0021-972X/01/$03.00/0 The Journal of Clinical Endocrinology & Metabolism Copyright © 2001 by The Endocrine Society
Vol. 86, No. 1 Printed in U.S.A.
Insulin Up-Regulates Vascular Endothelial Growth Factor and Stabilizes Its Messengers in Endometrial Adenocarcinoma Cells* ´ DE ´ RIQUE LAMIELLE, FABRICE LORCHEL, LAURENT BERMONT, FRE SYLVIE FAUCONNET, HIROYASU ESUMI, ALESSANDRO WEISZ, AND ´ RARD L. ADESSI GE Service d’Oncologie et d’Endocrinologie Mole´culaires (Centre Hospitalier Universitaire Besanc¸on and Institut d’Etude et de Transfert de Ge`nes) and Contract Recherche INSERM 96.01 (L.B., F.La., F.Lo., S.F., G.L.A.), 25000 Besanc¸on, France; Investigate Treatment Division, National Cancer Center Research Institute East (H.E.), Kashiwa, Chiba 277, Japan; and Istituto di Patologia Generale e Oncologia, Facolta` di Medicina e Chirurgia, Seconda Universita` di Napoli (A.W.), 80138 Naples, Italy ABSTRACT Angiogenesis is crucial for tumor growth and dissemination. Vascular endothelial growth factor (VEGF) is a potent angiogenic factor that promotes vascular growth and therefore tumoral growth and metastasis. Overweight, frequently associated with hyperinsulinemia, constitutes the major risk factor for endometrial carcinoma. Thus, elevated insulin levels may partly explain the increased risk of endometrial cancer observed in obese postmenopausal women. The aim of the present work was to test the role of insulin in the control of VEGF expression in endometrial carcinoma cells (HEC-1A). We have shown that insulin induced a biphasic expression of VEGF mes-
senger ribonucleic acid, with an early, but low, induction (4 h of stimulation) and a delayed, but high, induction (24 h). The delayed effect of insulin on VEGF expression involved transcriptional and posttranscriptional regulation, as evidenced by the increased rate of VEGF transcription and the prolonged half-life of VEGF messenger ribonucleic acid. Simultaneously we observed higher levels of VEGF protein in the conditioned medium of stimulated cells compared with unstimulated ones. Therefore, insulin could contribute to the increased risk of endometrial carcinoma due to its ability to induce VEGF expression and thus participate in the maintenance of an angiogenic phenotype. (J Clin Endocrinol Metab 86: 363–368, 2001)
A
NGIOGENESIS IS A multistep process essential for tumoral growth (1). The development of new blood vessels from preexisting capillaries requires the combined and coordinated actions of angiogenic factors, extracellular matrix components, as well as specific proteases. One of the most important mediators of angiogenesis is the vascular endothelial growth factor (VEGF), also called vascular permeability factor (VPF) or vasculotropin. VEGF has a specific mitogenic action on endothelial cells as well as a potent capacity for increasing vessel permeability. The VEGF protein characterized by Ferrara and Henzel (2) as well as by Gospodarowicz et al. (3) is a dimeric glycosylated protein of approximately 46 kDa that occurs as 5 isoforms of 121, 145, 165, 189, and 206 amino acids as a result of an alternatively spliced transcript from 1 gene (4, 5). The angiogenic activity of soluble isoforms depends on the presence of Flt-1 or KDR/Flk-1 tyrosine kinase receptors located in the cytoplasmic membrane of endothelial cells (6). In cycling human endometrium, VEGF expression is stronger in secretory than in proliferative endometrium and is prominent in the glandular compartment (7) suggesting hormonal regulation. In endometrial carcinoma, the mechanisms
responsible for angiogenesis are not well defined. Nevertheless, VEGF appears to notably contribute to tumor angiogenesis and thus to endometrial carcinoma growth (8 –10). The major risk factor of endometrial carcinoma is defined by overweight of more than 23 kg (11, 12). The metabolic disorders associated with obesity frequently involve hyperinsulinemia. Insulin is a growth factor known for its mitogenic activity with regard to carcinoma cell lines (13, 14). However, the biological role of insulin in endometrial adenocarcinoma is unclear, and although elevated insulin levels may partly explain the increased risk of endometrial cancer observed in obese postmenopausal women (12), the real implication of insulin in the growth of endometrial adenocarcinoma is still a matter of debate (15). Thus, the aim of our study was to investigate the effects of insulin on VEGF expression in endometrial carcinoma cells. The present work is the continuation of a former one (16) in which we showed that insulin-like growth factor I (IGF-I) increased VEGF expression in endometrial carcinoma cells. The similarities of the signaling pathways activated by these growth factors led us to study the potential role of insulin in tumoral angiogenesis via the regulation of VEGF expression.
Received June 12, 2000. Revision received September 22, 2000. Accepted October 5, 2000. Address all correspondence and requests for reprints to: Ge´rard L. Adessi, M.D., Ph.D., Baˆtiment Inserm, 240 route de Dole, 25000 Besanc¸on, France. E-mail:
[email protected]. * This work was supported by INSERM (Contract Grant 96.01) and the Association pour la Recherche sur le Cancer.
Materials and Methods Reagents Insulin and other biochemical reagents were purchased from Sigma (Saint Quentin, France) unless otherwise stated. Other chemicals were of the highest grade available.
363
364
BERMONT ET AL.
Cell lines and culture The HEC-1A, KLE, and RL 95–2 cell lines were purchased from American Type Culture Collection(Biovalley, Conches, France). HEC-1A cells were maintained in phenol red-free McCoy’s 5a medium supplemented with 5% FBS (Life Technologies, Inc., Cergy Pontoise, France) and a 1% antibiotic cocktail (10 mg/mL streptomycin, 10,000 U/mL penicillin, and 25 g/mL Amphotericin). KLE and RL 95–2 cells were maintained in phenol red-free DMEM/Ham’s F-12 medium supplemented with 10% FBS, 5 g/mL insulin, and a 1% antibiotic cocktail. Ishikawa cells (clone 3-H-12, no. 23) were provided by Dr. M. Nishida (Department of Obstetrics and Gynecology, Institute of Clinical Medicine, University of Tsukuba, Tsukuba, Japan) and maintained in DMEM with phenol red supplemented with 10% FBS and a 1% antibiotic cocktail. All cell lines were tested for the absence of mycoplasma before starting the experiments. The HEC-1A cell line was tested several times for the presence of estrogen receptors (ER) by an enzyme-linked immunosorbent assay (ELISA) method (Abbott, Rungis, France), and we failed to detect ER (⬍1 fmol/mg proteins). Castro-Rivera and Safe (17) have shown that HEC-1A cells expressed the wild-type form of the ER and were estradiol responsive. Thus, HEC-1A cells seem to have unsettled features. Therefore, to eliminate any estrogenic effect, HEC-1A cells as well as other cell lines were subcultured for 72 h in the presence of desteroided FBS (dextran-charcoal-treated FBS). For the experiments, cells were trypsinized and seeded in six-well plates (Falcon, Elvetec, Venissieux, France) for 72 h with 2% desteroided FBS, then placed in serum-free medium for 24 h before stimulation. HEC-1A cells were seeded in six-well plates with 1.5 ⫻ 105 and 2.5 ⫻ 105 cells/well for KLE, RL 95–2, and Ishikawa cells. Stimulations were performed in 1 mL of the appropriate serum-free medium supplemented with 1 mg/mL BSA and with or without insulin.
RNA isolation and Northern blotting analysis Total RNA from confluent cells was extracted with Tri-Reagent (Molecular Research Center, Inc., Euromedex, Souffelweyersheim, France). RNA samples (30 g) were electrophoresed through 1.2% agarose gels (SeaKem LE agarose, TEBU, Le Perray-en-Yvelines, France) for 3 h at 80 V, followed by 1 h at 100 V, then transferred to nylon membranes (Zeta-Probe GT Genomic, Bio-Rad Laboratories, Inc., Ivry-sur-Seine, France) using a vacuum blotting system and cross-linked to the membranes by heating for 1 h at 80 C. Blots were concomitantly hybridized to a VEGF complementary DNA (cDNA) probe (18) and to a 0.3-kb human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) probe for 48 h at 42 C with gentle agitation. The cDNA probes were labeled with [␣-32P]deoxy-CTP (NEN Life Science Products, Paris, France) using the random hexamer labeling method (Prime-a-Gene Labeling System, Promega Corp., Lyon, France). Final washes were in 0.3 ⫻ SSC (standard saline citrate)-0.1% SDS for 15 min at 55 C. Membranes were then exposed to the Imaging Screen-CS (Bio-Rad Laboratories, Inc.), and image analysis was performed using MultiAnalyst software (Bio-Rad Laboratories, Inc.).
JCE & M • 2001 Vol. 86 • No. 1
at 500 ⫻ g at 4 C for 5 min, and the lysis step was repeated once. The nuclei were then resuspended in 200 L in 40% glycerol, 5 mmol/L MgCl2, 0.1 mmol/L ethylenediamine tetraacetate, and 50 mmol/L TrisHCl, pH 8.3. The transcription assay was performed by adding to the nuclei suspension 200 L transcription buffer [10 mmol/L Tris-HCl (pH 8), 5 mmol/L MgCl2, 300 mmol/L KCl, and 0.5 mmol/L dithiothreitol] containing 1 mmol/L each of ATP, CTP, and GTP and 100 mCi [␣-32P]UTP (SA, 3000 Ci/mmol; NEN Life Science Products) and incubated for 30 min at 30 C. The sample volumes were adjusted to 1 mL with HSB buffer [10 mmol/L Tris-HCl (pH 7.4), 500 mmol/L NaCl, 50 mmol/L MgCl2, and 2 mmol/L CaCl2] containing ribonuclease-free deoxyribonuclease I (20 U/sample) and incubated for 15 min at 30 C. Each sample was then treated with 200 L Tris-SDS buffer [500 mmol/L Tris-HCl (pH 7.4), 125 mmol/L ethylenediamine tetraacetate, and 0.5% SDS] and 10 L of a 20 mg/mL protein K solution and incubated for 20 min at 42 C. Nuclear RNA was isolated by the addition of 1 mL Tri-Reagent (Molecular Research Center, Inc., Euromedex) and 200 L chloroform. After 10 min at room temperature, the mixture was centrifuged at 12,000 ⫻ g at 4 C for 15 min. The upper aqueous layer was removed and combined with an equal volume of isopropyl alcohol and incubated at ⫺70 C overnight. Afterward, samples were centrifuged at 10,000 ⫻ g at 4 C for 20 min, and the pellets were washed with 70% ethanol and dissolved in diethylpyrocarbonate-treated water. The labeled RNA obtained was used as a probe for hybridization. Before hybridization, RNA was heated at 80 C for 1 min. Nylon membranes (Zeta-Probe GT Genomic, Bio-Rad Laboratories, Inc.) containing 5 g each of VEGF and GAPDH cDNA were prepared. Five micrograms of DNA in 100 L sterile water were heated at 95 C for 10 min and placed on ice. The single stranded DNA was slot-blotted using a Millipore Corp. manifold (Bedford, MA), and the membranes were baked at 80 C for 1 h. Filters were prehybridized in 50% formamide, 5 ⫻ SSC, 7% SDS, 50 mmol/L sodium phosphate, 2% blocking reagent (Roche Molecular Biochemicals, Meylan, France), 0.1% N-lauryl sarcosine, and 50 g/mL yeast RNA at 42 C for 30 min. The filters were hybridized to the run-on products in 1 mL hybridization solution at 42 C for 48 h. They were washed successively in 2 ⫻ SSC-0.1% SDS and 1 ⫻ SSC-0.1% SDS at room temperature and finally in 0.3 ⫻ SSC-0.1% SDS at 55 C and were exposed to the Imaging Screen Cassette-CS (Bio-Rad Laboratories, Inc.). Densitometric scanning was performed with a BioRad Laboratories, Inc., GS-505 Molecular Imager system, and image analysis was conducted using MultiAnalyst software (Bio-Rad Laboratories, Inc.). The data were reported as relative increases in the transcription rate of VEGF after normalizing to GAPDH transcriptional rates.
Statistical analysis When appropriate, data from image analysis and VEGF protein quantification were expressed as the mean ⫾95% confidence limits. Statistical analysis to test the significance of the differences was performed using t test, and P ⬍ 0.05 was considered significant. Before performing the t test, the data were tested using an F test for their variance homogeneities.
VEGF messenger RNA (mRNA) stability After a 24-h incubation in a serum-free medium with or without 10 nmol/L insulin, actinomycin D (final concentration, 10 g/mL) was added to block transcription. Cells were harvested 0, 1, 2, 4, 6, and 8 h after the addition of actinomycin D, and Northern blot analysis was performed for VEGF mRNA expression. Densitometric scanning was conducted using the Bio-Rad Laboratories, Inc. GS-505 Molecular Imager system.
VEGF mRNA run-on analysis Nuclei were prepared from HEC-1A cells incubated with 10 nmol/L insulin for 4 or 24 h in 75-cm2 flasks. Untreated cells were incubated with McCoy’s 5a medium supplemented with 1 mg/mL BSA. At the end of the incubation cells were rinsed twice with ice-cold PBS and scraped. The cells were then pelleted at 500 ⫻ g at 4 C for 5 min, and the supernatants were discarded. Cells were lysed in ice-cold lysis buffer [10 mmol/L Tris-HCl (pH 7.4), 10 mmol/L NaCl, 3 mmol/L MgCl2, and 0.5% Nonidet P-40] and allowed to swell on ice for 5 min. The nuclei were pelleted
Results Induction of VEGF mRNA expression by insulin
HEC-1A cells were grown for various time periods with or without 10 nmol/L insulin, and expression of VEGF mRNA was then determined by Northern blotting. Figure 1 shows that the VEGF cDNA probe hybridizes with a 4.5- and a 3.7-kb mRNA, the former being more abundant than the latter. There was also a faint and inconstantly visible band at 5.2 kb that was not quantified. In insulin-treated cells, a biphasic response of VEGF mRNA was observed, with a faint peak after 4 h of stimulation and a more important one after 24 h. The maximum induction (3.5- and 1.9-fold for the 4.5and 3.7-kb mRNA, respectively) was reached after a 24-h incubation period with insulin compared with unstimulated cells for the same time period (P ⬍ 0.05; Fig. 1B). The dose-
REGULATION OF VEGF BY INSULIN IN ENDOMETRIAL CANCER
365
and 48-h incubation periods (1.8- and 2.1-fold; P ⬍ 0.002 and P ⬍ 0.01 respectively; Fig. 3A). The insulin effect after a 24-h incubation was dose dependent, and the maximum value was obtained with 100 nmol/L insulin (2.2-fold; P ⬍ 10⫺5; Fig. 3B). Transcriptional and posttranscriptional regulation of VEGF expression by insulin
To define the level of regulation of VEGF expression by insulin, the half-life as well as the transcription rate of VEGF mRNA in insulin-stimulated HEC-1A cells were determined. In unstimulated cells, the half-lives of the 4.5- and 3.7-kb mRNA were statistically different and were, respectively, 0.8 ⫾ 0.1 and 1.4 ⫾ 0.1 h (P ⬍ 0.01). The half-lives of both messengers were prolonged in HEC-1A cells treated for 24 h with 10 nmol/L insulin (Fig. 4) and were, respectively, 1.8 ⫾ 0.1 and 2.6 ⫾ 0.2 h. The differences (stimulated vs. unstimulated) were statistically significant for both mRNAs (P ⬍ 0.001). Concurrently, the run-on analysis showed that after 4 h of stimulation (Fig. 5), insulin induced a faint increase in the VEGF transcription rate (1.3-fold), which could be the cause of the faint peak observed by Northern blotting. In other respects, after 24 h of stimulation, insulin induced a 3.4-fold increase in the VEGF transcription rate compared with that in unstimulated cells (Fig. 5). However, the run-on analysis does not permit a study of both the 4.5- and 3.7-kb mRNAs separately, and the increase in the transcription rate therefore corresponds to both transcripts. Thus, it appeared that insulin regulated VEGF expression at the transcriptional and posttranscriptional levels in HEC-1A cells. Effect of insulin on VEGF mRNA expression in endometrial adenocarcinoma cell lines
FIG. 1. Time dependence of VEGF mRNA expression by insulin in HEC-1A. A, Cells were incubated for various periods with or without 10 nmol/L insulin, and VEGF mRNA was determined by Northern blotting. Filters were cohybridized with a GAPDH probe to assess loading differences. A representative result of three independent ones is shown. B, The band intensities corresponding to the 4.5- and 3.7-kb mRNAs were quantified using the Molecular Imager system and were reported in arbitrary units relative to VEGF mRNA expression at time zero after normalization for variances in GAPDH expression Results are expressed as the mean ⫾ 95% confidence limits of three independent experiments. *, P ⬍ 0.05, insulin stimulated vs. unstimulated.
dependent effect of insulin on VEGF mRNA expression in HEC-1A cells was then determined (Fig. 2A). A 24-h incubation period of HEC-1A cells with various concentrations of insulin showed an increase in VEGF mRNA expression (Fig. 2B). An ANOVA showed that insulin induced a significant increase in both the 4.5- and 3.7-kb VEGF mRNA (F ⫽ 16.3; P ⫽ 1.6 ⫻ 10⫺4 and F ⫽ 6.2; P ⫽ 5 ⫻ 10⫺3, respectively). Subsequently, quantification of VEGF protein in the conditioned medium of HEC-1A cells was performed. A significant increase in the amounts of soluble VEGF isoforms was observed in the conditioned medium of cells stimulated with 10 nmol/L insulin, compared with untreated cells, after 24-
The effect of insulin on VEGF mRNA expression was tested in several endometrial adenocarcinoma cell lines. Figure 6 shows that insulin enhanced VEGF mRNA expression after 4- and 24-h incubation periods in HEC-1A and Ishikawa cells, whereas there was no response in KLE and RL 95–2 cells, which, moreover, respectively expressed moderate and low levels of VEGF mRNA in these conditions. Discussion
In the present work we showed that insulin stimulated VEGF expression in HEC-1A cells in a biphasic manner, with an early increase (4 h of stimulation) characterized by low levels of VEGF mRNA and a delayed increase (24 h of stimulation) characterized by high levels of VEGF mRNA. The delayed and most important effect of insulin on VEGF expression involves transcriptional and posttranscriptional actions. As shown by Zelzer et al. (19), insulin can induce HIF-1␣/Aryl hydrocarbon Receptor Nuclear Translocator transcription complex formation, which specifically binds the hypoxia response element present in the promoter of numerous genes, such as VEGF or EPO (erythropoietin), and thus induces their transcription. However, the protein kinase B and mitogen-activated protein kinase signaling pathways usually activated by insulin could be implicated, directly or indirectly, in the regulation of VEGF expression. The mech-
366
BERMONT ET AL.
JCE & M • 2001 Vol. 86 • No. 1
FIG. 2. Dose-dependent effect of insulin on VEGF mRNA expression. A, HEC-1A cells were incubated for 24 h with various insulin concentrations, and VEGF mRNA was determined by Northern blotting. Filters were cohybridized with a GAPDH probe to assess loading differences. B, The band intensities corresponding to the 4.5- and 3.7-kb mRNAs were quantified using the Molecular Imager system and were reported in arbitrary units relative to VEGF mRNA expression at time zero after normalization for variances in GAPDH expression. Results are expressed as the mean ⫾ 95% confidence limits of three independent experiments. Differences were significant for all concentrations tested vs. unstimulated cells except for the 4.5-kb mRNA at 10⫺10 mol/L insulin, for which P ⬍ 0.075.
anisms responsible for VEGF mRNA stabilization are not well understood, but it appeared that 3⬘-untranslated regionspecific sequences were implicated. Indeed, Shih and Claffey (20) identified a hypoxia-sensible region in the 3⬘-untranslated region of VEGF mRNA that is able to specifically interact with heterogeneous nuclear ribonucleoprotein L and reduce mRNA degradation. As hypoxia shares with insulin the ability to regulate the expression of some stress-induced genes by similar mechanisms (19), it can be hypothesized that insulin also stabilizes VEGF mRNA via an increase in heterogeneous nuclear ribonucleoprotein L expression. The modest effect of insulin on VEGF protein expression could result from the behavior of the different VEGF isoforms. Thus, after the translation of VEGF messengers, only VEGF121 and a part of VEGF165 proteins are soluble forms in the extracellular medium and are detected by ELISA assays, whereas other isoforms escape from the ELISA quantification and could thus contribute to an underestimation of VEGF protein quantification. The VEGF cDNA probe hybridized with three different messengers of 5.2, 4.5, and 3.7 kb, which were reported in a previous work (16). The relation between these different messengers and the major VEGF isoforms (189, 165, and 121 amino acids) has not been completely elucidated, and a link can be substantiated between the 5.2-, 4.5-, and 3.7-kb mRNA and the 189-, 165-, and 121-VEGF isoforms, respectively.
However, the existence of an alternative initiation site of transcription 633 bp downstream of the main transcription start site (21) may enhance the complexity of VEGF regulation. The faint VEGF expression in both KLE and RL 95–2 cells could be the result of low levels of insulin receptor expression. Nagamani and Stuart (14) have shown the presence of insulin receptors in five human endometrial carcinoma cell lines (HEC-1A, HEC-1B, RL 95–2, KLE, and AN3 CA), but at low levels on RL 95–2 cells. However, insulin stimulated the cell growth of all cell lines tested, and it can be suggested that mitogenic and VEGF stimulations by insulin may be transduced by different pathways. Another explanation refers to the degree of differentiation of these cell lines. Fujimoto et al. (10) analyzed VEGF expression in endometrial cancers and found that VEGF protein levels were higher in well differentiated cells (G1) compared with those in moderately (G2) and poorly (G3) differentiated cells. The lowest levels of VEGF mRNA were observed in the KLE and RL 95–2 cell lines, which are, respectively, poorly and moderately differentiated cells, while HEC-1A and Ishikawa cells, which expressed higher levels of VEGF mRNA, are moderately to well differentiated cells. Thus, VEGF expression could be decreased during endometrial cancer progression with dedifferentiation and might therefore contribute to the early process of tumoral growth via angiogenic activity. In a previous
REGULATION OF VEGF BY INSULIN IN ENDOMETRIAL CANCER
FIG. 3. Time dependence (A) and dose dependence (B) of insulin on VEGF protein expression in conditioned medium of HEC-1A cells. Cells were incubated for various periods with or without 10 nmol/L insulin (A) or for a 24-h period with various concentrations of insulin (B). The ELISA quantification of VEGF protein was then performed in the conditioned medium and normalized relative to the total cellular protein concentrations. The quantifications were performed in triplicate in three independent experiments. *, P ⬍ 0.05, insulin stimulated vs. unstimulated cells.
work (16) we observed a similar profile of VEGF mRNA expression in these four cell lines stimulated with IGF-I, except in the Ishikawa cell line, in which IGF-I failed to increase VEGF mRNA expression. This difference could be the result of large amounts of IGF-binding proteins (IGFBPs), especially IGFBP-3 secreted by Ishikawa cells, which inhibit the action of IGF-I (22). Here, the low affinity of insulin for the IGFBPs does not impede insulin action on Ishikawa cells. Nevertheless, it cannot be excluded that insulin might regulate the expression of other angiogenic factors by endometrial carcinoma cells, namely VEGF B, VEGF C, VEGF D, or basic fibroblast growth factor, which could cooperate with VEGF to promote vascular growth. As the ability of insulin to induce VEGF expression is unlikely to be restricted to tumoral cells, such activity is likely to play a significant role in other pathologies characterized by neovascularization, such as diabetic retinopathy. In summary, insulin could contribute to vascular growth due to its ability to regulate VEGF expression in endometrial
367
FIG. 4. Stability of VEGF mRNAs. HEC-1A cells were grown for 24 h with or without 10 nmol/L insulin before adding actinomycin D (10 g/mL). Total RNA was extracted at the indicated time after actinomycin D addition, VEGF mRNA was determined by Northern blotting, and results were quantitated using a densitometer. Loading differences were corrected by normalization for variances in GAPDH expression, and relative VEGF mRNA expression was plotted as a percentage of the zero time value. One representative of three replicate experiments is shown.
FIG. 5. Transcriptional activation of VEGF by insulin. Nuclear run-on transcription analysis was carried out, as described in Materials and Methods, with nuclei of HEC-1A cells grown for 4 h or 24 h with or without 10 nmol/L insulin. The relative fold induction refers to the ratio of relative amounts of VEGF mRNA synthesized in treated cells relative to untreated cells after normalization for GAPDH mRNA expression. Values represent the means of 3 independent experiments. Error bars represent 95% confidence limits (*; P ⬍ 0.0003).
368
BERMONT ET AL.
JCE & M • 2001 Vol. 86 • No. 1
FIG. 6. Effect of insulin on VEGF mRNA expression in endometrial adenocarcinoma cell lines. Various confluent endometrial adenocarcinoma cell lines were incubated with or without 10 nmol/L insulin for 4 and 24 h. VEGF mRNAs expression was then determined by Northern blotting. The data obtained by densitometric analysis were normalized relative to GAPDH mRNA expression. One representative result of two independent experiments is shown.
carcinoma cells. Thus, the increased risk of endometrial carcinoma linked with severe obesity might be partially due to the hyperinsulinemia via the induction of VEGF expression, a potent angiogenic factor, by tumoral cells. Acknowledgments The excellent technical assistance of Ms. C. Vial, C. Colombain, E. Chezy, and C. Ferniot is acknowledged.
References 1. Folkman J 1992 The role of angiogenesis in tumor growth. Semin Cancer Biol. 3:65–71. 2. Ferrara N, Henzel WJ 1989 Pituitary follicular cells secrete a novel heparinbinding growth factor specific for vascular endothelial cells. Biochem Biophys Res Commun. 161:851– 858. 3. Gospodarowicz D, Abraham JA, Schilling J 1989 Isolation and characterization of a vascular endothelial cell mitogen produced by pituitary-derived folliculo stellate cells. Proc Natl Acad Sci USA. 86:7311–7315. 4. Houck KA, Leung DW, Rowland AM, Winer J, Ferrara N 1992 Dual regulation of vascular endothelial growth factor bioavailibility by genetic and proteolytic mechanisms. J Biol Chem. 267:26031–26037. 5. Tischer E, Mitchell R, Hartman T, Silva T, Gospodarowicz D, Fiddes JC, Abraham JA 1991 The human gene for vascular endothelial growth factor. Multiple protein forms are encoded through alternative exon splicing. J Biol Chem. 266:11946 –11954. 6. Vaisman N, Gospodarowicz D, Neufeld G 1990 Characterization of the receptors for vascular endothelial growth factor. J Biol Chem. 265:19461–19466. 7. Torry DS, Holt VJ, Keenan JA, Harris G, Caudle MR, Torry RJ 1996 Vascular endothelial growth factor expression in cycling human endometrium. Fertil Steril. 66:72– 80. 8. Guidi AJ, Abu-Jawdeh G, Tognazzi K, Dvorak HF, Brown LF 1996 Expression of vascular permeability factor (vascular endothelial growth factor) and its receptors in endometrial carcinoma. Cancer. 78:454 – 460. 9. Doldi N, Bassan M, Gulisano M, Broccoli V, Boncinelli E, Ferrari A 1996 Vascular endothelial growth factor messenger ribonucleic acid expression in human ovarian and endometrial cancer. Gynecol Endocrinol. 10:375–382.
10. Fujimoto J, Ichigo S, Hirose R, Sakaguchi H, Tamaya T 1998 Expressions of vascular endothelial growth factor (VEGF) and its mRNA in uterine endometrial cancers. Cancer Lett. 134:15–22. 11. Dubeau L 1993 Etiology and detection of gynecologic cancer. In: Morrow CP, Curtin JP, Townsend DE, eds. Synopsis of gynecologic oncology, 4th Ed. New York: Churchill Livingstone; 1–22. 12. Rose PG 1996 Endometrial carcinoma. N Engl J Med. 335:640 – 649. 13. Pearl ML, Talavera F, Gretz III HF, Roberts JA, Menon KMJ 1993 Mitogenic activity of growth factors in the human endometrial adenocarcinoma cell lines HEC-1A and KLE. Gynecol Oncol. 49:325–332. 14. Nagamani M, Stuart CA 1998 Specific binding and growth-promoting activity of insulin in endometrial cancer cells in culture. Am J Obstet Gynecol. 179:6 –12. 15. Troisi R, Potischman N, Hoover RN, Siiteri P, Brinton LA 1997 Insulin and endometrial cancer. Am J Epidemiol. 146:476 – 482. 16. Bermont L, Lamielle F, Fauconnet S, Esumi H, Weisz A, Adessi GL 2000 Regulation of vascular endothelial growth factor expression by insulin-like growth factor-I in endometrial adenocarcinoma cells. Int J Cancer. 85:117–123. 17. Castro-Rivera E, Safe S 1998 Estrogen- and antiestrogen-responsiveness of HEC-1A endometrial adenocarcinoma cells in culture. J Steroid Biochem Mol Biol. 64:287–295. 18. Chin K, Kurashima Y, Ogura T, Tajiri H, Yoshida S, Esumi H 1997 Induction of vascular endothelial growth factor by nitric oxide in human glioblastoma and hepatocellular carcinoma cells. Oncogene. 15:437– 442. 19. Zelzer E, Levy Y, Kahana C, Shilo BZ, Rubinstein M, Cohen B 1998 Insulin induces transcription of target genes through the hypoxia-inducible factor HIF-1␣/ARNT. EMBO J. 17:5085–5094. 20. Shih SC, Claffey KP 1999 Regulation of vascular endothelial growth factor mRNA stability in hypoxia by heterogeneous nuclear ribonucleoprotein L. J Biol Chem. 274:1359 –1365. 21. Akiri G, Nahari D, Finkelstein Y, Le SY, Elroy-Stein O, Levi BZ 1998 Regulation of vascular endothelial growth factor (VEGF) expression is mediated by internal initiation of translation and alternative initiation of transcription. Oncogene. 17:227–236. 22. Karas M, Danilenko M, Fishman D, LeRoith D, Levy J, Sharoni Y 1997 Membrane-associated insulin-like growth factor binding protein-3 inhibits insulin-like growth factor-I-induced insulin-like growth factor-I receptor signaling in Ishikawa endometrial cancer cells. J Biol Chem. 272:16514 –16520.
