Expression of vascular endothelial growth factor in renal cell carcinomas

Vascular endothelial growth factor (VEGF) is an angiogenic factor that may be involved in tumor growth and metastasis. Only a few data concerning the role of VEGF in renal cell carcinomas (RCCs) are available, and no studies have yet evaluated its prognostic value. The aim of the present study was to assess VEGF expression in a large series of renal tumors with a long follow-up, correlated with the usual histoprognostic factors and survival. VEGF immunostaining was performed on formalin-fixed, paraffin-embedded archival tissue from 74 renal carcinomas (62 conventional renal cell and 12 papillary carcinomas). Positivity of immunostaining was semi-quantitatively scored by two pathologists. Angiogenesis was evaluated by immunostaining with anti-CD34 antibodies on serial sections. Cytoplasmic VEGF expression was detected in tumor cells in 35% (26/74) of RCCs, including 18 out of the 62 (29%) conventional RCCs and 8 out of the 12 (67%) papillary carcinomas (P=0.02). In the group of conventional RCCs, VEGF expression was positively correlated with both nuclear grade (P=0.05) and size of the tumor (P=0.05). Furthermore, a significant correlation was observed between VEGF expression and microvascular count (P=0.04). Finally, cumulative survival rate was significantly lower in the group of patients with conventional RCCs expressing VEGF (log rank test, P=0.01). In the Cox model, VEGF expression was a significant independent predictor of outcome, as well as stage and nuclear grade. This study suggests that VEGF is involved in angiogenesis in conventional RCCs and appears to be a potential prognostic factor in these tumors. Received: 9 July 1999 / Accepted: 18 October 1999     

Divergent regulation of vascular endothelial growth factor and of erythropoietin gene expression in vivo

There is accumulating evidence from in vitro experiments that the gene expression of the vascular endothelial growth factor (VEGF) is, like that of the erythropoietin (EPO) gene, regulated by the oxygen tension and by divalent cations such as cobalt. Since the information about the regulation of VEGF gene expression in vivo is rather scarce, this study aimed to examine the influence of hypoxia and of cobalt on VEGF gene expression in different rat organs and to compare it with that on EPO gene expression. To this end male Sprague-Dawley rats were exposed to carbon monoxide (0.1% CO), hypoxia (8% O₂ ) or to cobalt chloride (12 and 60 mg/kg s.c.) for 6 h. mRNA levels for VEGF- 188, -164, and -120 amino acid isoforms in lungs, hearts, kidneys and livers were semiquantitated by RNase protection. For these organs we found a rank order of VEGF mRNA abundance of lung >> heart > kidney = liver. EPO mRNA levels were semiquantitated in kidneys and livers. Hypoxia, CO and cobalt increased EPO mRNA levels 60-fold, 140-fold and 5-fold, respectively, in the kidneys, and 11-fold, 11-fold and 3-fold, respectively, in the livers. None of these manoeuvres caused significant changes of VEGF mRNA in lung, heart or kidneys. Only in the livers did hypoxia lead to a significant (50%) increase of VEGF mRNA. These findings suggest that, in contrast to the in vitro situation, the expression of the VEGF gene in normal rat tissues is rather insensitive to hypoxia. In consequence, the in vivo regulation of the VEGF and the EPO genes appear to differ substantially, suggesting that the regulation of the VEGF and EPO genes may not follow the same essential mechanisms in vivo. Received: 31 July 1995/Received after revision: 20 November 1995/Accepted: 27 November 1995     

Expression of vascular endothelial growth factor induces an invasive phenotype in human squamous cell carcinomas

Inhibition of the vascular endothelial growth factor (VEGF) receptor Flk-1 has been shown to prevent invasion of experimental squamous cell carcinomas (SCC). To directly investigate the role of VEGF in tumor invasion, we stably transfected human SCC-13 cells, which are characterized by a noninvasive phenotype in vivo, with expression vectors containing murine VEGF(164) in sense (SCC/VEGF+) or antisense (SCC/VEGF-) orientation or with vector alone (SCC/vec). SCC/vec cells formed slowly growing, well-differentiated tumors with well-defined borders between tumor and stroma, after intradermal or subcutaneous injection. In contrast, SCC/VEGF+ tumors were characterized by rapid tumor growth, with small cell groups and single cells invading into the surrounding tissue, and by admixture of blood vessels and tumor cells in areas of tumor invasion. We detected an increase in tumor vessel density and size in VEGF-overexpressing tumors, resulting in a more than fourfold increase in total vascular areas. In contrast, SCC/VEGF- clones formed noninvasive, sharply circumscribed tumors with reduced vascular density. These findings demonstrate that selective VEGF overexpression was sufficient to induce tumor invasiveness, and they provide further evidence for an active role of the tumor stroma in cancer progression.     

Use of vascular endothelial cell growth factor gene transfer to enhance implantable sensor function in vivo

In the current study, we developed and validated a simple, rapid and safe in vivo model to test gene transfer and sensor function in vivo. Using the model, we tested the specific hypothesis that in vivo gene transfer of angiogenic factors at sites of biosensor implantation would induce neovascularization surrounding the sensor and thereby enhance biosensor function in vivo. As the in vivo site for testing of our gene transfer cell and biosensor function systems, the developing chorioallantoic membrane (CAM) of the embryo was utilized. Vascular endothelial cell growth factor (VEGF) was used as a prototype for angiogenic factor gene transfer. A helper-independent retroviral vector derived from Rous sarcoma virus (RSV), designated RCAS, was used for gene transfer of the murine VEGF (mVEGF) gene (mVEGF:RCAS) into the DF-1 chicken cell line (designated mVEGF:DF-1). Initially, the ability of VEGF:DF-1 cells to produce VEGF and RCAS viral vectors containing the mVEGF gene (mVEGF:RCAS) was validated in vitro and in vivo, as was the ability of the mVEGF:DF-1 cells to induce neovascularization in the ex ova CAM model. Using the system, we determined the ability of mVEGF:DF-1 cells to enhance acetaminophen sensor function in vivo, by inducing neovascularization at sites of sensor implantation in the ex ova CAM model. For these studies, acetaminophen sensors were placed on 8-day-old ex ova CAMs, followed by addition of media or cells (mVEGF:DF-1 cells or GFP:DF-1 cells) at the sites of biosensor implantation on the CAM. At 4 to 10 days after sensor placement, the biosensor function was determined by measuring sensor response to an intravenous injection of acetaminophen. Sensors implanted on CAMs with buffer or control cells (GFP:DF-1 cells) displayed no induced neovascularization around the sensor and had minimal/baseline sensor responses to intravenous acetaminophen injection (media, 133.33 +/- 27.64 nA; GFP:DF-1, 187.50 +/- 55.43 nA). Alternatively, the sensors implanted with mVEGF:DF-1 cells displayed massive neovascularization and equally massive sensor response to intravenous injection of acetaminophen (VEGF:DF-1, 1387.50 +/- 276.42 nA). These data clearly demonstrate that enhancing vessel density (i.e., neovascularization) around an implanted sensor dramatically enhances sensor function in vivo.     

Prognostic significance of vascular endothelial growth factor expression in clear cell renal cell carcinoma

The purpose of this investigation was to analyze and correlate the immunohistochemical pattern of vascular endothelial growth factor (VEGF) expression with the average of microvessel density (MVD) and other clinicopathologic parameters in clear cell renal cell carcinoma (CCRCC) in order to determine its prognostic significance. Surgical specimens of 93 CCRCC were immunohistochemically analyzed for VEGF expression, MVD with anti-CD31, and Ki 67 proliferative index. VEGF expression was recorded as the percentage of positive tumor cells (<75% and >75%) and as diffuse or perimembranous VEGF expression according to cytoplasmic distribution. Sixty-three (68%) RCC had <75% and 30 had (32%) >75% of VEGF expression. A diffuse cytoplasmic pattern of VEGF expression was found in 61(66%) RCC and a perimembranous one in 32 (34%) RCC. Statistical analysis showed that tumors with >75% of VEGF expression were characterized by lower MVD value (p=0.034), higher nuclear grade (p=0.018), and higher Ki 67 proliferation index (p=0.023). Moreover, a higher nuclear grade of tumor cells was characterized by diffuse cytoplasmic VEGF distribution (p=0.005). This tumor model did not confirm the postulated simple relationship between VEGF overexpression and angiogenesis through high microvessel count. However, the study results indicated that overexpression of VEGF was a worse histologic prognostic parameter in CCRCC.     

Vascular endothelial growth factor expression, angiogenesis, and necrosis in renal cell carcinomas

Rapidly growing tumors often develop necrosis. In the present study the expression of vascular endothelial growth factor (VEGF) was investigated and compared to microvessel density and necrosis of renal cell carcinomas. In the tumor-host interface the microvessel density was significantly increased compared to central tumor areas. Tumor necrosis was associated with a decrease of microvessel density and an increase of the VEGF protein expression within the perinecrotic rim. VEGF protein was focally upregulated in vital tumor tissue. An increase of the apoptotic rate of endothelia and vital tumor tissue in tumors with necrosis could not be detected. VEGF(121,165) mRNA was decreased in proliferatively active carcinomas compared to less proliferative tumors. Multicellular renal cell cancer spheroids as a model of chronic hypoxia developed central apoptosis but no necrosis. VEGF was upregulated in the spheroid. Tumor microvessels expressed matrix metalloproteinase -2 and -9 and an incomplete pericyte covering in comparison to tumor-free tissue indicating immature active angiogenesis. We conclude that highly proliferative renal cell carcinomas outgrow their vascular supply and develop chronic hypoxia inducing a decrease of proliferation and an increase of VEGF expression. However, chronic hypoxia does not cause significant necrosis or apoptosis. Tumor necrosis is more likely induced by acute hypoxia due to immature microvessels. Furthermore, VEGF expression associated with concomitant tumor necrosis may help identify renal cell carcinomas susceptible to antiangiogenic therapy.     

Comprehensive mutational analysis of the VHL gene in sporadic renal cell carcinoma: relationship to clinicopathological parameters

To delineate more precisely the somatic von Hippel-Lindau disease (VHL) gene alteration as well as to elucidate its etiologic role in renal tumorigenesis, we examined a total of 240 sporadic renal cell carcinomas (RCCs) for somatic VHL gene alterations by DNA-SSCP followed by sequencing, methylation-specific PCR assay, microsatellite LOH study, and Southern blot analysis. Intragenic mutation of the VHL gene was found exclusively in clear-cell or variant-type RCCs at a frequency of 51% (104/202). Hypermethylation of the VHL promoter region was detected in an additional 11 clear-cell RCCs. Microsatellite analysis demonstrated that LOH of the VHL locus was found in 140/155 (90%) informative clear-cell RCCs. The VHL gene therefore seems to be inactivated in a two-hit manner by intragenic mutation or hypermethylation plus allelic loss in clear-cell RCC. Genomic rearrangement of the VHL gene detected by Southern analysis was not found (0/216 cases); this is in contrast to germ lines in which Southern aberrations consisted of 7-19% of the mutations. Clinicopathologic data demonstrated that VHL mutation/LOH did not vary according to tumor progression in clear-cell RCC, including tumor diameter, stage, grading, distant metastasis, and lymph node metastasis. Interestingly, VHL mutation was significantly less frequent in RCCs occurring in younger (< or = 55 years) than that in older (> or = 56 years) patients. These data suggested that the inactivation of the VHL tumor-suppressor gene is a specific genetic change in clear-cell RCC, and that it may occur at an early or first step in the clear-cell tumorigenic pathway rather than as a late event.     

Expression of HGF/SF and Met protein is associated with genetic alterations of VHL gene in primary renal cell carcinomas

We analyzed the genetic alterations of VHL, HGF/SF, and Met genes and the expression pattern of HGF/SF and Met protein in 26 renal cell carcinomas (RCCs). We found five mutations of the VHL gene and frequent LOH (50%) only in non-papillary clear cell RCC. We found six cases in which the CpG island of VHL was methylated. In addition, one missense mutation of the HGF/SF gene was detected in clear cell RCC. HGF/SF and Met protein were expressed in 84.6% and 80.7% of RCCs, respectively. All of the cases with the genetic alterations of VHL or HGF/SF demonstrated strong expression of HGF/SF and Met protein in RCC cells. Statistically, genetic alterations of VHL and HGF/SF were significantly correlated with HGF/SF and Met expression (Fisher's exact test, p=0.022 and p=0.0070). Thus, these results strongly suggest that the expression of HGF/SF and Met protein is closely associated with the genetic alterations of VHL and HGF/SF in primary RCCs.     

Expression of vascular endothelial growth factor in basal cell tumours and in squamous cell carcinomas of canine skin

The expression of vascular endothelial growth factor (VEGF) was evaluated immunohistochemically in 20 basal cell tumours (BCTs) and 15 squamous cell carcinomas (SCCs) of canine skin. VEGF was identified in all the SCCs and was particularly striking in those occurring on the toe. On the other hand, VEGF was absent in the great majority of BCTs, occurring only in those of the solid type. The results suggest that presence of VEGF is a useful additional criterion for evaluating malignancy and growth potential in tumours of these types.     

Vascular endothelial growth factor receptor-2-mediated mitogenesis is negatively regulated by vascular endothelial growth factor receptor-1 in tumor epithelial cells

Vascular endothelial growth factor (VEGF) receptors are present on nonendothelial cells suggesting that VEGF may mediate nonendothelial effects during organogenesis and tumorigenesis. Here we show that VEGF receptor-1 (VEGFR-1) negatively regulates VEGFR-2-mediated proliferation via nitric oxide (NO) in an epithelial cancer cell line ECV304. Cell proliferation was assessed by [(3)H]thymidine incorporation, fluorescent-activated cell-sorting analysis, and cell number using a Coulter Counter. Total NO generated by the action of nitric oxide synthase was measured by Seivers NOA 280 Nitric Oxide Chemiluminescence Analyser. VEGF (1 ng/ml) stimulated DNA synthesis and increased ECV304 cell number in a manner that was inhibited by a neutralizing anti-VEGFR-2 antibody. In contrast, VEGF (50 ng/ml) stimulated NO release in a manner that was inhibited by functionally neutralizing anti-VEGFR-1 antibody. Blockage of the VEGFR-1 receptor signal with anti-VEGFR-1 stimulated DNA synthesis and increased cell number. Cell-cycle analysis showed that inhibition of VEGFR-1 increased the transition from G(1) to S phase whereas inhibition of VEGFR-2 blocked the VEGF-mediated transition from G(1) to S phase. Finally, the addition of NO donors suppressed both VEGF-mediated proliferation and the increase in growth after blockade of VEGFR-1. Conversely, inhibition of VEGF mediated NO release by nitric oxide synthase inhibitor, L-monomethyl-L-arginine, restored the mitogenic effect of VEGF. These findings identify a dose-dependent reciprocal regulatory mechanism for VEGF via its two receptors. It shows that VEGFR-1 induces cell cytostasis via NO and as such is a suitable target for molecular strategies suppressing tumorigenesis.     

Expression of vascular endothelial growth factor by macrophages is up-regulated in poorly vascularized areas of breast carcinomas

Angiogenesis is essential to the growth and metastasis of solid tumours. Vascular endothelial growth factor (VEGF) is a potent pro-angiogenic cytokine that is overexpressed in malignant tumours such as invasive carcinoma of the breast. The low oxygen tensions (hypoxia) present in these tumours are known to up-regulate the expression of VEGF by tumour cells. Human macrophages also respond to hypoxia by increasing their release of VEGF in vitro, although the effect of hypoxia on VEGF expression by macrophages in vivo has yet to be demonstrated. The present study compared the expression of VEGF by macrophages in areas of low and high vascularity in 24 invasive breast carcinomas (12 lobular, 12 ductal). The cellular distributions of VEGF protein, CD31 (vessels), and CD68 (macrophages) were compared in sequential sections for each tumour. In ten tumours, both tumour cells and macrophages were immunoreactive for VEGF protein. Use of non-isotopic in situ hybridization to localize VEGF mRNA showed that these cell types also expressed VEGF mRNA. No significant differences in the cellular distribution of VEGF protein were found between lobular and ductal carcinomas. In all tumours, macrophages accumulated in higher numbers in poorly vascularized than in highly vascularized areas. In VEGF-positive tumours, macrophages were immunoreactive for VEGF only in avascular areas where tumour cells also expressed VEGF. This suggests that VEGF expression by these two cell types may be regulated by the same microenvironmental stimuli in breast carcinomas. In addition, significantly more macrophages were present in poorly vascularized areas of VEGF-positive than VEGF-negative tumours. This suggests that VEGF may exert a chemotactic action on macrophages in vivo and guide their migration into avascular tumour sites.     

The impact of gender on progression of renal disease: potential role of estrogen-mediated vascular endothelial growth factor regulation and vascular protection

Male gender is associated with a more rapid progression of renal disease independent of blood pressure, dietary protein intake, or serum lipid levels. Recently, we reported a key role for the intrarenal vasculature in progressive renal disease (Kang D-H, Kanellis J, Hugo C, Truong L, Anderson S, Kerjaschki D, Schreiner GF, Johnson RJ: Role of endothelium in progressive renal disease. J Am Soc Nephrol 2002, 13:806-816). We hypothesized that estrogen-mediated preservation of the renal vasculature could account for the better renal outcome in female rats. We analyzed micro- and macrovascular changes in the 5/6 remnant kidney (RK) models both in male (n = 24) and female (n = 24) Sprague-Dawley rats up to 12 weeks after renal mass reduction. At 12 weeks, male and female RK rats had equivalent blood pressure, glomerular tuft area, and RK/body weight, but male rats showed worse renal function, proteinuria, glomerulosclerosis (%), and tubulointerstitial fibrosis. At 12 weeks peritubular capillary (PTC) EC proliferation and PTC density were higher in female RK rats whereas macrovascular changes in preglomerular vessels (smooth muscle cell proliferation, medial wall thickening, and adventitial fibrosis) were less prominent. The expression of vascular endothelial growth factor (VEGF) and VEGF type 2 receptor (flk-1) in renal cortex assessed by immunostaining were higher in female RK rats. To dissect the mechanism of sex hormone-induced vascular remodeling and VEGF regulation, we investigated the in vitro effect of 17 beta-estradiol (17 beta E, 10 nmol/L) on proliferation and VEGF expression of renal tubular cells (rat proximal tubular cells), vascular smooth muscle cells (VSMCs), and human umbilical vein endothelial cells (HUVECs). 17 beta E directly stimulated the proliferation of HUVECs, whereas it inhibited serum-induced proliferation of VSMCs. 17 beta E stimulated VEGF mRNA expression both in renal tubular cells and VSMCs. However, when cells were pretreated with a nitric oxide donor to simulate the in vivo condition, 17 beta E inhibited VEGF mRNA expression and protein release in VSMCs. In conclusion, female RK rats developed less glomerulosclerosis and renal failure compared to male RK rats in association with greater preservation of PTC and less preglomerular arteriopathy. Estrogen stimulated basal VEGF expression in renal tubular cells. We propose that estrogen may protect female rats in progressive renal disease by stimulating VEGF expression and maintaining a healthy intrarenal vasculature.     

Stromal expression of vascular endothelial growth factor correlates with tumor grade and microvessel density in mammary phyllodes tumors: a multicenter study of 185 cases

A retrospective review of 185 mammary phyllodes tumors (105 benign, 51 borderline, 29 malignant) from 4 centers was performed by immunohistochemistry to investigate the expression of vascular endothelial growth factor in the epithelial and stromal cells of mammary phyllodes tumors. The correlation of vascular endothelial growth factor with tumor grade, stromal cell nuclear pleomorphism, cellularity, mitotic rate, margin histomorphology, and the stromal microvessel density was evaluated. Vascular endothelial growth factor expression was found in the epithelium in 29% and in the stromal cells in 31% of cases. There was significant increase of vascular endothelial growth factor expression in the stromal cells with increasing degree of malignancy, but not the epithelium. Microvessel density in the stroma also showed significant correlation with tumor malignancy, and a correlation was shown with the stromal vascular endothelial growth factor expression. Statistical overlap of stromal vascular endothelial growth factor and microvessel density in predicting malignancy suggests that angiogenesis may be an effector mechanism for vascular endothelial growth factor. Assessment of stromal VEGF may be useful as an adjunctive diagnostic criterion in the histologic assessment of malignancy in phyllodes tumors.     

Plasma level and tissue expression of vascular endothelial growth factor in renal cell carcinoma: a prospective study of 50 cases

Vascular endothelial growth factor (VEGF) is the major factor involved in angiogenesis. Although it is known that one of the functions of VEGF is to regulate neovascularization in renal cell carcinomas, the relationship between the production of VEGF in tumor tissue and its concentration in blood has not yet been studied. The aims of this study were to determine, in a series of conventional renal cell carcinoma (CRCC) cases, (1) VEGF expression and VEGF pattern in tumor cells, (2) the relationship between VEGF expression/pattern and VEGF levels in plasma (pVEGF), and (3) the association with usual clinical and pathologic prognostic factors. Fifty patients operated on for CRCC by radical nephrectomy were included. Clinical and histologic parameters were studied. VEGF expression and VEGF pattern in tumor cells was immunohistochemically recorded. pVEGF levels and platelet count were analyzed in relation to clinical and histologic parameters. Intratumoral VEGF expression associated with a cytoplasmic VEGF pattern was significantly higher in patients with high pVEGF levels (P = .01). Both VEGF expression and pVEGF levels were significantly correlated with Fuhrman grade (P = .002 and P = .01, respectively) and tumor stage (P = .006 and P = .008, respectively). In addition, VEGF expression was also correlated with tumor necrosis (P = .001) and progression (P = .001). We demonstrated that in CRCC with tumor necrosis, VEGF expression, pVEGF levels, and platelet count were significantly higher than in CRCC with no tumor necrosis (P = .001, P = .03, and P = .001, respectively). Our results revealed that cytoplasmic VEGF expression and pVEGF levels are associated with usual prognostic factors and progression in CRCC, which may allow VEGF to be used as a prognostic marker for CRCC, especially in patients with VEGF-targeted therapy.     

In vitro and in vivo study of the expression vector encoding vascular endothelial growth factor

Vascular endothelial growth factor (VEGF) is an angiogenic cytokine with potential therapeutic applications in human diseases. It is a mitogen primarily for endothelial cells. The transfer of the cDNA encoding VEGF to ischemic tissues, which cannot be revascularized otherwise, represents a novel and promising approach to the treatment of vascular disorders. In this work the VEGF165 cDNA was cloned into the expression vector pSecTag2B. The activity of the construct was studied in cell culture as well as in vivo. Western blotting study showed that the cells transfected with the vector secreted significantly higher amounts of VEGF to the culture medium than the non-transfected cells. In vivo study revealed an increased number of new vessels in animals injected with vector encoding VEGF as compared with empty plasmid. Also, tumor cells transfected with the VEGF plasmid exhibited extensive vascularization.     

Human follicle fluid vascular endothelial growth factor concentrations are correlated with luteinization in spontaneously developing follicles

Vascular endothelial growth factor (VEGF) is a cytokine that induces angiogenesis. Angiogenesis is a prominent histologic component of the luteinization process. Luteinization is also characterized by granulosa cell progesterone secretion in response to the luteinizing hormone (LH) surge. Local VEGF production in human pre-ovulatory follicles, induced by LH, was postulated to be a luteinization mediator in women. To investigate this hypothesis, serum and fluid from the dominant follicle of 31 healthy regularly cycling multiparous women undergoing laparoscopic sterilization were obtained. VEGF was measured by enzyme- linked immunosorbent assay, and LH and progesterone were measured by radioimmunoassay. Follicle aspiration was performed at a median of 13 days from the last menstrual period (range 11-17 days). The median pre- ovulatory follicle diameter was 16 mm (range 11-23 mm). Follicle fluid VEGF concentrations (mean 6900 pg/ml, range 1200-17 100 pg/ml) were correlated positively with follicle fluid progesterone concentrations (mean 10 176 nmol/l, range 636-66780 nmol/l, r=0.62, P=0.002). This correlation was even tighter (r=0.87, P < 0.0001) when only samples from the 22 women in the earliest stages of follicle luteinization were considered. In these women serum LH concentrations were also correlated with follicle fluid VEGF concentrations (r=0.51, P=0.02). Our findings demonstrate the close dynamic relationship between VEGF production and early luteinization in human follicles during normal non-stimulated cycles.