Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Vascular endothelial growth factor acts in an autocrine manner in rhabdomyosarcoma cell lines and can be inhibited with all-trans-retinoic acid

Abstract

Vascular endothelial growth factor (VEGF) is a potent signalling molecule that acts through two tyrosine kinase receptors, VEGFR1 and VEGFR2. The upregulation of VEGF and its receptors is important in tumour-associated angiogenesis; however, recent studies suggest that several tumour cells express VEGF receptors and may be influenced by autocrine VEGF signalling. Rhabdomyosarcoma (RMS) is the most common paediatric soft-tissue sarcoma and is dependent on autocrine signalling for its growth. The alveolar subtype of RMS is often characterized by the presence of a PAX3-FKHR translocation and when introduced into non-RMS cells, the resultant fusion protein induces expression of VEGFR1. In our study, we examined the expression of VEGF and its receptors in RMS and autocrine effects of VEGF on cell growth. VEGF and receptor mRNA and protein were found to be expressed in RMS cells. Exogenous VEGF addition resulted in extracellular signal-regulated kinase-1/2 phosphorylation and cell proliferation and both were reduced by VEGFR1 blockade. Growth was also slowed by VEGFR1 inhibitor alone. Treatment of RMS cells with all-trans-retinoic acid decreased VEGF secretion and slowed cell growth, which was rescued by VEGF. These data suggest that autocrine VEGF signalling likely influences RMS growth and its inhibition may be an effective treatment for RMS.

Introduction

Vascular endothelial growth factor (VEGF) is an essential regulator of vasculogenesis and haematopoiesis during embryonic development (Fong et al., 1995; Shalaby et al., 1995; Ferrara, 1996; Damert et al., 2002; Risau, 1997). In adulthood, it drives angiogenesis in pregnancy and wound healing, as well as in pathological conditions including cancer, rheumatoid arthritis, ocular neovascular disorders and cardiovascular disease (Folkman, 1995; Isner and Losordo, 1999; Zachary et al., 2000). VEGF promotes vascular permeability (Keck et al., 1989) and stimulates endothelial cell migration, proliferation and survival of newly formed blood vessels (Neufeld et al., 1999; Zachary and Gliki, 2001; Claesson-Welsh, 2003). The effects of VEGF are transduced through two tyrosine kinase receptors, VEGFR1 (or fms-like tyrosine kinase receptor (Flt1)) (de Vries et al., 1992) and VEGFR2 (or kinase insert domain-containing receptor (KDR); Flk1 in mice) (Quinn et al., 1993). A third tyrosine kinase receptor, VEGFR3, does not bind VEGF, but instead binds VEGF-C and VEGF-D (Joukov et al., 1996; Orlandini et al., 1996). In addition, the effects of VEGF through VEGFRs may be modulated by neuropilin-1 (NRP1) and neuropilin-2 (NRP2), which act as coreceptors to the VEGFRs (Tammela et al., 2005). In addition to binding class 3 semaphorins, NRPs also bind the VEGF family ligands VEGF (NRP1/2), VEGF-B (NRP1), VEGF-C (NRP2) and placenta growth factor (PlGF; NRP1/2). NRP1 enhances VEGFR2 signalling and forms complexes with VEGFR1 and NRP2 is often coexpressed with VEGFR3 (Tammela et al., 2005).

VEGFR1 is likely constitutively inactive in endothelial cells (Gille et al., 2000), suggesting that VEGFR2 is the primary transducer of VEGF signalling during physiological angiogenesis (Waltenberger et al., 1994). In tumours, however, angiogenesis appears to be dependent on both VEGFR2 (Millauer et al., 1996) and VEGFR1 (Itokawa et al., 2002; Luttun et al., 2002) and tumour-associated angiogenesis is more potently inhibited by blocking both receptors rather than interrupting either alone (Lu et al., 2001). Owing to the role of VEGF in tumour angiogenesis, several approaches have been developed to interrupt its signalling. These include neutralizing antibodies to VEGF (Manley et al., 2004) and its receptors (Lu et al., 2001), soluble receptors to sequester VEGF (Holash et al., 2002), ribozymes to degrade transcripts of VEGF (Ke et al., 1998; Manley et al., 2004) and its receptors (Parry et al., 1999), compounds to decrease VEGF and receptor expression (Broggini et al., 2003) and small molecular kinase inhibitors (Manley et al., 2004).

As VEGF is a key mediator of tumour angiogenesis, most studies of VEGF and its receptors have primarily focused on endothelial expression of VEGFR1 and VEGFR2. However, growing evidence suggests that malignant cells also express VEGF receptors and are likely responsive to VEGF signalling (Dias et al., 2001; Broggini et al., 2003; Narendran et al., 2003). In leukaemia and malignant myeloma cells, VEGF autocrine loops likely aid in growth and survival (Dias et al., 2000, 2001, 2002; Podar et al., 2001, 2004). Similarly, some solid tumours may express both VEGF and its receptors and their growth and survival may be partially regulated by VEGF autocrine signalling (Bates et al., 2003; Wey et al., 2004). Recently, evidence has been presented for VEGF autocrine signalling through VEGFR1 in colon cancer (Bates et al., 2003). Bates et al. discovered that the colon cancer cell line LIM 1863 upregulates VEGF and VEGFR1 as cells progress to an invasive phenotype. Treatment with a monoclonal antibody against VEGFR1 led to diminished cell survival via an increase in apoptosis (Bates et al., 2003). Interestingly, VEGFR1 expression has been shown to be increased in cells transfected with PAX3-FKHR, a translocation characteristic of alveolar rhabdomyosarcoma (ARMS) (Barber et al., 2002).

Rhabdomyosarcoma (RMS) is the most common paediatric soft-tissue sarcoma, with an annual incidence of four to seven cases per million children under 16 years of age (Young et al., 1986). RMS is part of the larger family of ‘small round blue cell tumours’ of childhood and histologically resembles rhabdomyoblasts, the precursors of striated muscle (Merlino and Helman, 1999). There are two main subtypes of RMS, embryonal (ERMS) and alveolar (ARMS). ERMS tumours can be characterized by a loss of heterozygosity at 11p15.5 (Koufos et al., 1985), associated with a loss of maternal chromosomal material and duplication of paternal chromosomal material (Merlino and Helman, 1999). The majority of ARMS tumours have characteristic chromosomal rearrangements that translocate the PAX3 gene at 2q35 to FKHR at 13q14 (t(2;13)(q35;q14)) (Barr et al., 1993), or less frequently contain a t(1;13)(p36;q14) translocation that fuses PAX7 to FKHR (Davis et al., 1994). RMS tumours are known to express many autocrine signalling pathways, including insulin-like growth factor-II (IGF-II) (El-Badry et al., 1990; De Giovanni et al., 1995), basic fibroblast growth factor (bFGF) (Schweigerer et al., 1987; De Giovanni et al., 1995), epidermal growth factor (EGF) (De Giovanni et al., 1995), transforming growth factor-β (TGF-β) (De Giovanni et al., 1995) and myostatin (Ricaud et al., 2003). By blocking these pathways, RMS cells exhibit growth inhibition and occasionally are induced to differentiate (Schweigerer et al., 1987; El-Badry et al., 1990; De Giovanni et al., 1995; Ricaud et al., 2003).

In the present study, we report the first evidence of VEGF receptors in RMS. We found that VEGFR1 is expressed more frequently than VEGFR2 and VEGFR3. We also present evidence that RMS cell lines produce VEGF and that application of exogenous VEGF further increases cell growth. Treatment with a blocking antibody for VEGFR1 inhibits VEGF signalling and slows RMS proliferation. All-trans-retinoic acid (AtRA), previously shown to slow RMS growth (Crouch and Helman, 1991), decreases VEGF secretion from RMS cells. Furthermore, we show that this slowed growth of RMS cells upon AtRA treatment can be rescued by supplementing cells with VEGF. These data suggest that RMS cell growth is dependent on autocrine signalling, which includes VEGF. By interrupting these pathways, with VEGFR1 blocking antibodies and AtRA, RMS cell growth is reduced. Thus, the VEGF autocrine signalling pathway may represent novel therapeutic target for the treatment of RMS.

Results

RMS cell lines express VEGF, VEGFR and NRP mRNAs

RT–PCR analysis revealed that VEGF family receptor mRNAs are expressed in RMS cell lines. VEGFR1 mRNA could be detected in four of the six RMS lines assayed (RH4, RH6, RH18 and RD). RH28 and RH30 were the only lines in which VEGFR1 mRNA could not be detected (Figure 1a). This primer set amplified a region corresponding to VEGFR1 exons 6–8, a region that encodes the extracellular region of the VEGFR1 protein. By using a second primer set, which amplified a region corresponding to the transmembrane region of the protein, we confirmed that VEGFR1 mRNA is present in four of the six RMS lines (data not shown). In addition, expression of VEGFR2 mRNA was detected in RH28 cells (Figure 1b) and VEGFR3 mRNA was detected in RH4 and RH18 cell lines (Figure 1c). Additional RT–PCR reactions, using different primers and an increased number of cycles, confirmed that VEGFR2 was expressed only in RH28 cells (data not shown). The RMS cells lines also expressed mRNAs corresponding to the non-tyrosine kinase VEGF coreceptors, NRPs. NRP1 mRNA was expressed in four RMS lines (RH4, RH6, RH18 and RH30) and NRP2 mRNA was detected in all lines except for RH18 (Figure 1c).

Figure 1
figure1

Expression of VEGF and receptors in RMS cells. (a) RT–PCR for VEGFR1 demonstrated mRNA expression in four of six RMS lines (two alveolar, two embryonal). (b) RT–PCR for VEGFR2 detected mRNA in one of six RMS lines. (c) RT–PCR for VEGFR3 detected mRNA in two of six RMS lines. NRP1 mRNA was detected in four of six RMS lines and NRP2 mRNA was detected in five of six RMS lines. (d) RT–PCR for VEGF detected uniform expression of multiple VEGF transcripts across all RMS lines. (e) Membrane expression of VEGFR1 was assessed by Western blot. Protein was extracted as described in Materials and methods. Protein (150 μg) was separated by SDS–PAGE (8%), followed by transfer to nitrocellulose. The blot was probed with monoclonal antibody for VEGFR1, which demonstrated the presence of the receptor in three RMS cell lines. (f) Supernatant VEGF165 levels (after growth for 2 days) were quantified by ELISA. Results from ELISA were normalized by cell numbers as determined by Coulter counter. Data were assayed in triplicate and results are representative of three independent experiments

VEGF mRNA was expressed in all six RMS cell lines (Figure 1d). Using a previously published primer set (Suzuki et al., 1996), VEGF121, VEGF145 and VEGF165 mRNAs were clearly expressed (Figure 1d, upper panel). Additionally, we designed primers to amplify transcripts containing exons 6A and 7, which in theory are specific to VEGF183, VEGF189 and VEGF206. Upon amplification, only expressions of VEGF183 and VEGF189 mRNAs were seen. A similar transcript expression pattern was seen in all six RMS lines (Figure 1d, middle panel).

RMS cell lines express VEGFR1 protein

RH4, RH18, RH30 and RD were analysed by Western blot for the membrane expression of VEGFR1 (Figure 1e). Similar to results by RT–PCR, RH4, RH18 and RD expressed VEGFR1 and RH30 did not.

RMS cell lines secrete VEGF protein

All six RMS cell lines were analysed by ELISA for the production of VEGF165 (Figure 1f). Confirming the RT–PCR results, all six RMS lines produced VEGF165. However, baseline levels varied among the lines, with RH4 secreting the most VEGF165 (158 pg/ml/105 cells) and RD producing the least (25 pg/ml/105 cells).

ERK-1/2 are phosphorylated in response to VEGF165

As extracellular signal-regulated kinase (ERK)-1 and ERK-2 are phosphorylated following VEGF- and PlGF-induced activation of VEGFR1 (Podar et al., 2001; Selvaraj et al., 2003), we examined their phosphorylation states after a brief exposure of cells to the growth factor. ERK-1/2 phosphorylation is an indirect measure of receptor activity following VEGF165 stimulation. To demonstrate VEGFR1 activation in RMS cells, RH4 and RH30 were used as representative VEGFR1-positive and VEGFR1-negative cell lines, respectively. HeLa cells served as a negative control. After serum starvation for 24 h, with frequent washing and medium changes to reduce baseline activation of ERK-1/2, HeLa, RH4 and RH30 cells either remained in serum-free media, or were stimulated for 3 min with 10% fetal bovine serum (FBS) or 200 ng/ml VEGF165. This amount of VEGF165 used was shown to produce the strongest ERK-1/2 activation in preliminary experiments and compares to similar concentrations of VEGF and PlGF used previously to activate VEGFR1 (Podar et al., 2001; Selvaraj et al., 2003). As shown in Figure 2, ERK-1/2 was not activated in HeLa, RH4 or RH30 cells in serum-free conditions and all cell lines were positive for phospho-ERK-1/2 when stimulated with 10% FBS. More importantly, RH4 cells demonstrated ERK-1/2 activation in response to VEGF165 stimulation, whereas the proteins remained unphosphorylated in VEGFR1-negative HeLa and RH30 cells. Moreover, VEGF-induced ERK-1/2 phosphorylation was decreased by 36% when RH4 cells were preincubated with a blocking antibody for VEGFR1. This assay demonstrates that VEGFR1 is activated upon VEGF165 stimulation and that this activation can be inhibited with the addition of a blocking antibody for the receptor. Vinculin was used to normalize the bands instead of total ERK, because membranes were developed by AP staining. With this method, membranes cannot be stripped and reprobed. As a result, measurement of total ERK would have required separate lanes and would not have reflected loading differences of the phospho-ERK lanes.

Figure 2
figure2

VEGF induces ERK-1/2 phosphorylation in VEGFR1-expressing cells. After reaching 70% confluence, cells were serum starved and repeatedly washed over a 24 h period. Cells remained untreated or treated with either 10% FBS or VEGF (200 ng/ml) for 3 min. Where indicated, cells were preincubated with goat polyclonal anti-VEGFR1 blocking antibody (Ab; 20 μg/ml) for 1 h prior to VEGF treatment. Cells were immediately lysed and 50 μg of protein was subjected to SDS–PAGE (12.5%). ERK-1 and ERK-2 activities were analysed by Western blot using a rabbit polyclonal anti-phospho-MAPK (Thr202/Tyr204) antibody. To account for loading differences, a monoclonal anti-vinculin antibody was also used. Densitometric analysis was performed to determine differences in phospho-ERK-2 levels relative to vinculin between treatments in RH4 cells. Values are expressed relative to VEGF-treated cells (100%)

Addition of VEGF165 increases RMS proliferation

Addition of exogenous VEGF165 to RMS cell lines increased the proliferation of VEGFR1-positive lines (RH4, RH18 and RD) and promoted the survival of VEGFR2-positive RH28 cells, in a dose-dependent manner, as determined by counting total cells by Coulter Counter at treatment days 0 and 4 (Figure 3a). RH4, RH18 and RD cell numbers increased with VEGF165 treatment from day 0 to 4. In contrast, the RH28 cells had decreased in number by day 4; however, this decrease was mitigated by VEGF165. At the highest doses of VEGF165, cell numbers of all four RMS lines were significantly greater than those of nontreated cells (RH4: 205%, P<0.001; RH18: 82%, P<0.001; RH28: 131%, P<0.001; RD: 56%, P<0.01). This increase in growth could be blocked by the application of a blocking antibody for VEGFR1, as determined by Alamar Blue assay (Figure 3b and c). A 20 μg portion of anti-VEGFR1 antibody prevented the proliferation of VEGF165-stimulated RH4 cells by up to 97% (P<0.01).

Figure 3
figure3

VEGF signals through VEGFR1 to promote RMS growth. (a) VEGF promotes RMS cell growth in a dose-dependent manner. RMS cell lines RH4, RH18, RH28 and RD were plated in 24-well plates, allowed to recover and then serum starved (DMEM, 0% FBS) for 24 h. DMEM was then replaced and cells were either left untreated or incubated in the presence of VEGF165 (2, 5, 10, 50 or 100 ng/ml) and refreshed 48 h later. At 96 h post-treatment, cells were counted by Coulter counter. Data shown are the mean±s.d. of experiments performed in triplicate and are representative of three independent experiments for each cell line. The asterisks indicate significant differences (*P<0.05; **P<0.01; ***P<0.005; ****P<0.001) between treatments and untreated controls. (b, c) VEGFR1 blockade of RH4 cells abolishes VEGF165-induced proliferation. A total of 3.6 × 104 RH4 cells were plated in a 24-well plate, allowed to recover and then serum starved (DMEM, 0% FBS) for 24 h. Medium was changed and cells were either left untreated () or incubated with 100 ng/ml of VEGF165 () or 100 ng/ml of VEGF165 plus 20 μg of anti-VEGFR1 blocking antibody (Ab; ▪) with refreshment at 48 and 72 h post-treatment (b). Alternatively, treatments were 20 μg of nonspecific goat IgG, 100 ng/ml of VEGF165 plus goat IgG or 100 ng/ml of VEGF165 plus 20 μg of anti-VEGFR1 blocking antibody (Ab), with refreshment at 48 h (c). The number of viable cells was quantified by Alamar Blue assay at various time points. Asterisks indicate significant differences (*P<0.05; **P<0.01) between VEGF-treated and nontreated cells and number signs show significant differences (#P<0.05; ##P<0.01) between VEGF-treated cells plus or minus VEGFR1 Ab (b). Alternatively, comparison groups showed significance (*P<0.001) or were not significantly (NS) different (c). Data shown are the mean±s.d. of experiments performed in triplicate

VEGFR1 antibody treatment of RH4 cells decreases proliferation

As RH4 cells were shown to produce a large amount of VEGF165 compared to the other RMS cell lines, we hypothesized that ablation of autocrine VEGF binding to VEGFR1 may inhibit cell growth. Inhibition of autocrine VEGF signalling was assayed by treating RH4 cells with various doses of VEGFR1 blocking antibody, waiting 4 days, followed by counting cells by Coulter Counter (Figure 4). At the lowest doses of the antibody, compared to nontreated cells, RH4 growth increased by up to 17% (P<0.05). At higher antibody concentrations, however, RH4 growth decreased by up to 23% (P<0.01).

Figure 4
figure4

VEGFR1 blockade induces changes in RH4 growth. RH4 cells were seeded in 24-well plates and allowed to recover. Fresh DMEM containing 2% FBS was added and cells were either left untreated or incubated in the presence of increasing concentrations of anti-VEGFR1 polyclonal antibody (1, 2, 5, 10, 20 and 40 μg/ml). After 4 days, cells were trypsinized and counted by Coulter counter. At low doses of the VEGFR1 blocking antibody, cell numbers were higher than the control, whereas at high doses, cell numbers were lower than untreated cells. Data shown are the mean±s.d. of experiments performed in triplicate and results are representative of three independent experiments. Asterisks indicate statistically significant differences between treatment groups (*P<0.05; **P<0.01)

AtRA decreases VEGF165 production by RH4 cells

Retinoic acid has been previously shown to reduce VEGF production by various tumour types (Weninger et al., 1998; Kini et al., 2001) and slow RMS growth (Crouch and Helman, 1991; Brodowicz et al., 1999). We hypothesized, therefore, that VEGF production by RMS cells may be decreased with treatment by 5 μM AtRA, a dose that has previously been shown to arrest RMS cell growth (Crouch and Helman, 1991). An ELISA assay was used to compare VEGF165 production in AtRA-treated RH4 cells versus control cells treated with an equivalent volume of ethanol (Figure 5a). AtRA treatment decreased VEGF165 secretion by more than 50% (P<0.05).

Figure 5
figure5

VEGF rescues growth of AtRA-treated RH4 cells. In separate experiments, RH4 cells were seeded at a density of 5 × 105 cells/well in a 24-well plate in DMEM containing 10% FBS. (a) After 24 h, fresh DMEM containing 2% FBS and 5.0 μm AtRA or an equivalent volume of ethanol was added to the cells and DMEM and treatments were refreshed after 48 h. After an additional 48 h, cells were counted with a Coulter counter and supernatants were collected to detect VEGF165 by ELISA. Cells treated with AtRA produced less VEGF165 than ethanol-treated controls (*P<0.05). (b) After 24 h, fresh serum-free medium containing DMEM was added to the wells in addition to treatment conditions: 5.0 μm AtRA (▪), 200 ng VEGF165 to AtRA-treated cells (), or an equivalent volume of ethanol (EtOH; ), with refreshment 24 h post-treatment. At various time points (0=initial treatment), 10% Alamar Blue was added to each well and fluorescence was measured following 5 h of incubation. Treatment with 5.0 μm AtRA decreased the number of viable cells, as compared to ethanol-treated controls at all time points (P<0.001). Addition of 200 ng VEGF to AtRA-treated cells rescued the number of viable cells by day 2 (*P<0.05; **P<0.005; ***P<0.001). Data shown are the mean±s.d. of experiments performed in triplicate and results are representative of three independent experiments

Addition of VEGF165 to AtRA-treated RH4 cells rescues growth

As VEGF165 production by RH4 cells was shown to be decreased by AtRA treatment and this cell line appears to be supported by autocrine VEGF signalling, we reasoned that the decrease of this growth factor by AtRA may contribute to the slowing of RH4 cell growth. To examine this possibility, 200 ng VEGF165 was added to AtRA-treated RH4 cells and compared to cells treated solely with AtRA or an equivalent volume of ethanol. Using Alamar Blue to measure cellular metabolic activity, viability of RH4 cells in the three treatment conditions was examined over multiple time points (Figure 5b). After 1 day, compared to ethanol treatment, AtRA reduced the number of viable cells by 23% (P<0.001). After 6 days, there were 61% fewer viable AtRA-treated RH4 cells than ethanol-treated cells (P<0.001). VEGF165 supplementation rescued the number of viable AtRA-treated cells by 29% (P<0.05) at day 2 and at day 6, the percent rescue was 74% (P<0.001).

Discussion

Initially, research on VEGF signalling focused almost exclusively on its mitogenic and survival functions towards endothelial cells. More recently, however, mounting evidence suggests that malignant cells within certain tumour types also express VEGF receptors and respond to VEGF signalling (Dias et al., 2001; Broggini et al., 2003; Narendran et al., 2003). Both VEGFR1 and VEGFR2 have been detected in leukaemia and malignant myeloma and their growth and survival are decreased by inhibiting autocrine VEGF signalling (Dias et al., 2000, 2001, 2002; Podar et al., 2001, 2004). Tumour cell expression of VEGF receptors has also been seen in some solid tumours (Bates et al., 2003; Wey et al., 2004) and recent reports suggest that autocrine VEGF signalling may also influence solid tumour growth and survival. For example, our laboratory has recently shown that VEGFR1 signalling participates in hypoxia-mediated drug resistance in neuroblastoma (Das et al., in press).

Using the VEGF- and VEGFR1-expressing colon cancer cell line LIM 1863, Bates et al. (2003) treated cells with a neutralizing antibody against VEGFR1 that led to widespread apoptosis. This result was supported by a report by Fan et al. (2005), who demonstrated that the activation of VEGFR1 by VEGF or VEGF-B enhances colorectal cancer cell invasion. Interestingly, the VEGF-sequestering agent Avastin™ (Genentech) has been very successful at treating colorectal cancer (Brower, 2003; Kabbinavar et al., 2003; Willett et al., 2004) and perhaps this high degree of efficacy results partially from interfering with autocrine VEGF signalling (Mercurio et al., 2004). This suggests that anti-VEGF therapy may be successful at treating other cancers in which the tumour cells express VEGF receptors.

RMS might be such a tumour that responds well to anti-VEGF therapy. RMS xenografts were used in the earliest published mouse studies demonstrating the effectiveness of both Avastin™ (Kim et al., 1993; Borgstrom et al., 1996) and the more recent angiogenesis blocker VEGF Trap (Regeneron; Holash et al., 2002). Furthermore, a recent report hinted that RMS might express VEGF receptors. Barber et al. (2002) found that P19, HeLa and NIH3T3 cells transfected with PAX3-FKHR had induced expression of several genes, including VEGFR1. As PAX3-FKHR is the characteristic translocation of ARMS, we reasoned that RMS tumours may be supported by autocrine VEGF signalling. To address this, we characterized six RMS cell lines (four PAX3-FKHR-positive ARMS lines and two ERMS lines) by RT–PCR and ELISA and we have demonstrated that all six cell lines express VEGF. This result is perhaps not surprising based on previous reports demonstrating that angiogenesis is promoted in RMS xenografts in mice via VEGF (Kim et al., 1993; Borgstrom et al., 1996). Our study, however, also shows that all six RMS lines studied express one or more receptors for VEGF (VEGFRs and/or NRPs). To the best of our knowledge, this is the first time that tumour cell expression of VEGF receptors has been seen in RMS.

VEGFR1 mRNA was detected in only two of four ARMS lines (RH4 and RH18) and this was confirmed by Western blot. It had been expected that all four PAX3-FKHR-positive lines would express VEGFR1 mRNA, as the fusion protein has been shown to increase VEGFR1 expression (Barber et al., 2002). As cultured cell lines have an increased rate of genetic instability, the absence of VEGFR1 mRNA expression in RH28 and RH30 cell lines may simply be the result of deletion of genetic loci encompassing the VEGFR1 gene. It is possible, however, that these two cell lines may represent a distinct subset of ARMS (PAX3-FKHR+, VEGFR1−). Whether such an RMS subclassification exists in primary tumours and its potential value in clinical stratification, remains to be explored. Using a larger sample size, including both RMS cell lines and primary tumour samples, we are currently examining this question. Notably, VEGFR1 was also detected in both ERMS lines (RH6 and RD). As VEGFR1 is also a target of PAX3 (Barber et al., 2002), upregulation of PAX3 in ERMS lines (previously noted in RD cells; Barr et al., 1999) may lead to an increase in VEGFR1 expression, which would help to explain its expression in the ERMS cell lines.

Interestingly, other VEGF receptors were detected in the RMS lines. All cells that expressed VEGFR1 also expressed either NRP1 or NRP2 and VEGFR3 was coexpressed with NRP2 in RH4 cells. Coexpression of NRPs and VEGFRs has been previously reported in endothelial cells (Karkkainen et al., 2001) and neuroblastoma cell lines (Beierle et al., 2003, 2004; Das et al., in press). The detection of both NRPs and VEGFRs in RMS cell lines suggests that VEGFR expression in this tumour type likely has clinical relevance and is not merely a tissue culture artefact.

As the majority of RMS cell lines expressed VEGFRs, we assessed the responsiveness of these lines to exogenous VEGF165. Stimulating RH4 cells with VEGF165 induced ERK-1/2 phosphorylation, which could be decreased by 36% when cells were preincubated with a blocking antibody for VEGFR1. This demonstrates that VEGFR1 in RH4 cells (and possibly other RMS lines) is activated in response to addition of VEGF165. However, the antibody was unable to completely abolish VEGF-induced ERK activation. In contrast, Selvaraj et al. (2003), using the same antibody (albeit in a different cell system (monocytes)), demonstrated that ERK activation following stimulation of cells with PlGF could be reduced by approximately 75% with VEGFR1 blockade. We used 20 μg of antibody to block signalling from 200 ng of exogenously added VEGF165, while Selvaraj et al. used only 5 μg of antibody to block 250 ng of PlGF from binding to VEGFR1. This suggests that while VEGFR1 is likely the primary VEGF signalling molecule in RH4 cells, treatment with the blocking antibody is only partially effective. Perhaps the antibody may not block VEGF binding as effectively as it blocks PlGF. This is possible, as the effectiveness of this antibody was validated, in terms of its ability to inhibit VEGFR1 signalling, using PlGF, not VEGF.

The presence of NRPs may also help to explain the low rate of inhibition with blocking antibody treatment. NRP1 and NRP2 have been shown to bind to both VEGF and VEGFR1 (Fuh et al., 2000). We demonstrated that RH4 cells express both NRP1 and NRP2, so it is possible that the antibody may fail to disrupt VEGFR1–ligand binding when in complex with NRPs. This may explain why the ERK phosphorylation was only inhibited by 36% with pretreatment with VEGFR1 blocking antibody. Furthermore, we showed that both RH4 and RH30 cells express NRP1 and NRP2; however, upon VEGF stimulation, only RH4 cells exhibited ERK phosphorylation. Thus, it is not likely that the NRPs independently contribute to ERK phosphorylation in RMS, which is not expected since they have short intracellular domains that are unlikely to transduce signals (Neufeld et al., 2002). This was confirmed in a separate but related experiment. We treated RH4 cells with a VEGFR tyrosine kinase inhibitor (5 μM; Calbiochem, San Diego, CA, USA), which resulted in an 82% reduction in ERK-2 phosphorylation following VEGF stimulation (data not shown). This suggests that VEGFR1 is likely the receptor transducing the VEGF signal in RH4 cells.

By treating RH4 cells with a VEGFR1 blocking polyclonal antibody in the absence of additional growth factors, cell proliferation was reduced compared to the nontreated control. This suggests that autocrine VEGF signalling in RMS can be inhibited by blocking VEGFR1. At the highest doses, VEGFR1 blocking antibody inhibited growth of RH4 cells. As no exogenous growth factors were added, the blockage of VEGFR1 appears to have decreased autocrine signalling through the receptor. It is possible that RH4-produced PlGF also acts as an autocrine factor through VEGFR1, although this was not examined in our study. Thus, presumably signalling by both VEGF and PlGF is decreased by blocking VEGFR1 (Lyden et al., 2001).

Interestingly, low doses of the VEGFR1 blocking antibody appear to provide RH4 cells with a proliferative advantage, as the addition of 1 or 2 μg of antibody yields higher cell numbers than nontreated cells. While this effect appears to be counterintuitive, the biological nature of VEGFR1 may provide some insight. Aside from the membrane-spanning version of VEGFR1, alternative splicing yields soluble VEGFR1 (sVEGFR1). This soluble form competes with membrane-bound VEGFR1 for both VEGF and PlGF and it has been suggested that the soluble receptor may act as a sink for excess ligand (Kendall and Thomas, 1993; Kendall et al., 1996). Perhaps, in a manner similar to what has been suggested for soluble Her2 receptor (Kita et al., 1996), VEGFR1 at the cell surface may have limited accessibility of epitopes compared to its free molecule in solution. As a result, the VEGFR1 blocking antibody may preferentially bind to sVEGFR1 and at low antibody doses, antibody binding to sVEGFR1 may cause displacement of ligand. If so, low antibody doses may cause additional VEGF to be available to bind to membrane-spanning VEGFR1, resulting in higher cell numbers versus nontreated cells. When present in excess, antibody would likely bind both forms of VEGFR1 resulting in decreased signalling through the receptor. Alternatively, sVEGFR1 may bind to membrane-spanning receptors and act in a dominant negative fashion (Kendall et al., 1996); therefore, low doses of the antibody may disrupt this interaction thereby relieving this intrinsic repression of VEGFR1 signalling. These results raise a troubling issue. If receptor blocking antibodies are used as therapeutics, it is possible that inadequate doses may be fostering tumour growth rather than inhibiting it. We are currently working to elucidate this paradoxical effect and attempting to determine if blocking antibodies are effective agents in RMS treatment.

Previously, reports have demonstrated that, likely via decreases in AP1 activity (Diaz et al., 2000), retinoid treatment of various tumour types results in decreases in VEGF production (Weninger et al., 1998; Kini et al., 2001). Thus, we treated RH4 cells with AtRA to look for similar effects and to further investigate autocrine signalling in RMS. Using an ELISA, we demonstrated that AtRA treatment of RH4 cells decreased VEGF165 secretion by approximately half of baseline levels. We reasoned that the decline in growth and increase in apoptosis of RMS cells treated with AtRA may be a result of a reduction in VEGF secretion. To verify this, we supplemented AtRA proliferation-inhibited RH4 cells with exogenous VEGF165 to see if growth could be rescued. It should be noted that we used 5 μM AtRA to induce growth arrest in RH4 cells, a dose that may not be physiologically attainable. However, as cell lines inherently have resistance to pharmacologic doses of therapeutic agents, this dose was deemed appropriate for our studies. Moreover, this dose has been previously established as one that effectively inhibits RMS growth in vitro (Crouch and Helman, 1991; Brodowicz et al., 1999).

Previous experiments demonstrated that treatment with AtRA resulted in significantly decreased growth versus ethanol-treated control cells (Crouch and Helman, 1991; Brodowicz et al., 1999). We confirmed this finding in our experiment. However, upon addition of VEGF165, cells demonstrated significantly increased growth compared to cells treated solely with AtRA. To our knowledge, this experiment represents the first evidence that the growth of AtRA-treated RMS cells can be rescued upon the addition of exogenous growth factor.

This rescue experiment highlights two important points. First, the results further support our claim that RMS cell lines are responsive to changes in the VEGF signalling pathway. By reducing the amount of VEGF normally produced, cellular responses to exogenously added VEGF165 become very evident. Second, this demonstrates a potential mechanism by which AtRA inhibits RMS cell growth. Likely, the reduction of proliferative and antiapoptotic factors, such as autocrine VEGF signalling, results in decreased RMS growth and increased apoptosis. When an autocrine factor, like VEGF, is returned to the system, the growth of the RMS cells is rescued. This phenomenon is clearly demonstrable in serum-free conditions where there is an absence of growth-promoting factors.

While our data show that the growth of AtRA-treated RMS cells is rescued with supplementation with VEGF, this rescue is not complete. This suggests that in addition to VEGF, other prominent RMS autocrine signalling pathways, such as IGF-II, bFGF, EGF and TGF-β, may be inhibited by AtRA. In an early report of retinoid use in RMS, Crouch and Helman (1991) used AtRA to inhibit the growth of RD cells, which are known to produce IGF-II at a high rate (Zhang et al., 1998) and express IGF-I-R (Kalebic et al., 1994). They reasoned that the decrease in RD proliferation may be due to decreases in IGF-II signalling; however, no growth rescue was seen in cells treated with AtRA and IGF-II compared to those treated solely with AtRA (Crouch and Helman, 1991). Recently, however, it was shown that retinoic acid decreases protein levels and phosphorylation of insulin receptor substrate-1, a key adapter molecule for IGF-I-R, which serves as a docking module for several SH2 domain-containing proteins (del Rincon et al., 2003). This may help to explain why Crouch and Helman did not observe any restoration in growth with IGF-II treatment, as RA interference of the IGF-II pathway likely occurs not at the level of the ligand, but downstream of IGF-I-R. This suggests that AtRA may inhibit multiple autocrine signalling pathways in RMS, such as IGF-II and VEGF and may be an appropriate treatment when used alone or in combination with other therapies.

In summary, our experiments have shown that multiple RMS cell lines express both VEGF and receptors, suggesting a possible autocrine loop. To verify this, we showed that RMS cells are responsive to VEGF and signals are likely transduced via VEGFR1. By blocking VEGFR1, we were able to reduce RMS cell numbers, suggesting that we inhibited an autocrine VEGF signalling pathway. Furthermore, we reduced VEGF production through AtRA treatment and saw a reduction in RMS growth. This was likely due to decreases in autocrine signalling and as a result of reintroduction of VEGF, the cells showed rescued growth. Thus, our results demonstrate a novel autocrine signalling pathway in RMS and suggest that the VEGF pathway may be an appropriate target for the treatment of RMS.

Materials and methods

Cell lines and culture conditions

The human ARMS lines RH4, RH18, RH28 and RH30 and the human ERMS line RH6 were provided by Dr Tom Look. RD (human ERMS), SK-N-BE(2) (human neuroblastoma), HeLa (human cervical adenocarcinoma) and HUVECs (human umbilical vein endothelial cells) were obtained from American Type Culture Collection. The pre-B acute lymphoblastic leukaemia cell line A1 was a gift from Dr Melvin Freedman (Hospital for Sick Children, Toronto, ON, Canada). HeLa cells were used as negative controls for expression of VEGFR1 (Wakiya et al., 1996) and VEGFR2. HUVECs were used as positive controls for expression of VEGFR3, NRP1 and NRP2 (Beierle et al., 2004). A1 (Narendran et al., 2003) and SK-N-BE(2) (Langer et al., 2000) cells were used as positive controls for VEGFR1 expression. All RMS cell lines, SK-N-BE(2) and HeLa cells were grown in Dulbecco's modified Eagle's medium (DMEM; Wisent Inc., Montréal, QC, Canada) supplemented with 10% FBS (Sigma, St Louis, MO, USA) and 1% penicillin–streptomycin (Invitrogen, Burlington, ON, Canada). A1 cells were grown in suspension in 10 ml OPTI-MEM I (Invitrogen, Burlington, ON, Canada) supplemented with 5% FBS and 1% penicillin–streptomycin. Cells were kept at 37°C in a humidified atmosphere containing 5% carbon dioxide with free gas exchange.

Reagents and treatments

Recombinant human VEGF165 (rhVEGF165), VEGFR1 blocking antibody (goat polyclonal anti-VEGFR1) and nonspecific goat IgG (R&D Systems, Minneapolis, MN, USA) were reconstituted according to the manufacturer's recommendations. Briefly, rhVEGF165 was reconstituted with PBS supplemented with 0.1% human serum albumin (Bayer, Toronto, ON, Canada), which yielded a 10 μg/ml stock solution. VEGFR1 blocking antibody and nonspecific goat IgG were both reconstituted in PBS, producing 1 μg/μl stock solutions. AtRA (Sigma Chemical, St Louis, MO, USA) was resuspended in 95% filter-sterilized ethanol at a stock concentration of 5 mM.

RT–PCR analysis

RMS cell lines were grown as monolayers in Falcon 10 cm dishes (Becton Dickinson, Franklin Lakes, NJ, USA). At 80% confluence, total RNA was isolated using Trizol (Invitrogen, Burlington, ON, Canada), according to the manufacturer's protocol. Bone marrow-derived stromal cell RNA (positive control during the amplification of VEGFR2) was a gift from Dr Aru Narendran (Alberta Children's Hospital, Calgary, AB, Canada). RNA samples (250 ng) were incubated at room temperature for 10 min prior to being placed in a 96-well GeneAmp PCR System 9700 (Applied Biosystems, Foster City, CA, USA) at which point RNA was reverse transcribed at 42°C for 15 min, denatured at 99°C for 5 min and then cooled at 5°C for 5 min. All reagents for reverse transcription were obtained from Applied Biosystems (Foster City, CA, USA). PCR reactions were performed on a TGradient Thermocycler (Whatman Biometra, Goettingen, Germany) using AmpliTaq Gold DNA Polymerase (Applied Biosystems, Foster City, CA, USA) and gene-specific oligonucleotide primers (see Table 1 for cycling conditions and primer sequences). Primers for VEGF set #1 (Suzuki et al., 1996), VEGFR2 (Ziegler et al., 1999), VEGFR3, NRP1 and NRP2 (Beierle et al., 2004) have been previously published. All PCR reactions were preceded by a 10 min hot start at 95°C and were followed by a 10 min final extension step at 72°C. A 5 μl portion of reaction products (15 μl for VEGF) was run on agarose gels (4% for VEGF; 1% for all others) and visualized using ethidium bromide staining.

Table 1 RT–PCR primer sequences, cycling conditions and product sizes

Western blot analysis

For ERK-1/2, cells (cultured in six-well plates) were washed twice with cold PBS and lysed with 200 μl of cold lysis buffer (50 mM Tris/HCl (pH 7.4), 5 mM EDTA (pH 7.4), 150 mM NaCl, 1% Triton X-100) containing complete, Mini protease inhibitor cocktail tablets (Roche, Penzberg, Germany). Lysates were placed on ice for 30 min and clarified by microcentrifugation (14 000 r.p.m., 10 min, 4°C). A 12.5% SDS–PAGE gel was used to separate 50 μg/sample of protein, followed by transfer onto nitrocellulose membranes. Vinculin and phosphorylated forms of ERK-1/2 were detected by immunoblotting using a mouse monoclonal anti-vinculin (Upstate, Charlottesville, VA, USA), rabbit polyclonal anti-phospho-p44/42 MAP kinase (Cell Signaling Technology, Beverly, MA, USA) and appropriate alkaline phosphatase-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories, West Grove, PA, USA). An alkaline phosphatase staining solution (1 M Tris/HCl (pH 9.5), 1 M NaCl, 50 mM MgCl2) containing 5-bromo-4-chloro-3-indolyl phosphate and nitro blue tetrazolium (Pierce Biotechnology, Rockford, IL, USA) was used to visualize the protein. Band intensity of immunoblots was quantified using a FluorChem 8000 Imaging System and FluorChem 2.00 software (Alpha Innotech Corporation, San Leandro, CA, USA) and normalized to intensities of corresponding vinculin bands.

For VEGFR1, cells (cultured in 10 cm dishes) were washed twice with cold PBS and homogenized with 500 μl of cold lysis buffer (10 mM Tris/HCl (pH 7.4), 2 mM EDTA (pH 7.4), 250 mM sucrose) containing complete, Mini protease inhibitor cocktail tablets (Roche, Penzberg, Germany) using delicate syringing with a 29 gauge needle. Lysates were centrifuged at 4°C and 900 g for 10 min and supernatants were then collected and ultracentrifuged at 4°C and 32 000 r.p.m. for 80 min. Pellets were collected and dissolved in 100 μl of solubilization buffer (10 mM Tris/HCl (pH 7.4), 2 mM ETDA, 0.5% Triton X-100, 0.5% sodium deoxycholate) containing complete, Mini protease inhibitor cocktail tablets (Roche, Penzberg, Germany). An 8% SDS–PAGE gel was used to separate 150 μg/sample of protein, followed by transfer onto nitrocellulose membranes. VEGFR1 and β-actin were detected by immunoblotting using a mouse monoclonal anti-VEGFR1 (Chemicon, Temecula, CA, USA) or mouse monoclonal anti-β-actin (Sigma-Aldrich, Oakville, ON, USA) and a horseradish peroxidase-conjugated rabbit anti-mouse secondary antibody (Pierce Biotechnology, Rockford, IL, USA). Membranes were developed using the ECL chemiluminescence system (Amersham, Piscataway, NJ, USA). Ponceau staining of membranes was performed to confirm equal loading of protein.

ELISA analysis

To assay VEGF165 secretion, cells were seeded in 24-well plates with DMEM containing 10% FBS. After 24 h, medium was replaced by DMEM supplemented with 2% FBS. For measurement of baseline VEGF165, supernatants were collected 24 h later. For measurement of VEGF165 following AtRA treatment (initial seeding density of 7.5 × 105 RH4 cells), medium was supplemented with either ethanol or 5 μM AtRA at the switch to low serum. After 24 h, medium containing treatment conditions was refreshed and supernatants were collected 48 h later. Secreted VEGF165 was measured using the Quantikine Human VEGF Immunoassay (R&D Systems, Minneapolis, MN, USA) following the manufacturer's protocol. The optical density of each well was detected by means of a microplate reader detecting absorbance at 450 nm with a wavelength correction of 540 nm. Results were normalized by the number of cells per well, as measured by Coulter Counter (Coulter, Hialeah, FL, USA).

Cell counting

Growth of RH4, RH18, RH28 and RD cells in response to addition of rhVEGF165 or growth of RH4 cells in response to VEGFR1 blocking antibody was determined by counting total cells with a Coulter Counter (Coulter, Hialeah, FL, USA). To examine growth in response to rhVEGF165, RMS cells were seeded in Costar 24-well plates (Corning, Corning, NY, USA) with DMEM containing 10% FBS. The RMS lines required different initial densities (3.4 × 104 RH4, 2.2 × 104 RH18, 9.0 × 104 RH28 and 2.0 × 104 RD) to survive 4 days in serum-free medium and be responsive to exogenous growth factor. After cells had adhered to the wells (24 h), cells were serum starved by adding serum-free medium. A further 24 h later, rhVEGF165 was added to fresh serum-free medium at various doses. Media and rhVEGF165 were refreshed 48 h later and cells were trypsinized and counted after 4 days from the point of initial cytokine addition. To examine growth in response to anti-VEGFR1 blocking antibody, 5.3 × 104 RH4 cells were seeded in 24-well plates with DMEM containing 10% FBS. After 24 h, medium was removed and replaced with 500 μl of DMEM containing 2% FBS, supplemented with VEGFR1 antibody at various doses. Cells were counted after 6 days.

Alamar Blue assay

The Alamar Blue assay was used to determine the cell viability of RH4 cells at different time points under different treatment conditions. To examine growth of RH4 cells over a 6-day period, 3.6 × 104 RH4 cells were seeded in 24-well plates with DMEM containing 10% FBS. After recovering (24 h), cells were serum starved for 24 h. At this point (day 0), serum-free medium was refreshed and treatments (100 ng rhVEGF165 alone or in combination with 20 μg VEGFR1 blocking antibody) were applied. Medium and treatments were refreshed at days 2 and 4. In a separate experiment, cells were treated with 20 μg nonspecific goat IgG, 100 ng rhVEGF165 plus goat IgG or rhVEGF165 plus 20 μg VEGFR1 blocking antibody and viability was measured at day 4. To determine if VEGF rescues growth of AtRA-treated RH4 cells, 5.0 × 104 cells were seeded in 24-well plates with DMEM containing 10% FBS. Cells were allowed to recover for 24 h and then 1 ml of DMEM was added to each of the wells. Cells were treated with 5 μM AtRA, 5 μM AtRA plus 200 ng rhVEGF165, or nontreatment represented by an equivalent volume of ethanol. Medium and treatment conditions were refreshed 24 h later and every 2 days thereafter for the course of the experiment. Alamar Blue was added to the wells (rhVEGF165 addition time course: days 0, 2, 4, 6; rhVEGF165 addition end point: day 4; AtRA rescue: days 0, 1, 2, 3, 4, 5, 6) at a final concentration of 10% and after 5 h of incubation, fluorescence (excitation: 540 nm; emission: 590 nm; auto-cutoff: 570 nm) was detected using a microplate reader.

Statistical analysis

Significance between treatment conditions was determined by two-tailed, unpaired Student's t-test. A P-value of 0.05 or less was considered to be statistically significant.

Abbreviations

VEGF:

vascular endothelial growth factor

rhVEGF165:

recombinant human vascular endothelial growth factor (165 amino-acid isoform)

PlGF:

placenta growth factor

VEGFR:

vascular endothelial growth factor receptor

flt1:

fms-like tyrosine kinase receptor (or VEGFR1)

KDR:

kinase insert domain-containing receptor (or VEGFR2)

NRP:

neuropilin

RMS:

rhabdomyosarcoma

ARMS:

alveolar rhabdomyosarcoma

ERMS:

embryonal rhabdomyosarcoma

AtRA:

all-trans-retinoic acid

IGF-II:

insulin-like growth factor-II

bFGF:

basic fibroblast growth factor

EGF:

epidermal growth factor

TGF-β:

transforming growth factor

ERK:

extracellular signal-regulated kinase

References

  1. Barber TD, Barber MC, Tomescu O, Barr FG, Ruben S and Friedman TB . (2002). Genomics, 79, 278–284.

  2. Barr FG, Fitzgerald JC, Ginsberg JP, Vanella ML, Davis RJ and Bennicelli JL . (1999). Cancer Res., 59, 5443–5448.

  3. Barr FG, Galili N, Holick J, Biegel JA, Rovera G and Emanuel BS . (1993). Nat. Genet., 3, 113–117.

  4. Bates RC, Goldsmith JD, Bachelder RE, Brown C, Shibuya M, Oettgen P and Mercurio AM . (2003). Curr. Biol., 13, 1721–1727.

  5. Beierle EA, Dai W, Langham Jr MR, Copeland III EM and Chen MK . (2003). J. Pediatr. Surg., 38, 514–521.

  6. Beierle EA, Dai W, Langham Jr MR, Copeland III EM and Chen MK . (2004). J. Surg. Res., 119, 56–65.

  7. Borgstrom P, Hillan KJ, Sriramarao P and Ferrara N . (1996). Cancer Res., 56, 4032–4039.

  8. Brodowicz T, Wiltschke C, Kandioler-Eckersberger D, Grunt TW, Rudas M, Schneider SM, Hejna M, Budinsky A and Zielinski CC . (1999). Br. J. Cancer, 80, 1350–1358.

  9. Broggini M, Marchini SV, Galliera E, Borsotti P, Taraboletti G, Erba E, Sironi M, Jimeno J, Faircloth GT, Giavazzi R and D’Incalci M . (2003). Leukemia, 17, 52–59.

  10. Brower V . (2003). J. Natl. Cancer Inst., 95, 1425–1427.

  11. Claesson-Welsh L . (2003). Biochem. Soc. Trans., 31, 20–24.

  12. Crouch GD and Helman LJ . (1991). Cancer Res., 51, 4882–4887.

  13. Damert A, Miquerol L, Gertsenstein M, Risau W and Nagy A . (2002). Development, 129, 1881–1892.

  14. Das B, Yeger H, Tsuchida R, Torkin R, Gee MFW, Thorner P, Ho M, Shibuya M, Malkin D and Baruchel S . Cancer Res. (in press).

  15. Davis RJ, D’Cruz CM, Lovell MA, Biegel JA and Barr FG . (1994). Cancer Res., 54, 2869–2872.

  16. De Giovanni C, Melani C, Nanni P, Landuzzi L, Nicoletti G, Frabetti F, Griffoni C, Colombo MP and Lollini PL . (1995). Br. J. Cancer, 72, 1224–1229.

  17. de Vries C, Escobedo JA, Ueno H, Houck K, Ferrara N and Williams LT . (1992). Science, 255, 989–991.

  18. del Rincon SV, Rousseau C, Samanta R and Miller Jr WH . (2003). Oncogene, 22, 3353–3360.

  19. Dias S, Hattori K, Heissig B, Zhu Z, Wu Y, Witte L, Hicklin DJ, Tateno M, Bohlen P, Moore MA and Rafii S . (2001). Proc. Natl. Acad. Sci. USA, 98, 10857–10862.

  20. Dias S, Hattori K, Zhu Z, Heissig B, Choy M, Lane W, Wu Y, Chadburn A, Hyjek E, Gill M, Hicklin DJ, Witte L, Moore MA and Rafii S . (2000). J. Clin. Invest., 106, 511–521.

  21. Dias S, Shmelkov SV, Lam G and Rafii S . (2002). Blood, 99, 2532–2540.

  22. Diaz BV, Lenoir MC, Ladoux A, Frelin C, Demarchez M and Michel S . (2000). J. Biol. Chem., 275, 642–650.

  23. El-Badry OM, Minniti C, Kohn EC, Houghton PJ, Daughaday WH and Helman LJ . (1990). Cell Growth Differ., 1, 325–331.

  24. Fan F, Wey JS, McCarty MF, Belcheva A, Liu W, Bauer TW, Somcio RJ, Wu Y, Hooper A, Hicklin DJ and Ellis LM . (2005). Oncogene, 24, 2647–2653.

  25. Ferrara N . (1996). Eur. J. Cancer, 32A, 2413–2422.

  26. Folkman J . (1995). Nat. Med., 1, 27–31.

  27. Fong GH, Rossant J, Gertsenstein M and Breitman ML . (1995). Nature, 376, 66–70.

  28. Fuh G, Garcia KC and de Vos AM . (2000). J. Biol. Chem., 275, 26690–26695.

  29. Gille H, Kowalski J, Yu L, Chen H, Pisabarro MT, Davis-Smyth T and Ferrara N . (2000). EMBO J., 19, 4064–4073.

  30. Holash J, Davis S, Papadopoulos N, Croll SD, Ho L, Russell M, Boland P, Leidich R, Hylton D, Burova E, Ioffe E, Huang T, Radziejewski C, Bailey K, Fandl JP, Daly T, Wiegand SJ, Yancopoulos GD and Rudge JS . (2002). Proc. Natl. Acad. Sci. USA, 99, 11393–11398.

  31. Isner JM and Losordo DW . (1999). Nat. Med., 5, 491–492.

  32. Itokawa T, Nokihara H, Nishioka Y, Sone S, Iwamoto Y, Yamada Y, Cherrington J, McMahon G, Shibuya M, Kuwano M and Ono M . (2002). Mol. Cancer Ther., 1, 295–302.

  33. Joukov V, Pajusola K, Kaipainen A, Chilov D, Lahtinen I, Kukk E, Saksela O, Kalkkinen N and Alitalo K . (1996). EMBO J., 15, 290–298.

  34. Kabbinavar F, Hurwitz HI, Fehrenbacher L, Meropol NJ, Novotny WF, Lieberman G, Griffing S and Bergsland E . (2003). J. Clin. Oncol., 21, 60–65.

  35. Kalebic T, Tsokos M and Helman LJ . (1994). Cancer Res., 54, 5531–5534.

  36. Karkkainen MJ, Saaristo A, Jussila L, Karila KA, Lawrence EC, Pajusola K, Bueler H, Eichmann A, Kauppinen R, Kettunen MI, Yla-Herttuala S, Finegold DN, Ferrell RE and Alitalo K . (2001). Proc. Natl. Acad. Sci. USA, 98, 12677–12682.

  37. Ke LD, Fueyo J, Chen X, Steck PA, Shi YX, Im SA and Yung WK . (1998). Int. J. Oncol., 12, 1391–1396.

  38. Keck PJ, Hauser SD, Krivi G, Sanzo K, Warren T, Feder J and Connolly DT . (1989). Science, 246, 1309–1312.

  39. Kendall RL and Thomas KA . (1993). Proc. Natl. Acad. Sci. USA, 90, 10705–10709.

  40. Kendall RL, Wang G and Thomas KA . (1996). Biochem. Biophys. Res. Commun., 226, 324–328.

  41. Kim KJ, Li B, Winer J, Armanini M, Gillett N, Phillips HS and Ferrara N . (1993). Nature, 362, 841–844.

  42. Kini AR, Peterson LA, Tallman MS and Lingen MW . (2001). Blood, 97, 3919–3924.

  43. Kita Y, Tseng J, Horan T, Wen J, Philo J, Chang D, Ratzkin B, Pacifici R, Brankow D, Hu S, Luo Y, Wen D, Arakawa T and Nicolson M . (1996). Biochem. Biophys. Res. Commun., 226, 59–69.

  44. Koufos A, Hansen MF, Copeland NG, Jenkins NA, Lampkin BC and Cavenee WK . (1985). Nature, 316, 330–334.

  45. Langer I, Vertongen P, Perret J, Fontaine J, Atassi G and Robberecht P . (2000). Med. Pediatr. Oncol., 34, 386–393.

  46. Lu D, Jimenez X, Zhang H, Wu Y, Bohlen P, Witte L and Zhu Z . (2001). Cancer Res., 61, 7002–7008.

  47. Luttun A, Tjwa M, Moons L, Wu Y, Angelillo-Scherrer A, Liao F, Nagy JA, Hooper A, Priller J, De Klerck B, Compernolle V, Daci E, Bohlen P, Dewerchin M, Herbert JM, Fava R, Matthys P, Carmeliet G, Collen D, Dvorak HF, Hicklin DJ and Carmeliet P . (2002). Nat. Med., 8, 831–840.

  48. Lyden D, Hattori K, Dias S, Costa C, Blaikie P, Butros L, Chadburn A, Heissig B, Marks W, Witte L, Wu Y, Hicklin D, Zhu Z, Hackett NR, Crystal RG, Moore MA, Hajjar KA, Manova K, Benezra R and Rafii S . (2001). Nat. Med., 7, 1194–1201.

  49. Manley PW, Bold G, Bruggen J, Fendrich G, Furet P, Mestan J, Schnell C, Stolz B, Meyer T, Meyhack B, Stark W, Strauss A and Wood J . (2004). Biochim. Biophys. Acta, 1697, 17–27.

  50. Mercurio AM, Bachelder RE, Bates RC and Chung J . (2004). Semin. Cancer Biol., 14, 115–122.

  51. Merlino G and Helman LJ . (1999). Oncogene, 18, 5340–5348.

  52. Millauer B, Longhi MP, Plate KH, Shawver LK, Risau W, Ullrich A and Strawn LM . (1996). Cancer Res., 56, 1615–1620.

  53. Narendran A, Ganjavi H, Morson N, Connor A, Barlow JW, Keystone E, Malkin D and Freedman MH . (2003). Exp. Hematol., 31, 693–701.

  54. Neufeld G, Cohen T, Gengrinovitch S and Poltorak Z . (1999). FASEB J., 13, 9–22.

  55. Neufeld G, Kessler O and Herzog Y . (2002). Adv. Exp. Med. Biol., 515, 81–90.

  56. Orlandini M, Marconcini L, Ferruzzi R and Oliviero S . (1996). Proc. Natl. Acad. Sci. USA, 93, 11675–11680.

  57. Parry TJ, Cushman C, Gallegos AM, Agrawal AB, Richardson M, Andrews LE, Maloney L, Mokler VR, Wincott FE and Pavco PA . (1999). Nucleic Acids Res., 27, 2569–2577.

  58. Podar K, Catley LP, Tai YT, Shringarpure R, Carvalho P, Hayashi T, Burger R, Schlossman RL, Richardson PG, Pandite LN, Kumar R, Hideshima T, Chauhan D and Anderson KC . (2004). Blood, 103, 3474–3479.

  59. Podar K, Tai YT, Davies FE, Lentzsch S, Sattler M, Hideshima T, Lin BK, Gupta D, Shima Y, Chauhan D, Mitsiades C, Raje N, Richardson P and Anderson KC . (2001). Blood, 98, 428–435.

  60. Quinn TP, Peters KG, De Vries C, Ferrara N and Williams LT . (1993). Proc. Natl. Acad. Sci. USA, 90, 7533–7537.

  61. Ricaud S, Vernus B, Duclos M, Bernardi H, Ritvos O, Carnac G and Bonnieu A . (2003). Oncogene, 22, 8221–8232.

  62. Risau W . (1997). Nature, 386, 671–674.

  63. Schweigerer L, Neufeld G, Mergia A, Abraham JA, Fiddes JC and Gospodarowicz D . (1987). Proc. Natl. Acad. Sci. USA, 84, 842–846.

  64. Selvaraj SK, Giri RK, Perelman N, Johnson C, Malik P and Kalra VK . (2003). Blood, 102, 1515–1524.

  65. Shalaby F, Rossant J, Yamaguchi TP, Gertsenstein M, Wu XF, Breitman ML and Schuh AC . (1995). Nature, 376, 62–66.

  66. Suzuki K, Hayashi N, Miyamoto Y, Yamamoto M, Ohkawa K, Ito Y, Sasaki Y, Yamaguchi Y, Nakase H, Noda K, Enomoto N, Arai K, Yamada Y, Yoshihara H, Tujimura T, Kawano K, Yoshikawa K and Kamada T . (1996). Cancer Res., 56, 3004–3009.

  67. Tammela T, Enholm B, Alitalo K and Paavonen K . (2005). Cardiovasc. Res., 65, 550–563.

  68. Wakiya K, Begue A, Stehelin D and Shibuya M . (1996). J. Biol. Chem., 271, 30823–30828.

  69. Waltenberger J, Claesson-Welsh L, Siegbahn A, Shibuya M and Heldin CH . (1994). J. Biol. Chem., 269, 26988–26995.

  70. Weninger W, Rendl M, Mildner M and Tschachler E . (1998). J. Invest. Dermatol., 111, 907–911.

  71. Wey JS, Stoeltzing O and Ellis LM . (2004). Clin. Adv. Hematol. Oncol., 2, 37–45.

  72. Willett CG, Boucher Y, di Tomaso E, Duda DG, Munn LL, Tong RT, Chung DC, Sahani DV, Kalva SP, Kozin SV, Mino M, Cohen KS, Scadden DT, Hartford AC, Fischman AJ, Clark JW, Ryan DP, Zhu AX, Blaszkowsky LS, Chen HX, Shellito PC, Lauwers GY and Jain RK . (2004). Nat. Med., 10, 145–147.

  73. Young Jr JL, Ries LG, Silverberg E, Horm JW and Miller RW . (1986). Cancer, 58, 598–602.

  74. Zachary I and Gliki G . (2001). Cardiovasc. Res., 49, 568–581.

  75. Zachary I, Mathur A, Yla-Herttuala S and Martin J . (2000). Arterioscler. Thromb. Vasc. Biol., 20, 1512–1520.

  76. Zhang L, Zhan S, Navid F, Li Q, Choi YH, Kim M, Seth P and Helman LJ . (1998). Oncogene, 17, 1261–1270.

  77. Ziegler BL, Valtieri M, Porada GA, De Maria R, Muller R, Masella B, Gabbianelli M, Casella I, Pelosi E, Bock T, Zanjani ED and Peschle C . (1999). Science, 285, 1553–1558.

Download references

Acknowledgements

This work was funded by the Andrew Mizzoni Cancer Research Fund and the James Birrell Research Fund and through studentships of the Ontario Student Opportunity Trust Fund – Hospital for Sick Children Foundation Student Scholarship Program (MG, BD) and the Brainchild Foundation (BD). RT is supported in part by the Scotiabank Clinician-Scientist Training Program Fund and the Research Training Centre, Hospital for Sick Children, Toronto. DM is a Research Scientist of the National Cancer Institute of Canada/Canadian Cancer Society. We are grateful for helpful discussions with Drs Herman Yeger and Meredith Irwin.

Author information

Affiliations

Authors

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Gee, M., Tsuchida, R., Eichler-Jonsson, C. et al. Vascular endothelial growth factor acts in an autocrine manner in rhabdomyosarcoma cell lines and can be inhibited with all-trans-retinoic acid. Oncogene 24, 8025–8037 (2005). https://doi.org/10.1038/sj.onc.1208939

Download citation

Keywords

  • autocrine signalling
  • vascular endothelial growth factor
  • rhabdomyosarcoma
  • all-trans-retinoic acid

Further reading

Search

Quick links