Abnormalities in endothelial function, such as increased endothelial permeability to macromolecules and endothelial proliferation, may be involved in the pathogenesis of diabetic microvascular complications1. Considerable research has focused on the pathogenesis of endothelial dysfunction, but the exact mechanisms remain incompletely defined.
Vascular endothelial growth factor (VEGF) is an attractive candidate in this regard because VEGF markedly increases vascular permeability and potently stimulates the vascular endothelial cell to divide2,3. Its pathogenic role in diabetic retinopathy is widely accepted, and it predominantly mediates the neoangiogenesis associated with proliferative retinopathy4,5. Endothelial dysfunction, which is also associated with diabetic nephropathy, may play a pathogenic role in the development of diabetic renal disease6,7, but this remains speculative at this time.
In this study, we examined the effects of a high-glucose concentration in culture media on VEGF mRNA expression and protein production in rat mesangial cells. We also established whether VEGF production depends on the protein kinase C (PKC) pathway. In a human study, we determined the plasma and urine concentrations of VEGF and searched for any relationship between these values and the severity of diabetic nephropathy. To understand the role of VEGF in diabetic kidney disease further, renal biopsies from 20 patients with noninsulin-dependent diabetes mellitus (NIDDM) and 3 patients with renal cell carcinoma undergoing nephrectomy were stained with a rabbit anti-VEGF antibody.
METHODS
Rat mesangial cell culture
Glomeruli were isolated from male Sprague-Dawley rats weighing an average of 100 g. The kidney was removed and the glomeruli, isolated by differential sieving, were incubated in collagenase and trypsin-EDTA (Gibco Laboratories, Bethesda, MD, USA), as described previously8. Mesangial cells (MCs) were identified by their characteristic stellate shape in cell culture and confirmed by immunofluorescent microscopy for the presence of actin and myosin and the absence of factor VIII and cytokeratin (Synbiotics, San Diego, CA, USA). MCs were maintained in normal glucose (5.5 mmol/L) Dulbecco's Modified Eagle Medium (DMEM) supplemented with 2 mmol/LL-glutamine, 10 mmol/L N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid (HEPES), 100 U/mL penicillin, 100 U/mL streptomycin, and 17% heat-inactivated fetal bovine serum (FBS). For this study, cells in the fourth to sixth passages were incubated at 37°C in humidified 5% CO2 in air.
Experimental design
MCs were grown to subconfluence in 6-well plates in the growth media, and then cultured for 48 hours in serum-deprived medium containing 5 mmol/LD-glucose and 1% FBS, before being exposed to an experimental condition. Cells were divided into three experimental groups according to the glucose concentration: The normal glucose group (NG), served as control; the medium glucose group (MG) was incubated in 15 mmol/LD-glucose; and the high glucose group (HG) was incubated in 30 mmol/LD-glucose. As an osmotic control, MCs were cultured in media containing 5 mmol/LD-glucose and 25 mmol/LD-mannitol. MCs grown in triplicate were harvested at 0, 3, 6, and 24 hours for extraction of total RNA and protein. To evaluate whether VEGF production depends on the protein kinase C pathway, MCs were treated with 200 nmol/L calphostin C (Calbiochem, San Diego, CA, USA) for 60 minutes before changing to the 30 mmol/LD-glucose media. To determine the effect of PKC down-regulation, MCs were treated with 100 nmol/L phorbol 12-myristate 13-acetate (PMA, Sigma Chemical Corporation, St. Louis, MO, USA) for 24 hours, incubated in 30 mmol/LD-glucose for 3 hours, and harvested for total RNA and protein extraction. To avoid any confounding effects of serum on VEGF production, all experiments were performed in serum-free media.
Semiquantitative RT-PCR of VEGF transcripts
Total RNA was extracted from MCs using TRIzol LS reagent (Life Technologies, Gaithersburg, MD, USA) and the cDNA was synthesized by a reverse transcription reaction of 1 
g of total RNA with oligo-(dT) primers (Life Technologies, Gaithersburg, MD, USA). Next, 5 L cDNA was amplified by Taq DNA polymerase in a 25-L reaction volume containing 10 mmol/L Tris-HCl (pH 8.3), 50 mmol/L KCl, 1.5 mmol/L MgCl2, 0.2 mmol/L deoxyribonucleoside triphosphate, and 10 pmol of each primer. The polymerase chain reaction (PCR) profile consisted of initial denaturation at 94°C for 7 minutes, followed by 35 cycles (VEGF) or 30 cycles (
2-microglobulin [2m]) of denaturation at 94°C for 30 seconds, annealing at 58°C for 30 seconds, extension at 72°C for 90 seconds, and extension at 72°C for 7 minutes.
Two sets of rat VEGF primers, which amplify at least three splicing variants of rat VEGF mRNA (VEGF120, VEGF164, and VEGF188), were used. The first set (rVEGF1 primer), which includes introns between amplification sites from exon 3 to the 3' untranslated end, amplifies three splicing variants (VEGF120, VEGF164, and VEGF188). The expected lengths of their PCR products are 330 base pairs (bp) for VEGF120, 462 bp for VEGF164, and 514 bp for VEGF188. The nucleotide sequences of each primer are as follows: sense 5'-GACCCTGGTGGACATCTTCCAGGA-3' and antisense 5'-GGTGAGAGGTCTAGTTCCCGA-3'. The second set (rVEGF2 primer), which includes introns between amplification sites from exon 3 to exon 7, amplifies two splicing variants (VEGF164 and VEGF188). The expected lengths of their PCR products are 444 bp for VEGF164 and 372 bp for VEGF188. The nucleotide sequences of each primer are as follows: sense 5'-GACCCTGGTGGACATCTTCCAGGA-3' and anti-sense 5'-GTTTAACTCAAGCTGCCTCGC-3'. 2m was also amplified as an internal control, and the expected length of the PCR product was 188 bp. The nucleotide sequences of the primers are as follows: sense 5'-CAGATCTGTCCTTCAGCAAG-3' and antisense 5'-GGAGTAAACTGGTCCAGATG-3'.
Different numbers of PCR cycles were performed and plotted against the densitometry measurements of the resulting PCR products to define the range in which cycle number was linearly related to the amount of PCR-amplified product. Next, 5 L PCR reaction product were separated by electrophoresis through a 2% agarose gel with ethidium bromide. PCR product bands, visualized by ultraviolet light, were scanned at 300 dots per inch and densitometric analysis was done with the National Institutes of Health Image 1.61 software with available macro and linear regression analyzer. To confirm the identity of the VEGF cDNA product, each of the electrophoresed PCR bands was extracted with a DNA extraction kit (Qiagen, Valencia, CA, USA) and sequenced using an ABI automated DNA sequencing system (ABI Genetic Analyzer 310; PRISM, Branchburg Park, NJ, USA). Quantitation of the VEGF mRNA expression was corrected by 2m.
Measurement of VEGF protein using Western blot
After incubation in the different culture media, MCs were lyzed in lysis buffer and centrifuged for 15 minutes at 12,000 rpm. Next, 50 g protein was electrophoresed on 12% SDS-polyacrylamide gel and transferred to a nitro-cellulose membrane. The blot was incubated at room temperature on a shaker mechanism in 0.1 ng/L of an anti-VEGF monoclonal antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA). Detection of specific signals was performed using the ECL Western blotting detection reagents according to the manufacturer's recommendations (Amersham, Buckinghamshire, England).
Measurement of VEGF concentrations in plasma and urine of patients with diabetes
Seventy-three patients with NIDDM whose serum creatinine was less than 2.0 mg/dL were included for measurement of VEGF levels in plasma and urine according to the status of diabetic nephropathy. The amount of VEGF protein was measured by a competitive enzyme immunoassay (EIA) kit (ACCUCYTE Human VEGF; Cytimmune Science Inc., MD, USA). At that point, 50 L of a plasma or urine sample with an equal volume of diluent was dispensed in a 96-microwell plate, precoated with 25 L diluted biotinylated rabbit antihuman VEGF polyclonal antibody. The plate was incubated at room temperature for 3 hours, washed five times, and developed with 200 L of color reagent per well. The intensity of the color was measured in an enzyme-linked immunosorbent assay (ELISA) reader. The sensitivity of the ELISA for VEGF was 19.5 pg/mL.
Immunohistochemistry
This study was performed to analyze in detail the expression of VEGF protein in the diabetic kidney. Twenty renal biopsies from patients with NIDDM and three biopsies from patients with renal cell carcinoma undergoing nephrectomy were studied. In the group with diabetes, the average age of the patients was 50.9 
14 years (mean standard deviation [SD]); the average daily excretion of urine protein was 2.55 1.88 g (mean SD); and the average serum creatinine concentration was 1.56 1.12 mg/dL (mean SD). Renal tissue was obtained from percutaneous biopsy in patients with NIDDM and from tumor-free parts of the nephrectomy specimen in patients with renal cell carcinoma. Renal tissue was immediately fixed in 10% neutral buffered formalin, cast in paraffin, sliced in 3 m thick sections, and placed on a microscope slide. After removal of paraffin and dehydration in xylene and graded alcohols, the slide was immersed in distilled water. The kidney section was transferred to a 10-mmol/L citrate buffer solution for antigen retrieval at a pH of 6.0 and then microwaved for 10 min. After another water wash, 0.05% H2O2-methanol was applied for 15 minutes to block the endogenous peroxidase. The primary antibody, polyclonal anti-VEGF (Biogenex, San Ramon, CA, USA), was added at a 1:20 dilution for 2 hours at room temperature. Negative control sections were stained under identical conditions with the buffer solution substituting for the primary antibody. With an LSAB kit/HRP (DAKO, Carpinteria, CA, USA), kidney sections were sequentially treated with normal goat serum, primary antibody, link antibody, streptavidin-biotin horseradish peroxidase, and aminoethyl carbamisole (chromogen). Sections were then counterstained with Mayer's hematoxylin.
Statistics
Results were expressed as mean SD. Comparisons were made by analysis of variance (ANOVA) within groups and by Student's t-test between groups using a SPSS for Windows computer application. To analyze the correlation between urinary VEGF and various clinical parameters, simple regression analysis was used. P < 0.05 was considered statistically significant.
RESULTS
Effect of various glucose concentrations on VEGF gene expression
RT-PCR of mRNA from rat MCs using rVEGF1 primers showed three alternative splicing variants: bands of 330, 462, and 514 bp corresponding to VEGF120, VEGF164, and VEGF188, respectively Figure 1. Likewise, RT-PCR amplification using rVEGF2 primers revealed two alternative splicing variants: 377 and 439 bp fragments corresponding to VEGF164 and VEGF188, respectively Figure 1.
Figure 1.
RT-PCR of VEGF in cultured rat mesangial cells. The three alternative splicing variants encoding peptides VEGF120, VEGF164, and VEGF188 were identified using rVEGF1 primers. Two splicing variants, VEGF164 and VEGF188, were identified using rVEGF2 primers. VEGF120 and VEGF164 were expressed more highly than VEGF188, which was faintly expressed.
Full figure and legend (89K)We then addressed the effect of increasing glucose concentrations in the culture media on the level of the VEGF transcript. In the normal glucose group, low levels of VEGF isoforms were expressed, and the ratio of VEGF164 to 2m was 0.03 0.03. However, mRNA expression of VEGF was incrementally increased in a glucose concentration–dependent manner. In the medium glucose (15 mmol/LD-glucose) group, the ratio of VEGF164 to 2m was increased to 0.12 0.07 at 3 hours, 0.05 0.02 at 6 hours, and 0.03 0.01 at 24 hours. The VEGF164 to 2m ratio in the high glucose (30 mmol/LD-glucose) group was markedly increased to 0.48 0.10 at 3 hours (P = 0.016), 0.25 0.08 at 6 hours (P = 0.044), and 0.04 0.05 at 24 hours (Figure 2, Table 1).
Figure 2.
The effect of glucose concentration on mRNA expression of VEGF. VEGF expression was increased in high glucose media and reached a peak value in 30 mmol/LD-glucose at 3 hours' incubation. The expression of VEGF120 and VEGF164 was higher than that of VEGF188. NG, normal glucose; MG, medium glucose; HG, high glucose
Full figure and legend (118K)
The VEGF120 to 2m ratio was similar to that of VEGF164. In the medium glucose group, it was increased to 0.17 0.15 at 3 hours (L = 0.03) and 0.09 0.03 at 6 hours (L = 0.043). In the high glucose group, it was increased to 0.57 0.10 at 3 hours (L = 0.012) and 0.48 0.08 at 6 hours (L = 0.014) (Figure 2, Table 2). MCs exposed to 25 mmol/LD-mannitol for 3 hours showed no difference in the expression of VEGF mRNA compared with that found in the normal glucose group.
Effect of PKC inhibition on VEGF gene expression
To test whether VEGF expression is mediated by the PKC system, PMA and calphostin C were used. Treatment with calphostin C (200 nmol/L) for 1 hour before incubation in high-glucose media markedly decreased VEGF164 expression to 0.09 0.06 at 3 hours (P = 0.04) and 0.03 0.03 at 6 hours (P = 0.071) compared with no calphostin C. Similarly, VEGF120 expression was also decreased to 0.13 0.07 at 3 hours (P = 0.003) and 0.05 0.03 at 6 hours (P = 0.014). Treatment of MCs with PMA (100 nmol/L) for 24 hours in high-glucose media prevented the high glucose–induced increases in VEGF164 expression (0.08 0.05; P = 0.034) and VEGF120 expression (0.06 0.05; P = 0.035) (Figure 3, Tables 1 and 2).
Figure 3.
The effect of PKC inhibitor (calphostin C) and PKC down-regulator (PMA) on VEGF mRNA expression. VEGF mRNA expression was significantly decreased with calphostin C and PMA treatment. Mannitol had no effect on VEGF expression. 1: negative control, 2: normal glucose at 0 hour, 3: high glucose at 3 hours, 4: calphostin C in high glucose at 3 hours, 5: calphostin C in high glucose at 6 hours, 6: calphostin C in high glucose at 24 hours, 7: PMA in high glucose at 3 hours, and 8: mannitol at 3 hours.
Full figure and legend (55K)Western blot analysis of VEGF production in cultured MCs
VEGF protein in MCs was detected as a single band of approximately 42 kDa. Similar to the findings with VEGF mRNA, VEGF protein expression remained low in the normal glucose group but increased in a glucose concentration–dependent manner, peaking at 3 hours. Likewise, pretreatment with PMA or calphostin C markedly decreased production of VEGF protein in the high-glucose group. Moreover,D-mannitol had no effect on VEGF protein production at 3 h Figure 4.
Figure 4.
Western blot analysis of VEGF production. Distinct bands of VEGF dimers were detected at approximately 42 kDa. The VEGF peptide was increased in high glucose media, particularly at 3 hours' incubation. VEGF production was markedly suppressed in the calphostin C- or PMA-treated group. NG, normal glucose; MG, medium glucose; HG, high glucose; Cal, calphostin C-treated group; P, PMA-treated group; M, mannitol-treated group.
Full figure and legend (51K)VEGF concentration in plasma and urine in patients with diabetes
In the study, 73 patients with NIDDM whose serum creatinine level was less than 2.0 mg/dL were included. Patients were classified according to the status of their diabetic nephropathy (37 cases with normoalbuminuria, 23 with microalbuminuria, and 13 with overt proteinuria). Baseline clinical and biochemical characteristics of the study population are shown in Table 3. Plasma VEGF concentration was significantly higher in the patients with overt proteinuria (1219.2 463.9 pg/mL) than in patients with normoalbuminuria (937.0 429.7 pg/mL) and microalbuminuria (857.8 455.6 pg/mL) (P < 0.05). Plasma VEGF level did not differ between patients with normoalbuminuria and those with microalbuminuria.
Urinary excretion of VEGF was significantly increased according to the degree of microalbuminuria and proteinuria. Urinary VEGF concentration was 57.2 87.9 pg/mg Cr in normoalbuminuric patients, 122.7 75.1 in microalbuminuric patients, and 288.4 119.3 in overtly proteinuric patients (P < 0.05). The level of urinary VEGF was also weakly but significantly correlated with the levels of serum creatinine (r = 0.389; P = 0.001), microalbuminuria (r = 0.308; P < 0.05), proteinuria (r = 0.491; P < 0.0001), and creatinine clearance (r = -0.327; P < 0.05). In a multiple regression analysis, the 24-hour urine protein excretion and serum creatinine concentration were independently associated with the urinary VEGF level.
Immunohistochemical staining for VEGF in the diabetic kidney
In the normal kidney, VEGF protein was clearly detected in the glomerular visceral epithelial cells, evidenced by strong cytoplasmic staining, but not in mesangial and endothelial cells Figure 5a. In the renal tubule compartment, all positively stained cells showed a cytoplasmic pattern Figure 5b, which was focally increased in the distal tubular epithelial cells and faint in the proximal tubular cells. VEGF staining was markedly different in the diabetic vs. the normal kidney with changes primarily localized in the glomerular mesangium. Biopsy results showing mild changes of diabetic nephropathy stained more heavily for VEGF in the glomerular visceral epithelial and distal tubular cells than those of control kidneys Figure 5c. In contrast, the biopsy results showing advanced changes of diabetic nephropathy, such as severe mesangial sclerosis and mesangial expansion, showed markedly decreased and sometimes negative VEGF staining in globally sclerotic glomeruli Figure 5d. Interestingly however, VEGF staining in the tubule compartment, particularly the proximal segment, remained more intense in the biopsies of patients with late diabetic nephropathy than in those of early nephropathy Figure 5e.
Figure 5.
Immunohistochemistry of VEGF in control and diabetic kidneys. (A) Normal control glomeruli with positive staining for VEGF with labeling of visceral epithelial cells (short arrow). (B) Focally positive staining for VEGF in distal tubules in control kidney (P, proximal tubule; D, distal tubule, arrow). (C) Early diabetic glomeruli revealed stronger VEGF staining than control glomeruli in visceral epithelial cells (short arrow). (D) Advanced sclerotic glomeruli showed markedly decreased staining for VEGF. (E) Tubular staining for VEGF in the more advanced diabetic kidney. VEGF staining decreased in the glomerulus but markedly increased in the proximal tubular segment (arrowhead). Magnification: A, C, and D, 
400; B and E, 200.
DISCUSSION
We demonstrate that the rat mesangial cell cultured in high-glucose media transiently up-regulates its expression of VEGF mRNA and protein, which arise as three alternative splicing variants. This finding agrees with a previous report that showed that human mesangial cells are capable of producing VEGF9. Williams et al10 reported that vascular smooth muscle cells grown in high-glucose media overexpress VEGF through the PKC pathway. Considering that MCs share properties with smooth muscle cells, this raises the possibility that MCs can produce VEGF through a PKC-dependent mechanism.
The MC is apposed to the glomerular capillary endothelial cell, which possesses receptors for VEGF11,12. This anatomic relationship may allow the locally produced VEGF by MCs to directly influence glomerular endothelial cell function. Diabetes results in several pathobiologic changes, such as activation of PKC, up-regulation of cytokines and growth factors including transforming growth factor (TGF)-, and stimulation of the renin-angiotensin system13,14,15,16,17,18,19, which are all known to increase renal VEGF production. Therefore, the mesangial cell in the diabetic state may be plausibly expected to augment its production of VEGF. In the present study, VEGF expression peaked at 3 hours in the high-glucose group, and VEGF production was increased in a glucose concentration–dependent manner. This finding is similar to that of a study in which VEGF production peaked at 3 hours in vascular smooth muscle cells cultured in high-glucose solution10.
That high ambient glucose causes an early, but not sustained, release of VEGF in vitro seems to argue against a role for VEGF in diabetic nephropathy, but prolonged hyperglycemia and other consequences of the diabetic state in vivo may alter the relative balance among various growth factors and cytokines such that VEGF production may be sustained for a longer period20,21,22,23. In the present study, high glucose–induced VEGF synthesis was abolished by PKC inhibition or down-regulation, which suggests that PKC pathways are involved in VEGF production in MCs. It is thus possible that a PKC inhibitor may help to prevent overproduction of VEGF in diabetes.
Consistent with a previous study9, RT-PCR results revealed that rat MCs express mRNA splicing variants corresponding to the VEGF120, VEGF164, and VEGF188 isoforms. All three transcripts, especially those of VEGF120 and VEGF164, appear to be up-regulated by high-glucose media. VEGF120 and VEGF164 are secreted isoforms that bind with high affinity to the vascular endothelium through two specific tyrosine kinase receptors24. Although differences in the biologic activities of these different splicing variants have been reported, it remains unclear whether they perform different biologic functions in vivo12. Alternative splicing of VEGF mRNA occurs in a tissue-specific manner, and the relative abundance of each variant differs among the tissue types12,25. In the kidney, VEGF164 is the most abundant product of the VEGF gene, but in this study, glucose-induced expression of VEGF120 was slightly higher than that of VEGF164 in rat MCs.
VEGF is strongly implicated in the pathogenesis of diabetic microvascular complications because it increases vascular permeability to macromolecules and stimulates monocyte chemotaxis and tissue factor production, all of which can contribute to microvascular complications26,27. Although the role of VEGF in regulating glomerular permeability has not yet been defined, it can potentially mediate glomerular proteinuria in diabetic nephropathy. The mechanism of increased vascular permeability by VEGF may involve stimulation of collagenase production and proteolytic disruption of the endothelial basement membrane28. Antonetti et al29,30 reported that vascular permeability in experimental diabetes is associated with reduced endothelial occludin, a tight junction protein between endothelial cells.
With regard to vascular permeability, Williams et al [abstract; Diabetes 45(Suppl 2):66A, 1996] showed that an acute infusion of VEGF into experimental animals markedly increased sciatic nerve and aortic albumin permeability. It is tempting to speculate that the VEGF produced by MCs may act as a paracrine factor to influence glomerular protein permselectivity.
In this study, urinary excretion of VEGF in people with type 2 diabetes was significantly increased with progression from microalbuminuria to overt proteinuria and was correlated with the levels of serum creatinine and the degree of proteinuria. Whether the increase in urinary VEGF excretion is a primary cause or a secondary effect of diabetic nephropathy remains unanswered.
It has been reported that the kidney, particularly the glomerulus, is a major source of VEGF production in humans31,32. This study showed that the VEGF protein is localized primarily in the glomerular epithelial cells and to a lesser extent in the distal tubular epithelial cells in the normal kidney, consistent with the findings of previous reports33,34. One can speculate that VEGF, constitutively produced by the podocyte, may play some role in regulating the glomerular permeability to protein.
VEGF staining in the diabetic kidney demonstrated marked differences with respect to degree of the glomerular mesangial expansion. Interestingly, biopsies of patients who had experienced mild changes of diabetic nephropathy showed greater VEGF staining in the glomerulus than biopsies of control specimens. Although this does not establish causality, up-regulation of VEGF may contribute to the early changes of diabetic nephropathy. However, the small number of patients in this study with an early diabetic lesion precludes any definitive conclusions about the role of VEGF in this stage of the disease.
Pathologic findings differed somewhat from those of urinary VEGF levels. Even though VEGF staining in the glomerulus decreased with worsening of diabetic nephropathy, VEGF excretion in the urine increased with progression of proteinuria. Patients with early diabetic nephropathy, evidenced by microalbuminuria, may produce most of their VEGF from glomerular visceral epithelial cells. In contrast, patients with more advanced nephropathy, evidenced by global sclerosis and overt proteinuria, may produce much of their VEGF in the tubules, especially proximally, which demonstrated a marked up-regulation of VEGF staining. It is possible that as nephropathy progresses, increased tubular epithelial VEGF production more than compensates for decreased glomerular VEGF production.
The renal tubule is subject to both direct and indirect pathogenetic influences because of its position in the nephron and its reabsorptive function35,36,37. In the diabetic state, the tubule, particularly the proximal segment, is directly exposed to glomerular effluents including glucose, large amounts of protein, advanced glycosylation end products, and other substances. Many of these activate various cytokines and growth factors including TGF- and VEGF.
Conversely, tubular ischemia can result from increased metabolic demand on the tubules or reduced peritubular blood flow38. In this regard, the S3 segment of the proximal tubule may be particularly vulnerable to ischemia, which is a potent inducer of VEGF production39. Additional studies will be needed to formally address these issues.
In summary, high-glucose concentration in culture media transiently increases VEGF mRNA expression and protein production in rat MCs through a PKC-dependent mechanism. Urinary excretion of VEGF significantly increases according to the stage of diabetic nephropathy, and it also weakly but significantly correlates with the level of serum creatinine, creatinine clearance, and degree of proteinuria. Pronounced up-regulation of VEGF is found in the glomerular epithelial cell in early diabetic nephropathy, whereas a marked increase in VEGF is found in the proximal tubular cell in more advanced nephropathy.
References
| 1. | Stehouwer CD, Lambert J, Donker AJM & Hinsbergh V. Endothelial dysfunction and pathogenesis of diabetic angiopathy. Cardiovasc Res 1997; 34: 55–68. | Article | PubMed | ISI | ChemPort | |
| 2. | Connolly DT, Heuvelman DM, Nelson R, Olander JV, Eppley BL, Delfino JJ, Siegel NR, Leimgruber RM & Feder J. Tumor vascular permeability factor stimulates endothelial growth and angiogenesis. J Clin Invest 1989; 84: 1470–1478. | PubMed | ISI | ChemPort | |
| 3. | Connolly DT, Olander JV, Heuvelman D, Nelson R, Monsell R, Siegel N, Haymore BL, Leimgruber R & Feder J. Human vascular permeability factor: isolation from U937 cells. J Biol Chem 1989; 264: 20017–20024. | PubMed | ISI | ChemPort | |
| 4. | Gilbert RE, Vranes D, Berka JL, Kelly DJ, Cox A, Wu LL, Stacker SA & Cooper ME. Vascular endothelial growth factor and its receptors in control and diabetic rat eyes. Lab Invest 1998; 78: 1017–1027. | PubMed | ISI | ChemPort | |
| 5. | Mathews MK, Merges C, Mcleod S & Lutty GA. Vascular endothelial growth factor and vascular permeability changes in human diabetic retinopathy. Invest Ophthalmol Vis Sci 1997; 38: 2729–2741. | PubMed | ISI | ChemPort | |
| 6. | Stehouwer CDA & Schaper NC. The pathogenesis of vascular complications of diabetes mellitus: One voice or many? Eur J Clin Invest 1996; 26: 535–543. | Article | PubMed | ISI | ChemPort | |
| 7. | Stehouwer CDA & Donker AJM. Urinary albumin excretion and cardiovascular disease risk in diabetes mellitus: Is endothelial dysfunction the missing link? J Nephrol 1993; 6: 72–92. |
| 8. | Cha DR, Feld SM, Nast C, Lapage J & Adler SG. Apoptosis in mesangial cells induced by ionizing radiation and cytotoxic drugs. Kidney Int 1996; 50: 1565–1571. | PubMed | ISI | ChemPort | |
| 9. | Lijima K, Yoshikawa N, Connolly DT & Nakamura H. Human mesangial cells and peripheral blood mononuclear cells produce vascular permeability factor. Kidney Int 1993; 44: 959–966. | PubMed | ISI | ChemPort | |
| 10. | Williams B, Gallacher B, Pate H & Orme C. Glucose induced protein kinase C activation regulates vascular permeability factor mRNA expression and peptide production by human vascular smooth muscle cells in vitro. Diabetes 1997; 46: 1497–1503. | PubMed | ISI | ChemPort | |
| 11. | Neufeld G, Cohen T, Gengrinovitch S & Poltorak Z. Vascular endothelial growth factor (VEGF) and its receptors. FASEB J 1999; 13: 9–22. | PubMed | ISI | ChemPort | |
| 12. | Hammes HP, Lin J, Bretzel RG, Brownlee M & Breier G. Up-regulation of vascular endothelial growth factor/vascular permeability factor receptor system in experimental background retinopathy of the rat. Diabetes 1998; 47: 401–406. | PubMed | ISI | ChemPort | |
| 13. | Mauer SM, Steffes MW & Brown DM. The kidney in diabetes. Am J Med 1981; 70: 603–612. | Article | PubMed | ISI | ChemPort | |
| 14. | Donnell MP, Kasiske BL, Kim Y, Atluru D & Keane WF. Lovastatin inhibits proliferation of rat mesangial cells. J Clin Invest 1993; 91: 83–87. | PubMed | |
| 15. | Nakamura T, Fukui M, Ebihara I, Osada S, Nagaoka I, Tomino Y & Koide H. mRNA expression of growth factors in glomeruli from diabetic rats. Diabetes 1993; 42: 450–456. | PubMed | ISI | ChemPort | |
| 16. | Shankland SJ, Scholey JW, Ly H & Thai K. Expression of transforming growth factor- 1 during diabetic renal hypertrophy. Kidney Int 1994; 46: 430–442. | PubMed | ISI | ChemPort | |
| 17. | Sharma K & Ziyadeh FN. Renal hypertrophy is associated with up-regulation of TGF-1 gene expression in diabetic BB rat and NOD mouse. Am J Physiol 1994; 267: F1094–F1101. | PubMed | ISI | ChemPort | |
| 18. | Young BA, Johnson RJ, Alpers CE, Eng E, Gordon K, Floege J, Couser WG & Seidel K. Cellular events in the evolution of experimental diabetic nephropathy. Kidney Int 1995; 47: 935–944. | PubMed | ISI | ChemPort | |
| 19. | Derubertis FR & Craven DA. Activation of protein kinase C in glomerular cells in diabetes: mechanisms and potential links to the pathogenesis of diabetic glomerulopathy. Diabetes 1994; 43: 1–8. | PubMed | ISI | ChemPort | |
| 20. | Williams B, Quinn BA, Gallacher B & Lodwick D. Angiotensin II increases vascular permeability factor gene expression by human vascular smooth muscle cells. Hypertension 1995; 25: 913–917. | PubMed | ISI | ChemPort | |
| 21. | Brogi E, Wu T, Namiki A & Isner JM. Indirect angiogenic cytokines up-regulate VEGF and bFGF gene expression in vascular smooth muscle cells, whereas hypoxia up-regulates VEGF expression only. Circulation 1994; 90: 649–652. | PubMed | ISI | ChemPort | |
| 22. | Stavri GT, Zachary IC, Baskerville PA, Martin JF & Erusalimsky JD. Basic fibroblast growth factor up-regulates the expression of vascular endothelial growth factor in vascular smooth muscle cells: synergistic interaction with hypoxia. Circulation 1995; 92: 11–14. | PubMed | ISI | ChemPort | |
| 23. | Shweiki D, Itin A, Soffer D & Keshet E. Vascular endothelial growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis. Nature 1992; 359: 843–845. | Article | PubMed | ISI | ChemPort | |
| 24. | Zachary I. Vascular endothelial growth factor: How it transmits its signal. Exp Nephrol 1998; 6: 480–487. | Article | PubMed | ISI | ChemPort | |
| 25. | Bacic M, Edwards NA & Merrill MJ. Differential expression of vascular endothelial growth factor (vascular permeability factor) forms in rat tissues. Growth Factors 1995; 12: 11–15. | PubMed | ISI | ChemPort | |
| 26. | Clauss M, Gerlach H, Brett J, Wang F, Familetti PC, Pan YCE, Olander JV, Connolly DT & Stern D. Vascular permeability factor: A tumor derived polypeptide that induces endothelial cells and monocyte procoagulant activity and promotes monocyte migration. J Exp Med 1990; 172: 1535–1545. | Article | PubMed | ISI | ChemPort | |
| 27. | Clauss M, Weich H, Breier G, Knies U, Rockl W, Waltenberger J & Risau W. The vascular endothelial growth factor receptor Flt-1 mediates biological activities. J Biol Chem 1996; 271: 17629–17634. | Article | PubMed | ISI | ChemPort | |
| 28. | Unemori EN, Ferrara N, Bauer EA & Amento EP. Vascular endothelial growth factor induces interstitial collagenase expression in human endothelial cells. J Cell Physiol 1992; 153: 557–562. | Article | PubMed | ISI | ChemPort | |
| 29. | Antonetti DA, Barber AJ, Khin S, Lieth E, Tarbell JM & Gardner TW. Vascular permeability in experimental diabetes is associated with reduced endothelial occludin content: vascular endothelial growth factor decreases occludin in retinal endothelial cells. Penn State Retina Research Group.Diabetes 1998; 47: 1953–1959. | PubMed | ISI | ChemPort | |
| 30. | Antonetti DA, Barber AJ, Hollinger LA, Wolpert EB & Gardner TW. Vascular endothelial growth factor induces rapid phosphorylation of tight junction proteins occludin and zonula occludens 1.A potential mechanisms for vascular permeability in diabetic retinopathy and tumors. J Biol Chem 1999; 274: 23463–23467. | Article | PubMed | ISI | ChemPort | |
| 31. | Gröne HJ, Simon M & Gröne EF. Expression of vascular endothelial growth factor in renal vascular disease and renal allografts. J Pathol 1995; 177: 259–267. | PubMed | ISI | ChemPort | |
| 32. | Shulman K, Rosen S, Tognazzi K, Manseau EJ & Brown LF. Expression of vascular permeability factor (VPF/VEGF) is altered in many glomerular diseases. J Am Soc Nephrol 1996; 7: 661–666. | PubMed | ISI | ChemPort | |
| 33. | Monacci WT, Merrill MJ & Oldfield EH. Expression of vascular permeability factor/vascular endothelial growth factor in normal renal tissues. Am J Physiol 1993; 264: C995–C1002. | PubMed | ISI | ChemPort | |
| 34. | Simon M, Gröne HJ, Jöhren O, Kullmer J, Plate H, Risau W & Fuchs E. Expression of vascular endothelial growth factor and its receptors in human renal ontogenesis and in adult kidney. Am J Physiol 1995; 268: F240–F250. | PubMed | ISI | ChemPort | |
| 35. | Zhang SL, Filep JG, Hohman TC, Tang SS, Ingelfinger JR & Chan JSD. Molecular mechanisms of glucose action on angiotensinogen gene expression in rat proximal tubular cells. Kidney Int 1999; 55: 454–464. | Article | PubMed | ISI | ChemPort | |
| 36. | Remuzzi G, Ruggenenti P & Benigni A. Understanding the nature of renal disease progression. Kidney Int 1997; 51: 2–15. | PubMed | ISI | ChemPort | |
| 37. | Abbate M, Zoja C, Corna D, Capitanio M, Bertani T & Remuzzi G. In progressive nephropathies, overload of tubular cells with filtered proteins translates glomerular permeability dysfunction into cellular signals of interstitial inflammation. J Am Soc Nephrol 1998; 9: 1213–1224. | PubMed | ISI | ChemPort | |
| 38. | Schrier RW, Harris DC, Chan L, Shapiro JI & Caramelo C. Tubular hypermetabolism as a factor in the progression of chronic renal failure. Am J Kidney Dis 1988; 12: 243–249. | PubMed | ISI | ChemPort | |
| 39. | Takagi H, King GL, Robinson GS, Ferrara N & Aiello LP. Adenosine mediates hypoxic induction of vascular endothelial growth factor in retinal microvascular endothelial cells. Invest Ophthalmol Vis Sci 1996; 37: 2165–2176. | PubMed | ISI | ChemPort | |
Acknowledgments
This work was supported by a special grant from Korea University and presented in abstract form at the American Society of Nephrology meeting in Miami, Florida in 1999. We are grateful for the excellent technical support of Chae Seung Lim.
