG (Fig. 8A, panel b). Since Rap1 is localized in the perinuclear Golgi region, that is at a distance from the plasma membrane, and the de novo synthesis of cytokines, e.g. TGF-␤, EGF or PDGF, would take some time to reach the perinuclear region, it is most likely that these cytokines may not be involved for Rap1b induction. Nevertheless, the role of PDGF in cellular proliferation linked to up-regulation of small GTPases has been elucidated in other studies, where an ϳ2fold increased expression of Rap1b was observed in response to the exposure of smooth muscle cells to PDGF (69,70). With respect to PDGF, its up-regulation has been reported in glomeruli of diabetic rats as well as mesangial cells exposed to HG (71,72). In human MCs, HG has been shown to induce an overexpression of TGF-␤ through activation of PDGF loop, which in turn leads to cellular proliferation and mesangial ECM production (31,73). Such effects were reversed in MCs treated with neutralizing anti-PDGF antibodies (31). However, in the present study the anti-PDGF antibody failed to reverse the HG-induced activation of Rap1b (Fig. 8A, panel c), suggesting that the PDGF is not involved in the Rap1b-B-Raf pathway that downstream modulates the ECM-fibronectin synthesis or expression. It is also interesting to note that like the results of Rap1b expression studies (Fig. 6) certain amino acid residues, i.e. Ser 17 or Thr 61 are also critical for HG activation of Rap1b, since overexpression of dominant negative Rap1b/S17N and/ T61R notably reduced the Rap1bGTP (Fig. 8A, panel d). Also, similar to the expression studies, the Rap1b activation was enhanced in cells overexpressing Rap1b. The HG-induced activation was GTPase-specific, since Rap2b was unaffected (Fig.  8B), suggesting that the latter may not involve the cAMPdependent PKA pathway. This is also the case in studies with platelets where Rap2B, unlike Rap1a and Rap1b, was found not to be phosphorlyated by PKA (74).
Analogous to the above findings, the expression studies revealed no change in Rap2a or Rab2b mRNA levels ( Fig. 9) in MCs subjected to HG ambience. Interestingly, Rap1a expression exhibited a dose-dependent response to HG treatment, suggesting that the homologous Rap1a and Ra1b proteins respond to the HG milieu in a similar manner and add to the list of other Ras-related GTPases, e.g. Rad and Gem (75,76), that are relevant to pathogenesis of diabetic complications. Moreover, Rad has also been shown to contribute to insulin resistance as well (77). Above all, the findings of this investigation support the notion that the small GTP-binding proteins, like Rap1b, are relevant to the pathogenesis and progression of diabetic nephropathy, which is characterized by excessive synthesis of ECM proteins, glomerulosclerosis, and ultimately renal failure (Fig. 10) (78 -80).