Vascular endothelial growth factor (VEGF) is induced by stress. We determined whether chemotherapy (genotoxic stress) could induce expression of VEGF and VEGF receptors (VEGFR) in human colorectal cancer cells. The colorectal cancer cell lines HT29, RKO, and HCT116 were acutely exposed to increasing doses of oxaliplatin or 5-fluorouracil for 2, 6, and 24 h in vitro. Expression of VEGF ligand family members, VEGFRs, and signaling intermediates was determined by reverse transcription-PCR and Northern and Western blotting. The effect of oxaliplatin on VEGF-A transcriptional activity was determined by promoter assays. Acute exposure of human colorectal cancer cells to oxaliplatin led to a marked induction of VEGF-A mRNA and protein, whereas 5-fluorouracil alone or when added to oxaliplatin did not cause a further increase in VEGF levels. VEGF-A promoter activity was induced by oxaliplatin exposure. Expression of VEGF-C, placental growth factor, VEGFR-1, and neuropilin-1 levels were also increased when cells were treated with oxaliplatin. Oxaliplatin led to an increase in Akt and Src activation in HT29 cells. In contrast, Akt activation did not change in RKO cells whereas phospho-Src and phospho-p44/42 mitogen-activated protein kinase was dramatic increased by oxaliplatin. Inhibition of Akt or Src activation with wortmannin or PP2 blocked induction of VEGF-A by oxaliplatin in HT29 or RKO cells, respectively. VEGFRs may reflect the adaptive stress responses by which tumor cells attempt to protect themselves from genotoxic stress. Neutralization of prosurvival responses with anti-VEGF therapy might explain, in part, some of the beneficial effects of anti-VEGF therapy when added to chemotherapy. [Mol Cancer Ther 2008;7(9):3064–70]

Vascular endothelial growth factor (VEGF) has been shown to be the major mediator of physiologic and pathologic angiogenesis (1). There are five distinct mammalian members of the VEGF family [VEGF-A, VEGF-B, VEGF-C, VEGF-D, and placental growth factor (PlGF)] and three structurally homologous tyrosine kinase receptors (VEGFR-1, VEGFR-2, and VEGFR-3). Neuropilin-1 (NRP-1) and neuropilin-2 (NRP-2), originally identified as nonkinase receptors for the semaphorins, also bind specific isoforms of VEGF family members (2). Although the VEGF ligands and receptors have been extensively studied in developmental and pathologic angiogenesis, VEGFRs have recently been found to be present and functional on human tumor cell lines (37).

Expression of VEGF is regulated by numerous mediators and environmental conditions, an observation reinforcing its importance in physiologic and pathologic processes. One of the most potent mediators of VEGF-A is hypoxia, a stress condition that is thought to signal cells to up-regulate VEGF expression to induce vasodilation and angiogenesis in an attempt to alleviate the hypoxic conditions in tissues. Further studies showed that other stressors can induce VEGF expression, including cell density, glucose deprivation, and low pH (810).

Although VEGF and its receptors are most commonly associated with angiogenesis, expression and activation of VEGFRs on tumor cells have been associated with increased cell survival and intracellular signaling (11).

We hypothesized that chemotherapy, as a form of stress, would induce expression levels of VEGF family ligands and receptors in colorectal cancer cells.

Reagents

Oxaliplatin (Eloxatin; Sanofi-Aventis) and 5-fluorouracil (5-FU) were purchased from the pharmacy at The University of Texas M. D. Anderson Cancer Center. Antibodies for Western blot analysis were as follows: polyclonal goat anti-VEGF-A and polyclonal goat anti-VEGF-B167/186 (R&D Systems), polyclonal goat anti-Flt-1 (VEGFR-1; EMD Biosciences), rabbit anti-VEGF-C (Zymed Laboratories), polyclonal rabbit anti-PlGF (Abcam), polyclonal rabbit anti-NRP-1 and mouse monoclonal NRP-2, PTEN (Santa Cruz Biotechnology), polyclonal phospho-Src (Tyr416), phospho-p44/42 mitogen-activated protein kinase (Thr202/Tyr204), phospho-Akt (Ser473), monoclonal phospho-p38 mitogen-activated protein kinase (Thr180/Tyr182; Cell Signaling Technology), dactinomycin (actinomycin D) and PP2 (EMD Biosciences), wortmannin (Sigma), and UO126 (Cell Signaling Technology).

Cell Lines and Culture Conditions

HT29, RKO, and HCT116 human colon cancer cells were obtained from the American Type Culture Collection. Cell lines were cultured in MEM supplemented with 10% fetal bovine serum (FBS), penicillin-streptomycin, vitamins, sodium pyruvate, l-glutamine, nonessential amino acids (Life Technologies), and HEPES (MP Biomedicals) at 37°C in 5% CO2 and 95% air. Cells were confirmed to be free of Mycoplasma using the Mycoplasma Detection Kit (Cambrex). Results from all studies were confirmed in at least three independent experiments. All the experiments were done when cells reached 50% to 60% confluence.

Western Blot Hybridization

Cells were lysed in a protein lysis buffer [20 mmol/L sodium phosphate (pH 7.4), 150 mmol/L sodium chloride, 1% Triton X-100, 5 mmol/L EDTA, 5 mmol/L phenylmethylsulfonyl fluoride, 1% aprotinin, 1 μg/mL leupeptin, and 500 μmol/L sodium orthovanadate]. Proteins were subjected to electrophoresis on polyacrylamide gels and transferred to nylon membranes (Millipore) as described previously (12). After blocking with 3% or 5% milk in 0.1% polysorbate 20 in TBS, the membranes were probed with primary antibodies in 1% or 5% milk, washed, and treated with secondary antibody labeled with horseradish peroxidase. Protein bands were visualized with a commercially available chemiluminescence kit (Amersham Biosciences). Recombinant human VEGF165, VEGF-B, VEGF-C, and PlGF protein (R&D Systems) were used as a positive controls.

Isolation of mRNA and Northern Blot Analysis

Total RNA was extracted from 60% to 70% confluent tumor cells growing in culture using TRIzol reagent according to the manufacturer's protocol (Life Technologies). Northern blotting analysis was done as described previously (12). Each complementary deoxyribonucleic acid probe was radiolabeled with [α-32P] deoxyribonucleotide triphosphate by random-primer technique using a commercially available kit (Amersham Biosciences). After prehybridization of blottings for 3 to 4 h at 65°C in rapid hybridization buffer (Amersham Biosciences), the membranes were hybridized overnight at 65°C with the cDNA probe for VEGF. The probed nylon membranes were washed at 65°C with 30 mmol/L sodium chloride, 3 mmol/L sodium citrate (pH 7.2), and 0.1% SDS. Autoradiography was then done with GAPDH as an internal loading control.

Reverse Transcription-PCR

Reverse transcription-PCR was done as described previously (13). Briefly, PCR amplification of VEGFR-2 and VEGFR-3 was done under the following conditions: 95°C for 5 min, 35 cycles of 30 s denaturation at 95°C, 30 s annealing at 57°C, and 1 min extension at 72°C. Products were analyzed by electrophoresis of 20 μL of each PCR reaction mixture in a 1.5% agarose gel, and bands were visualized by ethidium bromide staining. The following primers were used: VEGFR-2 primer set 1 (402 bp): sense primer (5′-CATCACATCCACTGGTATTGG-3′) and antisense primer (5′-GCCAAGCTTGTACCATGTGAG-3′), VEGFR-2 primer set 2 (777 bp): sense primer (5′-CTGGCGGCACGAAATATCCTCTTA-3′) and antisense primer (5′-GGGCACCATTCCACCAAAAGAT-3′), and VEGFR-3 (381 bp): sense primer (5′-CCCACGCAGACATCAAGACG-3′) and antisense primer (5′-TGCAGAACTCCACGATCACC-3′).

Effect of Colorectal Cancer-Derived Conditioned Medium on Oxaliplatin Treatment

Conditioned medium from HT29, RKO, and HCT116 was prepared as follows: cells were grown to 50% to 60% confluence and treated with or without oxaliplatin for 6 and 24 h in 1% MEM-FBS. The conditioned medium was collected, centrifuged, and concentrated 10-fold using Centricon centrifugal filters (Millipore) according to the manufacturer's instructions.

VEGF Transcriptional Activity and mRNA Half-life Studies in Response to Oxaliplatin

The role of transcriptional regulation of VEGF by oxaliplatin was examined using transient transfection with a VEGF promoter-luciferase reporter construct. The following plasmids were used: pGL3-VEGF (containing the human VEGF promoter linked to the firefly luciferase reporter gene) and pRLTK (internal control containing the herpes simplex thymidine kinase promoter linked to a constitutively active Renilla luciferase reporter gene; ref. 14). Because efficiency of transfection in HT29 cells is <5%, we used RKO cells for this study. RKO (0.1 × 105) cells were plated per well in 24-well plates supplemented with complete medium. Immediately before transfection, 300 μL fresh medium was added to each well. RKO cells were cotransfected with 1 μg pGL3-VEGF and 1 μg pRLTK (control for transfection efficiency); cotransfection of pGL3 and pRLTK was used as a negative control using FuGENE (Roche Diagnostics) as indicated by the manufacturers. Cells were incubated in the transfection medium for 24 h; cells were then incubated in 1% MEM-FBS for overnight serum starvation at 37°C. Cells were then treated with oxaliplatin (0.2 and 2 μmol/L) for 8 h in 1% MEM-FBS, and lysates were obtained. Cells were harvested with passive lysis buffer Dual-Luciferase Reporter Assay System (Promega), and luciferase activity was determined. To further investigate the role of transcription on VEGF mRNA induction, transcriptional activity in HT29 and HCT116 was blocked with actinomycin D (5 μg/mL) for 1 h before treatment with oxaliplatin. The respective cells were harvested 24 and 6 h after treatment. Total RNA was then extracted and analyzed by Northern blotting.

To investigate the effect of oxaliplatin on the half-life of VEGF mRNA, HT29 cells were incubated for 24 h with or without oxaliplatin. Further transcription was blocked by the addition of actinomycin D (5 μg/mL). Total RNA was extracted from the cells at 0, 1, 2, 3, 6, and 8 h after the addition of actinomycin D, and Northern blotting analysis was done. The half-life of VEGF mRNA was determined by plotting relative VEGF mRNA expression levels on a semilogarithmic axis versus time (Microsoft Excel v2003-SP2; Microsoft).

Effect of Oxaliplatin on Signaling Pathways Involved in VEGF Induction

To determine the signaling intermediates activated by oxaliplatin, cells were grown for various time points after exposure to oxaliplatin. Cells were lysed for analyses of signaling intermediates including phosphorylated and total extracellular signal-regulated kinase 1/2 (p44/p42), P38, Jun NH2-terminal kinase mitogen-activated protein kinase, Akt, and Src. PTEN levels were also examined.

Effect of Akt or Src Inhibition on Oxaliplatin Induction of VEGF

To examine the specific role of Akt or Src on VEGF induction, HT29 or RKO cells were pretreated with wortmannin (200 nmol/L) or PP2 (10 μmol/L) for 1 h before oxaliplatin exposure (2 μmol/L) in 1% MEM-FBS. Cells were then added with or without oxaliplatin (2 μmol/L) and incubated for 24 h. Conditioned medium was collected after 24 h, centrifuged, and concentrated and 10-fold VEGF levels were determined by Western blotting. Cell viability was determined by MTT assay.

Extraction of Nuclear Proteins

HT29 and RKO cells (60% confluent) were incubated in 1% MEM-FBS overnight. Cells were then treated with or without oxaliplatin (2 μmol/L) for 4 and 24 h. Nuclear protein extraction was done according to the manufacturer's protocol (Active Motif) and Western blotting was done for the HIF-1α and NF-κB p65 subunit.

Statistical Analyses

All statistical analyses were done with InStat Statistical Software (version 2.03; GraphPad Software) and Microsoft Excel (v2003-SP2; Microsoft). P values < 0.05 were considered to be statistically significant.

Densitometric Quantification

Densitometric analysis of autoradiographs were done by using NIH Image Analysis software (V1.62) from the NIH to quantify the results of Northern and Western blotting analyses.

Effect of Chemotherapy on VEGF Expression in Colorectal Cancer Cell Lines

HT29 cells were treated for 2, 6, and 24 h with increasing concentrations of oxaliplatin or 5-FU to achieve the clinically relevant levels of 2 μmol/L or 2 μg/mL, respectively. Oxaliplatin treatment resulted in a 3-fold up-regulation of VEGF mRNA expression at clinically relevant dose of oxaliplatin (2 μmol/L) at 24 h as determined by Northern blotting (Fig. 1A). Exposure of cells to oxaliplatin for shorter durations (2 and 6 h) and to 5-FU at all time points did not lead to changes in VEGF mRNA expression. The combination of 5-FU and oxaliplatin did not cause any further increase of VEGF levels in HT29 cells compared with oxaliplatin alone at any time point (data not shown). Therefore, further studies focused on the use of oxaliplatin. Cells were exposed to oxaliplatin (0.2 and 2 μmol/L) for 24 h, and both whole-cell lysates and conditioned medium were collected and analyzed by Western blotting. VEGF-A protein levels in conditioned medium were increased nearly 4-fold following oxaliplatin treatment. Exposure of cells to 5-FU (0.2 and 2 μg/mL) did not increase VEGF-A protein levels (data not shown). To confirm that this finding was not restricted to a single cell line, we examined VEGF expression in two additional colorectal cancer cell lines, RKO and HCT116. Similar to results with HT29 cells, oxaliplatin led to a more than 2-fold increase in VEGF mRNA expression at 6 h in both additional cell lines. Analysis of conditioned medium by Western blotting showed similar results to those of HT29 cells; oxaliplatin led to a 3- to-5-fold increase in VEGF protein levels in RKO and HCT116 cells (Fig. 1B and C).

Figure 1.

Effect of oxaliplatin on VEGF expression in colorectal cancer cells. A, HT29 cells were treated with oxaliplatin (0.2 and 2 μmol/L) in 1% MEM-FBS. Oxaliplatin significantly up-regulated VEGF mRNA and protein levels at 24 h as determined by Northern blotting and Western blotting analyses, respectively. B and C, similar results were found in RKO and HCT116 cells.

Figure 1.

Effect of oxaliplatin on VEGF expression in colorectal cancer cells. A, HT29 cells were treated with oxaliplatin (0.2 and 2 μmol/L) in 1% MEM-FBS. Oxaliplatin significantly up-regulated VEGF mRNA and protein levels at 24 h as determined by Northern blotting and Western blotting analyses, respectively. B and C, similar results were found in RKO and HCT116 cells.

Close modal

Effect of Oxaliplatin on VEGF-B, VEGF-C, and PlGF Expression in Colorectal Cancer Cell Lines

Because exposure to 5-FU did not change VEGF-A levels, we decided to focus on effects of oxaliplatin on cells. HT29 cells were treated with oxaliplatin (0.2 and 2 μmol/L) for 6 and 24 h. VEGF-B protein levels in HT29 cells were unaltered by oxaliplatin (data not shown). HT29 cells treated with oxaliplatin (0.2 and 2 μmol/L) for 6 and 24 h; whole-cell lysates and conditioned medium were collected and Western blotting showed a 2- to-3-fold up-regulation of VEGF-C and PlGF protein levels in conditioned medium (Fig. 2A; but not in cell lysates; data not shown). To confirm that this finding was not restricted to a single cell line, we examined expression of VEGF-B, VEGF-C, and PlGF in a second colorectal cancer cell line, RKO. Similar to results with HT29 cells, oxaliplatin increased VEGF-C and PlGF protein expression in conditioned medium 4- to-5-fold at 24 h (Fig. 2B). VEGF-B expression did not change (data not shown).

Figure 2.

Effect of oxaliplatin treatment on VEGF-C and PlGF expression in conditioned medium from colorectal cancer cells. A, HT29 cells were treated with oxaliplatin (0.2 and 2 μmol/L) in 1% MEM-FBS. Western blotting analysis showed that oxaliplatin increased VEGF-C and PlGF levels in conditioned medium. B, similar results were found in RKO cells.

Figure 2.

Effect of oxaliplatin treatment on VEGF-C and PlGF expression in conditioned medium from colorectal cancer cells. A, HT29 cells were treated with oxaliplatin (0.2 and 2 μmol/L) in 1% MEM-FBS. Western blotting analysis showed that oxaliplatin increased VEGF-C and PlGF levels in conditioned medium. B, similar results were found in RKO cells.

Close modal

Effect of Chemotherapy on VEGFR Levels in Colorectal Cancer Cell Lines

Treatment of HT29 cells with oxaliplatin (0.2 and 2 μmol/L) for 6 and 24 h led to 2-fold increase in VEGFR-1 expression at 6 h and NRP-1 expression at 24 h (Fig. 3A). VEGFR-2 and VEGFR-3 mRNA were not expressed in HT29 cells and remained undetectable after oxaliplatin treatment (Fig. 3B). To confirm that this finding was not restricted to a single cell line, we examined VEGFRs in a second colorectal cancer cell line, RKO. Similar to results with HT29 cells, expression of VEGFR-1 and NRP-1 were also up-regulated 3-fold by oxaliplatin as determined by Western blotting (Fig. 3C). NRP-2 protein levels did not change with oxaliplatin treatment in either cell line (data not shown).

Figure 3.

Effect of oxaliplatin treatment on VEGFR expression in colorectal cancer cells. A, HT29 cells were treated with oxaliplatin (0.2 and 2 μmol/L) in 1% MEM-FBS. Western blotting analysis showed that oxaliplatin treatment up-regulated VEGFR-1 and NRP-1 protein expression. B, reverse transcription-PCR analysis showed that oxaliplatin treatment did not induce expression of VEGFR-2 or VEGFR-3 in HT29 cells. Human umbilical vein endothelial cells are shown as a positive control. C, similar results were found in RKO cells.

Figure 3.

Effect of oxaliplatin treatment on VEGFR expression in colorectal cancer cells. A, HT29 cells were treated with oxaliplatin (0.2 and 2 μmol/L) in 1% MEM-FBS. Western blotting analysis showed that oxaliplatin treatment up-regulated VEGFR-1 and NRP-1 protein expression. B, reverse transcription-PCR analysis showed that oxaliplatin treatment did not induce expression of VEGFR-2 or VEGFR-3 in HT29 cells. Human umbilical vein endothelial cells are shown as a positive control. C, similar results were found in RKO cells.

Close modal

Effect of Oxaliplatin on VEGF Transcriptional Activity and mRNA Half-life

We next examined the effect of oxaliplatin on the transcriptional regulation of VEGF. As transfection efficiency in HT29 cells is very low, we used RKO to investigate the effect of oxaliplatin on VEGF promoter activity. Oxaliplatin led to a 2.5-fold increase in induction of VEGF promoter activity in RKO cells (Fig. 4). We then sought to confirm that the increase in VEGF by oxaliplatin was transcriptionally regulated. Actinomycin D was used to block transcription in HT29 cells, and this blockade completely abolished induction of VEGF mRNA by oxaliplatin. Similar results were obtained in a second cell line, HCT116 (data not shown). Additionally, we studied the effect of oxaliplatin on VEGF mRNA stability. After HT29 cells were incubated in the presence or absence of oxaliplatin, actinomycin D was added to block further transcription. The half-life of VEGF mRNA treated with oxaliplatin and actinomycin D was similar to that in untreated cells (data not shown).

Figure 4.

Effect of oxaliplatin on gene transcription of VEGF in colorectal cancer cells. To determine the effect of oxaliplatin exposure on VEGF promoter activity, RKO cells were transiently cotransfected with pGL3-VEGF (a VEGF promoter-luciferase reporter construct) and pTLRK (control for transfection efficiency). RKO cells were treated with oxaliplatin after overnight serum starvation, and luciferase activity was determined at 8 h. The activity of the VEGF promoter-reporter construct treated with oxaliplatin was approximately 2.5 times greater than controls.

Figure 4.

Effect of oxaliplatin on gene transcription of VEGF in colorectal cancer cells. To determine the effect of oxaliplatin exposure on VEGF promoter activity, RKO cells were transiently cotransfected with pGL3-VEGF (a VEGF promoter-luciferase reporter construct) and pTLRK (control for transfection efficiency). RKO cells were treated with oxaliplatin after overnight serum starvation, and luciferase activity was determined at 8 h. The activity of the VEGF promoter-reporter construct treated with oxaliplatin was approximately 2.5 times greater than controls.

Close modal

Effect of Oxaliplatin on Signal Transduction Pathways in Colorectal Cancer Cells

To identify the signaling pathways activated by oxaliplatin, we treated HT29 cells with oxaliplatin for various durations and analyzed signaling intermediates by Western blotting. Oxaliplatin induced Akt activation with peak phosphorylation occurring at 60 min and decreased PTEN. Phospho-Src was also increased by oxaliplatin exposure (Fig. 5A). In contrast, phosphorylation of extracellular signal-regulated kinase 1/2, p38, and Jun NH2-terminal kinase remained unchanged throughout the period studied in HT29 cells (data not shown). RKO cell signaling in response to oxaliplatin was different than that observed in HT29 cells. In RKO cells, Akt activation was not increased by oxaliplatin, whereas phospho-Src and phospho-extracellular signal-regulated kinase 1/2 were markedly increased. Induction of phospho-p38 was minimal (Fig. 5B).

Figure 5.

Effect of oxaliplatin on signal transduction pathways in colorectal cancer cells. A, HT29 cells were treated with oxaliplatin for 5, 15, 30, and 60 min. Western blotting showed that oxaliplatin induced Akt activation, with peak phosphorylation occurring at 60 min and decreased PTEN. Phospho-Src increased at 30 min. B, signaling in second colorectal cancer cell line, RKO, was investigated. In contrast to HT29 cells, in RKO cells, phospho-Akt did not change after oxaliplatin exposure, whereas phospho-extracellular signal-regulated kinase 1/2 and phospho-Src were dramatically increased. Phospho-p38 was minimally increased in RKO cells exposed to oxaliplatin. C, to determine whether Akt activation secondary to oxaliplatin exposure mediated the induction of VEGF in HT29 cells, cells were treated with wortmannin followed by a 24 h incubation with oxaliplatin. Western blotting analysis revealed that inhibition of Akt activation with wortmannin blunted the oxaliplatin-mediated up-regulation of VEGF protein expression. D, because Src was markedly activated by oxaliplatin in RKO cells, we determined the effect of Src inhibition on VEGF induction by oxaliplatin in this cell line. Western blotting analysis revealed that inhibition of Src activation blunted the oxaliplatin-mediated up-regulation of VEGF protein expression.

Figure 5.

Effect of oxaliplatin on signal transduction pathways in colorectal cancer cells. A, HT29 cells were treated with oxaliplatin for 5, 15, 30, and 60 min. Western blotting showed that oxaliplatin induced Akt activation, with peak phosphorylation occurring at 60 min and decreased PTEN. Phospho-Src increased at 30 min. B, signaling in second colorectal cancer cell line, RKO, was investigated. In contrast to HT29 cells, in RKO cells, phospho-Akt did not change after oxaliplatin exposure, whereas phospho-extracellular signal-regulated kinase 1/2 and phospho-Src were dramatically increased. Phospho-p38 was minimally increased in RKO cells exposed to oxaliplatin. C, to determine whether Akt activation secondary to oxaliplatin exposure mediated the induction of VEGF in HT29 cells, cells were treated with wortmannin followed by a 24 h incubation with oxaliplatin. Western blotting analysis revealed that inhibition of Akt activation with wortmannin blunted the oxaliplatin-mediated up-regulation of VEGF protein expression. D, because Src was markedly activated by oxaliplatin in RKO cells, we determined the effect of Src inhibition on VEGF induction by oxaliplatin in this cell line. Western blotting analysis revealed that inhibition of Src activation blunted the oxaliplatin-mediated up-regulation of VEGF protein expression.

Close modal

To determine whether Akt activation secondary to oxaliplatin exposure mediated the induction of VEGF, HT29 cells were treated with wortmannin, an inhibitor of phosphoinositide 3-kinase, followed by a 24 h incubation with oxaliplatin. Western blotting analysis revealed that inhibition of Akt activation blunted the oxaliplatin-mediated up-regulation of VEGF protein expression (Fig. 5C), whereas PP2, a Src inhibitor, did not (data not shown). Because Src activation was markedly increased in RKO cells, we used PP2 to block Src activation before oxaliplatin exposure. Similar to the results with wortmannin in HT29 cells, blockage of Src activity in RKO cells blocked induction of VEGF-A by oxaliplatin (Fig. 5D). Similarly, blockade of extracellular signal-regulated kinase activation with U0126 also blocked induction of VEGF by oxaliplatin (data not shown). Cell viability after exposure to oxaliplatin and signaling inhibitors decreased by 20% to 40%, as expected.

Our studies showed that acute exposure of colon cancer cells to oxaliplatin led to induction of several members of the VEGF family of ligands including VEGF-A, VEGF-C, and PlGF. As VEGF-A is the target of most therapies that are designed to inhibit VEGF signaling, we focused on this important factor for subsequent studies. We found that VEGF-A was induced by oxaliplatin at the transcriptional level, without any alteration in mRNA stability. Furthermore, oxaliplatin led to activation of the phosphoinositide 3-kinase/Akt pathway; the induction of VEGF by oxaliplatin appeared to be mediated by this pathway, as blockade of Akt activation with wortmannin blocked the induction of VEGF by oxaliplatin. Interestingly, we did not find any induction of VEGF by 5-FU, suggesting that induction of VEGF by chemotherapy is chemotherapy drug type specific.

In addition to induction of VEGF ligands, we also found that VEGFR-1 and NRP-1 in tumor cells were also induced by exposure to oxaliplatin. The recognition that VEGFRs are present on tumor cells has led to studies investigating the role of these receptors on tumor cell function. Studies on colorectal cancer cells from our laboratory have shown that VEGFR-1 mediates tumor cell migration and invasion. The role of NRP-1 on colorectal cancer cells, however, is less well defined. NRP-2, a coreceptor for VEGF tyrosine kinase receptors, mediates colorectal cancer cell migration, invasion, and cell survival. However, NRP-2 was not induced by oxaliplatin in our study. We did not find expression of VEGFR-2 or VEGFR-3 in our colorectal cancer cell lines. Although others have noted expression of these VEGFRs on colorectal cancers, we have not been able to detect these by use of multiple reverse transcription-PCR primer sets or antibodies. The induction of VEGF ligands and receptors on tumor cells simultaneous with activation of Akt and Src suggests that oxaliplatin leads to autocrine signaling in an effort to enhance cell survival pathways. However, other stress pathways such as hypoxia-inducible factor-1 and nuclear factor-κB were not activated in our studies in response to oxaliplatin exposure (data not shown).

Other investigators have shown that chemotherapy and irradiation can induce VEGF-A and other angiogenic factors in tumor cells. Human melanoma cells treated with dacarbazine led to an increase in secreted VEGF-A and interleukin-8 (15) in one study. In that study, the investigators found similar induction of the promoter activity of these two angiogenic molecules at levels similar to those we observed in our studies. In a follow-up study, those investigators showed that dacarbazine-resistant melanoma cell lines showed increased growth in vivo with increased microvessel density (16). They also showed that Raf-mitogen-activated protein kinase activation was necessary for both VEGF-A and interleukin-8 secretion and promoter activity, whereas in our studies we found that Akt was responsible for the induction of VEGF by oxaliplatin. Other studies have shown that UV irradiation or photodynamic therapy can increase VEGF secretion from keratinocytes or prostate cancer cells, respectively (17, 18). Lastly, irradiated tumor cells showed increased expression levels of VEGF. Importantly in that study, sublethal irradiation actually led to an induction of in vivo tumor growth, hypothesized to be secondary to increased secretion of VEGF (19). In contrast to all the aforementioned studies, our study investigated the role of oxaliplatin in the induction of other members of the VEGF family of ligands. Additionally, we studied the effect of acute exposure of oxaliplatin on induction of VEGFRs on tumor cells, including receptors that are hypothesized to be involved in cell survival (11, 20).

The implications of the observation that chemotherapy can induce VEGF ligands and receptors are clinically relevant. In patients with colorectal cancer, non-small cell lung cancer, and breast cancer, the addition of bevacizumab (Avastin; Genentech), a monoclonal antibody to VEGF, to specific chemotherapeutic regimens, improves the effects of chemotherapy (2123). Numerous potential mechanisms of action of anti-VEGF therapy have been proposed (24), but it is possible that the addition of anti-VEGF therapy to chemotherapeutic regimens that induce VEGF may neutralize the effects of VEGF on the tumor vasculature (angiogenesis) and on tumor cells (survival) and may thus optimize the effects of chemotherapy. Interestingly, in our studies, 5-FU did not increase VEGF secretion in colorectal cancer cells. Although it would be informative to determine whether specific chemotherapeutic regimens in patients (and mice) led to increases in circulating VEGF levels, such a determination would be exceedingly difficult, as numerous other factors (such as tumor burden, platelet counts, etc.) could also affect circulating VEGF levels. Regardless, our studies and those of others provide further support for the use of combination anti-VEGF with oxaliplatin-containing regimens in metastatic colorectal cancer. As noted previously, in addition to induction of VEGF-A, oxaliplatin led to induction of other ligands (VEGF-C) and receptors (VEGFR-1 and NRP-1). Thus, it remains to be determined whether specific approaches to inhibit the VEGF pathway will have an advantage over another approach (VEGF antibody versus tyrosine kinase inhibitors) based on these findings.

L.M. Ellis, grant support from Sanofi-Aventis, Amgen, and ImClone; consultant to ImClone Genetech. No other potential conflicts of interest were disclosed.

Grant support: NIH grants T-32 09599 (N.A. Dallas, A.D. Yang, G. van Buren, and E.R. Camp) and CA112390 (L.M. Ellis).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Pamela McAlpin and Diane Hackett (Department of Scientific Publications) and Rita Hernandez (Department of Surgical Oncology) for editorial assistance.

1
Hicklin DJ, Ellis LM. Role of the vascular endothelial growth factor pathway in tumor growth and angiogenesis.
J Clin Oncol
2005
;
23
:
1011
–27.
2
Roy H, Bhardwaj S, Yla-Herttuala S. Biology of vascular endothelial growth factors.
FEBS Lett
2006
;
580
:
2879
–87.
3
Fan F, Wey JS, McCarty MF, et al. Expression and function of vascular endothelial growth factor receptor-1 on human colorectal cancer cells.
Oncogene
2005
;
24
:
2647
–53.
4
Gray MJ, Van Buren G, Dallas NA, et al. Therapeutic targeting of neuropilin-2 on colorectal carcinoma cells implanted in the murine liver.
J Natl Cancer Inst
2008
;
100
:
109
–20.
5
Nguyen QD, Rodrigues S, Rodrigue CM, et al. Inhibition of vascular endothelial growth factor (VEGF)-165 and semaphorin 3A-mediated cellular invasion and tumor growth by the VEGF signaling inhibitor ZD4190 in human colon cancer cells and xenografts.
Mol Cancer Ther
2006
;
5
:
2070
–7.
6
Rivat C, Rodrigues S, Bruyneel E, et al. Implication of STAT3 signaling in human colonic cancer cells during intestinal trefoil factor 3 (TFF3) and vascular endothelial growth factor-mediated cellular invasion and tumor growth.
Cancer Res
2005
;
65
:
195
–202.
7
Wey JS, Fan F, Gray MJ, et al. Vascular endothelial growth factor receptor-1 promotes migration and invasion in pancreatic carcinoma cell lines.
Cancer
2005
;
104
:
427
–38.
8
Koura AN, Liu W, Kitadai Y, Singh RK, Radinsky R, Ellis LM. Regulation of vascular endothelial growth factor expression in human colon carcinoma cells by cell density.
Cancer Res
1996
;
56
:
3891
–4.
9
Shi Q, Le X, Wang B, et al. Regulation of vascular endothelial growth factor expression by acidosis in human cancer cells.
Oncogene
2001
;
20
:
3751
–6.
10
Shweiki D, Neeman M, Itin A, Keshet E. Induction of vascular endothelial growth factor expression by hypoxia and by glucose deficiency in multicell spheroids: implications for tumor angiogenesis.
Proc Natl Acad Sci U S A
1995
;
92
:
768
–72.
11
Bachelder RE, Crago A, Chung J, et al. Vascular endothelial growth factor is an autocrine survival factor for neuropilin-expressing breast carcinoma cells.
Cancer Res
2001
;
61
:
5736
–40.
12
Jung YD, Liu W, Reinmuth N, et al. Vascular endothelial growth factor is upregulated by interleukin-1β in human vascular smooth muscle cells via the p38 mitogen-activated protein kinase pathway.
Angiogenesis
2001
;
4
:
155
–62.
13
Liu W, Ahmad SA, Jung YD, et al. Coexpression of ephrin-Bs and their receptors in colon carcinoma.
Cancer
2002
;
94
:
934
–9.
14
Jung YD, Nakano K, Liu W, Gallick GE, Ellis LM. Extracellular signal-regulated kinase activation is required for up-regulation of vascular endothelial growth factor by serum starvation in human colon carcinoma cells.
Cancer Res
1999
;
59
:
4804
–7.
15
Lev DC, Ruiz M, Mills L, McGary EC, Price JE, Bar-Eli M. Dacarbazine causes transcriptional up-regulation of interleukin 8 and vascular endothelial growth factor in melanoma cells: a possible escape mechanism from chemotherapy.
Mol Cancer Ther
2003
;
2
:
753
–63.
16
Lev DC, Onn A, Melinkova VO, et al. Exposure of melanoma cells to dacarbazine results in enhanced tumor growth and metastasis in vivo.
J Clin Oncol
2004
;
22
:
2092
–100.
17
Kosharskyy B, Solban N, Chang SK, Rizvi I, Chang Y, Hasan T. A mechanism-based combination therapy reduces local tumor growth and metastasis in an orthotopic model of prostate cancer.
Cancer Res
2006
;
66
:
10953
–8.
18
Kosmadaki MG, Yaar M, Arble BL, Gilchrest BA. UV induces VEGF through a TNF-α independent pathway.
FASEB J
2003
;
17
:
446
–8.
19
Chung YL, Jian JJ, Cheng SH, et al. Sublethal irradiation induces vascular endothelial growth factor and promotes growth of hepatoma cells: implications for radiotherapy of hepatocellular carcinoma.
Clin Cancer Res
2006
;
12
:
2706
–15.
20
Bates RC, Goldsmith JD, Bachelder RE, et al. Flt-1-dependent survival characterizes the epithelial-mesenchymal transition of colonic organoids.
Curr Biol
2003
;
13
:
1721
–7.
21
Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer.
N Engl J Med
2004
;
350
:
2335
–42.
22
Miller K, Wang M, Gralow J, et al. Paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer.
N Engl J Med
2007
;
357
:
2666
–76.
23
Sandler A, Gray R, Perry MC, et al. Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer.
N Engl J Med
2006
;
355
:
2542
–50.
24
Ellis LM. Mechanisms of action of bevacizumab as a component of therapy for metastatic colorectal cancer.
Semin Oncol
2006
;
33
:
S1
–7.