Epidermal Growth Factor Suppresses Insulin-Like Growth Factor Binding

Home , Epidermal growth factor

(CANCER RESEARCH54, 3160-3166, June 15, 19941 Epidermal Growth Factor Suppresses Insulin-like Growth Factor Binding Protein 3 Levels in Human Papillomavirus Type 16-immortalized Cervical Epithelial Cells and Thereby Potentiates the Effects of Insulin-like Growth Factor 1' Joan R. Hembree,2Chapla Agarwal, and Richard L. Ecke& Departments of Physiology and Biophysics (J. R. H., C. A., R. L E.J, Demwtology [R. L. E.J, Reproductive Biology fR L. E.J, and Biochemistiy [J. R. H., R. L E.], Case Western Reserve University School ofMedicine, CIeveIand@Ohio 44106-4970

ABSTRACT growth. The types and quantities of IGFBPs produced are dependent on cell type and are influenced by a variety of hormones and growth Human ectocervical epithelial cells are a primary target for infection by factors (7â€”9).The effects of changes in IGFBP levels and/or type oncogemc papillomaviruses, which are strongly implicated as causative on IGF-I action are complex, in some cases potentiating (10â€”13) agents in the genesis of cervical cancer. Growth factors have been Impli and in others inhibiting (10, 14â€”16)IGF-I action. In the case of cated as agents that stimulate proliferation and enhance the possibility of malignant transformation. In the present study we utilize several human inhibition, it appears that soluble IGFBPs sequester IGFs and paplllomavirus (HPV) type 16-Immortalized eCtOCerVIcaIepithellal cell prevent their interaction with cell surface-associated IGFBPs lines to Investigate the effects of epidermal growth factor (EGF) and and/or IGF receptors (14â€”16). Potentiation of IGF-I action by insulin-like growth factor I (IGF-I) on cell proliferation and the produc IGFBP-3 has recently been attributed to association of IGFBP-3 tion of IGF binding proteins (IGFBPs). ECE16-1 cells, an HPV16-immor with the cell surface and subsequent alteration of IGFBP-3 and talized/nontumorigenic cell line, maintaIned in defined medium, produce IGF-I receptor activity (17). The studies indicate that IGF activity and release high levels ofIGFBP-3 (38/42 kDa) as well as smaller amounts is tightly controlled by IGFBPs, suggesting that regulation of ofa 24-kDa IGFBP. Supplementation ofdefined medium with EGF causes IGFBPs by other growth factors could have profound effects on a dose-dependent increase In cell growth and a concomitant decrease in IGF-mediated cellular responses. the levels of IGFBP-3 released Into the culture medium. EGF suppression of IGFBP-3 is maintained even when EGF-stimulated cell growth is The ectocervical epithelium is the target of the oncogenic HPV suppressed 67% due to the simultaneous presence of 3 ag/mI of TGFÃ˜1, forms (HPV16, HPV18, HPV31, and HPV33) which are etiological Indicating that EGF suppression ofIGFBP-3 levels is independent of EGF agents in the development ofcervical carcinoma (18â€”20).Members of effects on cell growth. EGF suppression of IGFBP-3 production is carve the EGF and IGF families are thought to contribute to the process of lated with a reduction in IGFBP-3 mRNA leveL In the presence of EGF, malignant transformation (21â€”24).However, the role of such growth the growth response of the cells to ng amounts of IGF-I Is significantly factors in the process of cancer development in cervical epithelial enhanced. Moreover, the simultaneous presence of both EGF and IGF-I cells is poorly understood. Members of the EGF and IGF families are reduces the level of IGFBP-3 more emciently than EGF alone. We also recognized mitogens for epithelia in vitro. At least two members of observe that the IGFBP-3 level is decreased and the 24-kDa IGFBP level the EGF family, TGFa (25) and amphiregulin (26) are produced by is Increased In HPV16-posltlve tumorigenic versus nontumorigenic cell epithelial cells. IGF-I is not produced by epithelial cells (27, 28); lines. This is the first report ofEGF acting as a positive regulator of IGF-I however, IGFs derived from the underlying fibroblasts accumulate in action via the IGFBPs. On the basis of these findings, we propose that EGF stimulates ECE16-1 cell growth via a dual-action mechanism by (a) the epithelial cell layers. IGFBPs have been reported to be produced stimulating growth directly via the EGF mitogenic pathway and (b) by the simple epithelium of the kidney (2), intestine (28), and breast stimulating growth indirectly by reducing the levels of inhibitory IGFBPs epithelium (4, 29, 30). However, the types of IGFBPs produced by and thereby potentiating the effects ofIGF-L In addition, the observation cervical epithelium, a nonkeratinizing, squamous epithelial cell that more highly transformed cell types produce lower levels of IGFBP-3 type, have not been characterized. and higher levels of 24-kDa IGFBP suggests that tumor cells in more In this study we examine the interplay between EGF and IGF-I in advanced cervical cancers may have an altered response to IGF-L the regulation of cervical epithelial cell growth and IGFBP production using an HPV16-immortalized ectocervical epithelial cell line, ECE16-1 (31). ECE16-1 cells were generated by transfection of INTRODUCTION primary cultured ectocervical epithelial cells and are a well-charac The mitogenic activity of the IGFs4 is influenced by the presence of terized model for the study of HPV16 effects on cervical cell function IGFBPs (1). Both IGFBPs and IGFs are produced by a wide variety (31, 32). We present evidence that EGF stimulates ECE16-1 cell of cells (2â€”6),suggesting that the balance of these factors in the growth via a dual-action mechanism by (a) stimulating growth di cellular microenvironment plays an important role in regulating cell rectly via the EGF mitogenic pathway and (b) stimulating growth indirectly by reducing the levels of inhibitory IGFBPs and thereby Received 7/9/93; accepted 4/15/94. potentiating the effects of IGF-I. IGFBPs represent a previously The costs of publication of this article were defrayed in part by the payment of page unrecognized link between the EGF and the IGF signaling systems. charges. This article must therefore be hereby marked advertisement in accordance with A reduction in IGFBP levels via an EGF-dependent mechanism 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by a grant from the American Institute for Cancer Research may be an important mechanism for the promotion of cervical (R. L. E.) and utilized the facilities of the Skin Diseases Research Center of Northeast tumor development. Ohio (NIH AR49750).

2 Supported as a postdoctoral trainee in the Metabolism Training Program (NIH DK07319). 3 To whom requests for reprints should be addressed, at Department of Physiology and MATERIALS AND METHODS Biophysics, Case Western Reserve University School of Medicine, 2109 Adelbert Road, Cleveland, OH 44106-4970. Chemicals. Dulbecco's modified Eagle's medium, F-12 medium, nones 4 The abbreviations used are: IGF, insulin-like growth factor, IGFBP, insulin-like sential amino acids, L-glutamine, trypsin, and antibiotics were purchased from growth factor binding protein; IGF-I, insulin-like growth factor I; HPV, human papillo mavirus; EGF, epidermal growth factor; TGFa, transforming growth factor a; DM, Gibco (Grand Island, NY). Fetal calf serum, insulin, bovine serum albumin, defined medium; TBS, 10 msi Tris-HC1(pH 7.4), 150 mat Naa, and 0.2% sodium azide; hydrocortisone, T3-transferrin, and adenine were obtained from Sigma (1cm cDNA, complementary DNA. icalCo.(St. Louis,MO).CholeratoxinwaspurchasedfromICNBiomedicals 3160

(Costa Mesa, CA). Human recombinant EGF (Upstate Biotechnology Inc., days with intensifying screens. The intensity of the signal was determined by Lake Placid, NY), human recombinant IGF-I (Collaborative Biomedical Prod densitometry. ucts, Bedford, MA), or human TGF@1(R & D Systems, Minneapolis, MN) Nucleic Acid Methods. RNA was isolated as previously described (38) was added to medium as indicated in the text. (3-[1@I]iodotyrosyl)IGF-I and from ECE16-1 cells treatedas described in the text. PolyadenylatedRNA, [32P]dCl'P were purchased from Amersham (Arlington Heights, IL). selected from total RNA by binding to oligodeoxythymidylate cellulose, was Prestained molecular weight standards were purchased from Bio-Rad Labors fractionated on a formaldehyde-containing agarose gel (39) and transferred to tories (Richmond, CA). R7 rabbit anti-human IGFBP-3 antiserum was kindly Biodyne A membranes. The membrane was hybridized with cDNAs encoding provided by Dr. R. L Baxter (Royal Prince Alfred Hospital, Camperdown, human IGFBP-3 or glyceraldehyde-3-phosphate dehydrogenase labeled with Australia). EGF neutralizing antibody AB-236-NA was obtained from R & D [32PJdCTPby random priming. Autoradiograms of Northern blots were quan Systems. Human IGFBP-3 cDNA, spanning nucleotides â€”50to + 1264 base titated by densitometry. Statistics. Data were analyzed using analysis of variance and Dunnett's pairs relative to the translation start codon (33) was kindly provided by Dr. D. multiple range test for comparison with a single control group. Differences R. Powell (Baylor College of Medicine, Houston, TX). Human glyceralde between two groups were analyzed by Student's t test for two independent hyde-3-phosphate dehydrogenase partial cDNA (34) was kindly provided by samples. Experiments were performed in duplicate or triplicate and repeated at Dr. R. Wu (Cornell University, Ithaca, NY). Proteins were transferred to least twice or as indicated in the figure legends. nitrocellulose membranes (0.45 pm; Schleicher and Schuell, Keene, NH) and RNA was transferredto BiodyneA membranes(ICN Biomedicals,Costa Mesa, CA). Densitometry of autoradiograms was performed using a SciScan RESULTS 5000 optical scanner (United States Biochemicals, Cleveland, OH). EGF and TGF@1Regulation of ECE16-1 Cell Proliferation. To Cell Proliferation Studies. ECE16-1 cells were grown as previouslyde investigate the effects of individual growth factors on ECE16-1 cell scribed (31). Cultures were maintained in Dulbecco's modified Eagle's mcdi um:F-12 medium (3:1) supplementedwith 5% fetal calf serum, 5 pg/cd of proliferation, it was necessary to devise a serum- and growth factor insulin, 0.1 flMcholera toxin, 5 @Wmloftransferrin, 1 nMT3, 10 ng/ml of free medium for cell maintenance. As shown in Fig. 1, ECE16-1 cell EGF,0.18 mMadenine,nonessentialaminoacids, L-glutamine,andantibiotics number declines slightly when the cells are grown in DM medium or (100 units/nil of penicillin, 100 pg/cd of streptomycin, and 5 @.tg/cdof in DM supplemented with 3 ng/ml of [email protected] contrast, 3 days of gentamicin). For proliferation studies the cells were plated at 10,000 cells/cm2 treatment with 20 ng of EGF/ml increases cell number by 2.5-fold into 35-mm culture dishes in the above described medium. After a 24-h (day 3) whereas coincubation with EGF and TGF(31 reduced EGF attachment period, the medium was changed to a DM. DM consists of Thai stimulated growth by 67%. The level of TGF@31chosen for these becco's modified Eagle's medium:F-12 (3:1) supplemented with 1 mg/ml of studies, 3 ng/ml, maximally inhibits EGF-stimulated ECE16-1 cell bovine serum albumin, 50 g@g/mlofascorbic acid, nonessential amino acids, growth (not shown). L-glutamine,5 p@g/mloftransferrin,1 flMT3, 0.18 mMadenine, and antibiotics. EGF and TGF@31Regulation of IGFBP Levels. Conditioned After a 24-h incubation in DM, DM Â±growth factor was added and fresh medium was harvested from cells on days 1, 2, and 3 of the experi medium was added at subsequent 24-h intervals. For cell counting, cells were ment shown in Fig. 1 and IGFBP content was evaluated by ligand blot harvested with trypsin and then resuspended in an isotonic solution containing analysis (Fig. 2). ECE16-1 cells growing in DM release two major 3.7% formaldehyde for Coulter counting. Medium conditioned for 24 h by species of IGFBPs, a 24-kDa band and a complex spanning 38/42 cells was collected, centrifuged to remove cell debris, and stored at â€”20Â°C until analysis of IGFBP content. ECE16-D1 and ECE16-D2 cells (HPV16- kDa. In addition, a faint band at about 36 kDa was present in some positive, immortalized/nontumorigenic)5and CaSki and SiHa cells (HPV16- conditioned medium samples. As determined by densitometry, EGF positive, transformed/tumorigenic) (35, 36) were maintained as outlined for treatment for 2 days significantly reduced IGFBP levels compared to ECE16-1cells. cells growing in DM. The densitometry-derived levels are listed in Analysis of IGF Binding PrOteIns. The composition and relative abun Table 1. TGF@31also decreased IGFBP production, an effect that was dance of IGFBPs released into medium conditioned by ECE16-l cells were apparent after 48 h of treatment. Combined treatment with EGF and determined by ligand blot analysis as described by Hossenlopp et aL (37). Briefly, each sample ofconditioned medium was diluted in tracking dye so that an equivalent volume of medium, normalized based on cell number, was â€”â€¢â€”DM T 25 @ loaded in each lane within a geL The volume loaded ranged from 5,000 to â€”A 10,000 cell equivalents/laneinvariousexperiments.Thesampleswere heated r â€”aâ€” to 100Â°Cfor 10 min and 50 p.1/lanewere electrophoresed through a 10% 0 @ polyacrylamide gel under nonreducing conditions (37). In addition to the 20 @â€¢â€”E@+T@B1 conditioned medium samples, a 50-pJ sample from a standardized pool of K human cord serum was loaded onto each gel and used as an internal standard to facilitate comparisons between blots. Proteins were transferred to nitrocel 0 lulose by electrophoresis for 60 V for 16 h, followed by 70 V for 2 h. After the @15 +QF membranewas blocked, 1@I-IGF-I(specific activity, 2000 Ci/mmol) was addedat 2500 cpm/cm2membrane.Afterincubationwith â€˜@I-IGF-Ifor18 h @ at 4Â°C,themembrane was washed and exposed to KOdakX-Omat film for @1: 3 to 7 days at â€”80Â°Cwithintensifying screens. Immunological Detection of IGFBP-3. Samples of conditioned medium were concentrated50-foldusingCentricon10 microconcentratorsandsamples were electrophoresedona 10%acrylamidegel undernonreducingconditions, blotted to nitrocellulose, and incubated with an IGFBP-3-specific, R7 antisera (3). The nitrocellulose membranes were blocked with TBS-M (TBS containing DAY OF TREATMENT 2% milk fat) for 20 h, incubated with R7 antiserum diluted 1:750 in TBS-M for Fig. 1. EGF and TGF@1regulation of ECEI6-1 cell proliferation. Cells were plated 16 h, and then washed and incubatedwith 1@'I-proteinA(1 X 10@cpm)for into 35-mm dishes in serum-supplemented medium and allowed to attach overnight. On 1 to 2 h. The blots were thenwashedwithTBS containing0.2%TritonX-100 day -1, the cells were washed and shifted to DM. On day 0, DM or DM supplemented and 1%gelatin andthe blots were exposed on KodakX-Omatfilm for 2 to 4 with 20 ng/ml of EGF, 3 ag/nil of TGF@1,or 20 ng/ml of EGF + 3 ng/ml of TGF@1 (+GF, plus growth factors) was added. Fresh medium and growth factors were added at 24-h intervals. At the indicated time points cells were harvested for cell counting. Cell 5 C. K. (1oo and R. L Eckert, unpublished observations. counts are expressed as the mean Â±SE (bars) of three separate experiments. 3161

EGFANDThE EFFECFSOF IGF-l

The level of mRNA encoding IGFBP-3 was also evaluated. Cells 49.5, were grown in DM or DM supplemented with 20 ng of EGF/ml for 3

- .@ 0 days. Polyadenylated RNA was prepared, fractionated on an RNA I @â€˜ denaturing agarose gel, and transfer blotted to nitrocellulose. The membrane was then hybridized with an IGFBP-3-specific cDNA (33). i*I4@1 ; ,.,, As shown in Fig.5A, a single2.6-kilobasemRNA specieswas 32.5' detected. EGF treatment reduced the levels of IGFBP-3 mRNA by about 60% (P < 0.01). Densitometric analysis of the IGFBP-3 species 27.5' identified by Western and Northern blot revealed a 50 to 60% EGF

@ @â€˜4 suppression of IGFBP-3 at both the protein and mRNA level (Fig. SC, Cs EGFTGFBEGF OM E@TGFBE@ OM E@ TGFBEGFOM Western and Northern blots). These results indicate that EGF regu L@ T@JL TGFBIL@@T@Bj

DAYI DAY2 DAY3 3.0 80 Fig. 2. EGF and TGF@1 regulation of IGFBP levels in the cell culture medium. 0 @oâ€”38/42 kD. l@np Medium, conditioned for 24 h, was harvested from the cultures shown in Fig. 1 on days 70 @ 1, 2, and 3 and assayed for the presence of IGFBPs by ligand blotting (37). A known 0 24 kOs l@@P amount of human cord serum (CS) was coetectrophoresed as an internal control. Ordinate, 2.5 â€˜4 @.â€”can numb.r migration of the molecular weight markers. 60

â€˜4 @ I 2.0 50 Table 1 Quantitation of EGF and TGFI3I effects on ECEJ6-1 cell JGFBP level 2 Ligand blots from three separate experiments performed as described in Fig. 2 were 0 40@ quantitated by densitometric scanning. Values are presented in arbitrary density units and 0 are the mean Â±SE. 1.5 30J 338/42-kDaTreatmentDay 1Day 2Day levelEGF38IGFBP 20 D 0.3Â°[email protected] Â±9.936.9 Â±2.8Â°6.0 Â± I .0 2.7Â°EGF/[email protected] Â±20.778.9 Â±12.5â€•32.1 Â± . 0 10 2.3Â°DM55.5 Â±22.221.1 Â±6.4Â°8.7 Â± @ 3.524-kDa Â±0.9v.113.8 0.850.3 Â± . 0.5 â€˜ . . . 0 levelEGF17.1IGFBP 0 10 20 30 40 1TGFj3125.3 Â±2.625.9 Â±3.47.9 Â± [email protected] Â±10.926.8 Â±2.616.8 Â± EQF (rig/mi) 0.6Â°DM14.2 Â±4.212.3 Â±6.35.2 Â± Â±I20.3 Â±4.922.7 Â±3.6 Fig. 3. Concentration-dependent regulation of ECE16-1 cell proliferation and IGFBP leveis by EGF. ECE16-1 cells were plated in 35-mm dishes as in Fig. 1. After equilibra a Significant difference (P < 0.05) compared to the DM control. tion in DM for 24 h, they were treated with the indicated concentrations of EGF for 72 h with daily medium changes. At 72 h, the cells were harvested and counted. The results are expressed as fold change compared to cells grown in DM (i.e., no EGF). IGFBP levels (38/42 kDa, D 24 kDa, i@)were determined by scanning ligand blots of 24 h TGFf31 decreased IGFBP to levels similar to that observed with EGF conditioned medium derived from the last 24 h of treatment and expressing the level in alone. In addition, treatment with EGF or EGF + TGF@31for 3 days arbitrary density units. resulted in a significant reduction in the level of the 24-kDa IGFBP (Table 1). The EGF reduction in IGFBP levels was concentration-dependent 50 and inversely correlated with increased proliferation (Fig. 3). Half maximal growth stimulation and 38/42-kDa IGFBP reduction was observed at 3 to 5 ng/mI of EGF; maximal proliferation and 38/42 kDa 40 IGFBP reduction was observed at @10ng/ml. 0 Antibody Neutralization Eliminates EGF Effects on ECE16-1 Cell Proliferation and IGFBP Production. Incubationof ECE16-1 K 30 cells with an EGF-specific antibody eliminates the EGF growth stim â€˜4 ulatory effect (Fig. 44) and the EGF-dependent down-regulation of E IGFBP levels (Fig. 4B). These results indicate that the effects are specific for EGF. TGFa, another member of the EGF family, also

inhibits the production of IGFBPs and stimulates ECE16-1 cell pro DM EGF @GF + liferation; moreover, the TGFa response is countered by a TGFa 10 AB specific antibody (not shown). Characterization of the 38/42-kDa IGF-binding Species. The IGF binding activity migrating at 38/42 kDa is the predominant species regulated by EGF. We therefore decided to characterize these 0 DM EGF EGF binding activities. The mobility of this cluster of bands suggested it AB might be IGFBP-3. To test this hypothesis, culture medium condi Fig. 4. EGF-neutralizing antibody abrogates EGF effects. ECE16-1 cells were allowed . tioned by ECE16-1 cells during the last 24 h of a 72-h incubation was to attach overnight in serum-supplemented medium and then incubated in DM for 24 h. concentrated, electrophoresed, transferred to nitrocellulose, and incu Treatment was continued in DM or DM containing 20 ng/ml of EGF or 20 ngJml of EGF + 240 ,@gof EGF-neutralizing antibody and fresh medium was added daily. After bated with R7, an antiserum specific for IGFBP-3. This antiserum a 72 h treatment, the cells were harvested and counted (A). Cell number is shown as the detected the 38/42-kDa species produced by ECE16-1 cells (Fig. 5A). mean Â±SE(bars) of three separate experiments. Medium from the last 24 h of treatment was collected and analyzed for the presence of IGFBPs. B, representative autoradiogram This species was reduced by 55% (P < 0.01) in medium conditioned from one of three experiments.Arrow,broad band (38/42 kDa) correspondingto by cells treated for 3 days with 20 ng/ml of EGF (Fig. 5A). IGFBP-3. 3162

Northern blot

OM E@

A Western blot

DME@

Fig. 5. EGF suppression of IGFBP-3 protein and mRNA levels. In A, ECE16-l cells plated in serum supplemented medium were equilibrated in DM for @-3.@I 24 h and subsequently treated with DM or DM + 20 ng/ml of EGF for 72 h. Conditioned culture medium 49.5> (last 24 h) was collected and concentrated, and the IGFBP-3 concentration in the medium was assayed K@FBP4 by Western blot using an anti-human IGFBP-3 anti serum followed by incubation with â€˜@I-proteinA. Ordinate, position of the molecular weight markers. 32.5 > Arrow, IGFBP-3 band. In B, polyadenylated RNA was prepared from parallel dishes, fractionated by gel 27.5> electrophoresis, transferred to membranes, and hy bridized with a 32P-labeled cDNA clone specific for IGFBP-3 (33). The IGFBP-3 hybridization signal was normalized based on the glyceraldehyde-3-phos phate dehydrogenase (GAPDH) signal. Glyceralde hyde-3-phosphate dehydrogenase is not regulated by GAPDH-@@@ EGF.The hybridizedbands(arrows)correspondto the expected size of the IGFBP-3 (2.6 kilobases) and glyceraldehyde-3-phosphate dehydrogenase (1.0 Id lobase) transcripts. C, estimate of the relative levels of IGFBP-3 protein and IGFBP-3 mRNA levels de rived by scanning a series of blots represented by those shown in A and B. A parallel series of ligand blots (not shown) were also scanned. The results are C 12 18 30 expressed in arbitrary units. EGF treatment signifi N@thsm canfly reduced IGFBP-3 protein and mRNA levels Mot @ (P < 0.01). Bars, SE. 10 15 25 S S S @.12 )@ 20 a a I- a- I. a .0 I- 9 :@is a a

@ >1 6 10 a C C .i V 0 02 0 3

0 0 0 j 1

lates IGFBP-3 levels, at least in part, by regulating the level of The increased responsiveness of ECE16-1 cells to IGF-I growth IGFBP-3 mRNA. A parallel reduction in the level of IGFBP-3 was stimulation coincides with an EGF-dependent decrease in IGFBP-3 observed when measured by densitometric analysis of ligand blots levels (see Fig. 3). As shown in Fig. 7, A and B, addition of 3 ng/ml (Fig. SC). EGF to cultures containing DM or various concentrations of IGF-I EGF Potentiation of IGF-I Mitogenic Effect Is Associated with results in a 50 to 85% decrease in IGFBP-3 levels, a decrease Decreased IGFBP-3 Levels. Our results indicate that EGF regulates that is evident in the presence or absence of IGF-I (Fig. 7B). the level of IGFBP-3 protein, suggesting that EGF may regulate the Interestingly, the suppression is more dramatic at higher IGF-I ability of IGF-I to stimulate cell growth. To test this possibility, we concentrations. measured the effects of cotreatment with a submaximal growth stim EGF Regulation of IGFBP-3 Levels in Other Cervical Cell ulatory concentration (3 ng/ml; see Fig. 3) of EGF and various Lines. To determine whether EGF regulation of IGFBP levels is a concentrations of IGF-I on ECE16-1 cell growth (Fig. 6). The results common feature in human cervical epithelial cells, we treated ECE16- are expressed as the fold increase relative to the respective control Dl, ECE16-D2, CaSki, and SiHa cells in the presence or absence of (either DM for the â€”EGFcurve or 3 ng/ml of EGF for the +EGF 20 ng of EGF/ml for 3 days. The 24-h conditioned medium from the curve). The method of normalization corrects for basal mitogenic last day of treatment was collected and assayed for IGFBPs by ligand. acitivity of EGF, revealing the contribution ofIGF-I to cell growth. At blot. As shown in Fig. 8, EGF treatment decreased IGFBP-3 produc 2 to about 10 ng of IGF-I/ml, the presence of EGF significantly (P < tion in each cell line, but the extent of decrease varied. As determined 0.05) increased cell sensitivity to IGF-I, compared to cells treated with by densitometry, EGF treatment reduced IGFBP-3 levels by 27% in IGF-I alone. These results indicate that IGF-I mitogenic activity is ECE16-D1 and ECE16-D2 cells, by >50% in ECE16-1 cells, by 93% enhanced by EGF. in CaSki cells, and by 60% in SiHa cells. 3163

to alter IGF responsiveness (17). However, the concomitant decrease 1.7 in both IGFBP-3 mRNA and secreted protein in EGF-treated

0 1.6 ECE16-1 cells argues against repartitioning of IGFBP as a major g mechanism for EGF suppression of IGFBP-3 levels. The mode of 0 IS E 0 I .4 A a 1.3 a 49.5' I .2

0 II 1.1 K@4- @lISINN Â£@sN@@Ia

I .0 27.5@ 0 5 10 15 20 25 30 35 40 IGFâ€”i(rig/mi) 4 Fig. 6. IGF-I stimulation of ECE16-l cell proliferation is enhanced by EGF. ECE16-1 0.5 1 2 358102040 0 5 1 2 3 58102040 cells were seeded in serum-containing medium and shifted to DM as in Fig. 1. After a 24-h equilibration in DM, the cultures were treated with the indicated concentration of @-l(nWn@) I@-1 (na/ni) IGF-I Â±3 ng/ml of EGF for 72 h with daily medium changes. The results [mean Â±SE , 3ag/sdE@ (bars) of two experiments] are expressed as â€œfoldchangefrom controlâ€•(i.e., no EGF for the â€”EGFcurve,EGF alone for the +EGF curve). â€˜,significantdifference (P < 0.05) B between corresponding -EGF and +EGF treatment groups. 30 .-.â€” -EGF

â€˜43 â€”0â€” +EQF When normalized per cell for cells maintained in DM, the relative 4.. 25 C level of IGFBP-3 varies in the cell lines as follows: ECE16-D2 > ECE16-D1 > ECE16-1 >> CaSki >> SiHa. The CaSki and SiHa cells @,20 produce very low levels of IGFBP-3. In fact, in these transformed cell a-

lines (CaSh and SiHa), the 24-kDa IGFBP species and not IGFBP-3 -@ 15 was the major IGFBP produced. Moreover, the 24-kDa species in V creased by 25 to 50% in response to EGF treatment in the transformed @10, cells (compared to a slight suppression in ECE16-1, ECE16-D1, and â€˜43 C ECE16-D2). In addition, CaSki cells produce additional bands at .:@@jTIo\J OOo_o@ V 5 28â€”32kDa and SiHa produces a band at 28 kDa. 0

0 DISCUSSION 0 10 20 30 40 EGF, IGFs, and Cell Proliferation. The IGFBPs are known to IGFâ€”l (ng/ml) play an important role in the modulation of IGF action at the cellular Fig. 7. EGF and IGF-I regulation of IGFBP-3 levels. ECE16-1 cells were seeded in level (1). Many cell types display altered production of IGFBPs in serum-containing medium and shifted to DM as in Fig. 1. After a 24-h equilibration in response to hormones and growth factors (7â€”9).A major goal in our DM, the cultures were treated for 72 h with the indicated concentration ofIGF-I Â±3ng/ml of EGF. Conditioned medium from the last 24 h treatment period was harvested and laboratory is to identify and understand mechanisms that may con assayed for IGFBPs by ligand blot (A). Arrowheads, position of the molecular weight tribute to the progression of cervical cancer. In the present study, we markers. Arrow, IGFBP-3 bands (38/42 kDa). The autoradiographs were scanned with a densitometer and the results expressed in arbitrary density units [mean Â±SE(bars) of two have utilized HPV16-immortalized and HPV16-transformed cervical separate experiments] (B). epithelial cell lines (31) as models to study the effects of EGF on the IGF signaling system. Members of the EGF and IGF families are potent stimulators of epithelial cell growth (25, 26, 40â€”42).Although ECE16-D2 ECE16-D1 ECE16-1 CaSki SiHa - + - .@. _â€”9:- _ + _ + epithelial cells do not produce IGF-I (27, 28), IGFs have been shown to associate with epithelial cells in vivo (43) and stimulate epithelial IGF8P-3-@ I cell growth in vitro (40, 42). In vivo, IGFs are synthesized and

released by fibroblasts and stimulate the growth of surrounding epi 323> thelial cells via a paracrine mechanism (27). Binding of IGFBPs to IGFs in the extracellular environment reduces the free IGF concen 273> @ tration (1); thus, an alteration in the extracellular IGFBP composition â€˜I.. . @â€”. â€” â€”. 4-24kDa and/or concentration can profoundly influence IGF effects on cells. Fig. 8. Production of IGFBPs by immortalized versus immortalized/transformed hu Cooperation between EGF and IGF in Regulation of Cell Pro man cervical epitheial cell lines. ECE16-1, ECE16-D1, and ECE16-D2 are HPV16- liferation. The present studies demonstratethat the EGF and IGF positive, immortalized, non-tumorigenic cell lines,5whereas CaSki and SiHa are HPV16- pathways interact to enhance cell proliferation. Treatment of ECE16-1 positive transformed tumorigenic lines (35, 36). All lines were plated in serum supplemented medium and then shifted to DM as outlined in Fig. 1. After a 24-h cells with EGF results in a dramatic reduction in the levels of equilibration in DM, the cultures were treated in the presence or absence of 20 ng of IGFBP-3. This reduction is mediated, at least in part, by a reduction EGF/mI for 72 h with daily medium changes. Medium from the last 24 h of treatment was in the level of mRNA encoding IGFBP-3, indicating that EGF influ harvested and assayed for IGFBP levels by ligand blot. Equal quantities of conditioned medium were loaded within each cell type for â€”EGFand +EGF treatment. However, the ences IGFBP-3 mRNA synthesis and/or turnover. Further studies will loading of different numbers of cell equivalents does not permit normalization between be required to identify the level of IGFBP-3 mRNA regulation (syn cell types. Some differences between EGF treated and nontreated groups are underesti mated due to the overexposure of the ligand blot to enable detection of less abundant thesis or turnover). It should be noted that a redistribution of IGFBPs IGFBPs. Arrowheads, position of the molecular weight markers. Arrow, IGFBP-3 bands between soluble and membrane-associated phases has been reported (38/42 kDa). 3164

@ IGPRP-3 â€¢ K(;@ receptors (50). Increased number of EGF receptors has been observed @t@ci in ECE16-1 cells compared to normal ectocervical epithelial cells I n (52). Thus, the EGF family appears to play an important autocrine â€˜if. growth-deregulatory role that may contribute to malignant progres sion in epithelia. Increased levels of EGF-like proteins in epithelial @ If AlGP@RECI It @ - . tumors might lead to suppression of IGFBP-3 levels and thus a pI,,rna Mrrih,anc potentiation IGF mitogenic action. IGFBP Production in Tumorigenic and Nontumorigenic Cer Intracellular Signal(s) Ititrticvlltilar vical Cell Lines. To look for potential changes in IGFBP production Sigmil(s) during tumor progression, we examined IGFBP levels in a series of â€œ@â€˜OICFBP-3I immortalized or immortalized/transformed HPV16-positive cervical @@1I_._.__ II'roliferiitio,i cell lines. Our comparison of three HPV16-immortalized cervical cell /@ .. .Z lines (ECE16-1, ECE16-D1, and ECE16-D2)5 and two HPV16-posi tive immortalized and transformed cell lines (CaSki and SiHa) (35, @ I mRNA 36), show that although the EGF suppression of IGFBP-3 levels is maintained in SiHa and CaSki cell lines, these lines release much less Fig. 9. Model describing the interaction between the IGF- and EGF-signaling path ways. IGF-I is normally in equilibrium between an extracellular IGFBP-3 bound pool, a IGFBP-3 than the immortalized lines. In fact, the predominant form of small amount of free IGF-l, and a pool bound to the type I IGF membrane-associated IGFBP in CaSki and SiHa cells is the 24-kDa IGFBP. These results receptor (IGF-R) (some IGFBP-3/IGF complex may also be associated with the mem brane, not shown) (17). EUF interacts with and activates the EGF receptor (EGF-R) suggest the following possibilities regarding the role of IGFBP-3 initiating the EGF signaling cascade that eventually results in cell division. Our results during cervical cancer progression: (a) EGF may promote malignant suggest that EGF, presumably acting via its receptor, down-regulates the level of IGFBP-3 progression by facilitating IGF-I effects (via IGFBP-3 down-regula mRNA and, in turn, reduces the amount of IGFBP-3 released from the cell. This, in turn, increases the amount of free extracellular IGF-I which is able to bind to the type I IGF-l tion) and (b) IGFBP-3 levels are lower in more malignant tumor cells, receptorand stimulatecellgrowth.OtherIGFBPsare ignored,althoughtheyare also suggesting that IGF-I may be a more potent mitogen in more highly likely to contribute to the net biological response. The net effect is enhanced growth due transformed cells. to stimulation via both pathways. The cervical cancer model used in these studies presents an inter esting system for the study of dysregulation of IGFBP production regulation of IGFBP-3 mRNA levels by EGF is currently being during malignant progression and suggest that a reduction in IGFBP-3 investigated. production may contribute to the process. It will be important to The decrease in IGFBP-3 is correlated with an increased respon determine whether similar changes are observed during progression in siveness of the cells to stimulation of growth with IGF-I. Based on other types of cancers. these results, we propose a dual mode of EGF action in which it simultaneously stimulates growth via direct effects on the EGF re ACKNOWLEDGMENTS ceptor-dependent signaling system and reduction in IGFBP-3 levels (Fig. 9). IGFBP-3 is thought to bind and sequester IGFs such that they The authors thank Drs. Peter LaCelle and Jean Welter for reviewing the cannot obtain access to the cell surface IGF-I receptor (14â€”16).The manuscript. net effect of the EGF-dependent reduction in IGFBP-3 is to enhance IGFs mitogenic stimulation, presumably by increasing the amount of REFERENCES free IGF-I available to bind the cell surface IGF receptor. In this 1. Baxter, R. C., and Martin, J. L. Binding proteins for the insulin-like growth factors: context it is interesting to note that overexpression of IGFBP-3 has structure, regulation and function. Prog. Growth Factor Res., 1: 49â€”68,1989. been shown to suppress cell growth, suggesting that IGFBP-3 may 2. Cohick, W. S., and Clemmons, D. R. Regulation of insulin-like growth factor binding protein synthesis and secretion in a bovine epithelial cell line. Endocrinology, 129: function in some cell types as a growth suppressor protein (44). 1347â€”1354,1991. Although the present paper is the first report of EGF acting as a 3. Martin, J. L., and Baxter, R. C. Insulin-like growth factor-binding proteins (IGF-BPs) positive regulator of IGF-I action via the IGFBPs, others have re produced by human skin fibroblasts: immunological relationship to other human IGF-BPs. Endocrinology, 123: 1907â€”1915,1988. ported similar cooperation in other growth factor signaling systems 4. DeLeon, D. P., Bakker, B., Wilson, D. M., Hints., R. L, and Rosenfeld, R. G. (45â€”47).However, care must be taken in extending our results with Demonstration of insulin-like growth factor (IGF-l and II) receptors and binding protein in human breast cancer cell lines. Biochem. Biophys. Res. Commun., 152: EGF in human cervical epithelial cells to other cell types, because 398â€”405,1988. IGFBPs may be modulated by EGF in a cell type-specific manner. For 5. Clemmons, D. R., Thrailkill, K. M., Handwerger, S., and Busby, W. H., Jr. Three example, IGFBP-3 levels are enhanced by EGF in porcine granulosa distinct forms of insulin-like growth factor binding proteins are released by decidual cells in culture. Endocrinology, 127: 643â€”650,1990. cells (45). 6. Schmid, C., Zapf, J., and Froesch, E. R. Production ofcarrier proteins for insulin-like EGF, IGFs, and Cervical Tumor Growth. The fact thatthe EGF growth factors (IGF5)by rat osteoblastic cells. Regulation by IGFI and cortisol. FEBS signaling pathway regulates the available levels of IGF (via regulation LeU., 244: 328â€”332, 1989. 7. Thrailkill, K. M., Clemmons, D. R., Busby, W. H., Jr., and Handwerger, S. Differ of IGFBP-3 mRNA levels) suggests this pathway may play an im ential regulation of insulin-like growth factor binding protein secretion from human portant role in the genesis of cervical tumors. In vivo, TGFa and other decidual cells by IGF-l, insulin and relaxin. J. Clin. Invest., 86: 878â€”883, 1990. 8. Conover, C. A., Liu, F., Powell, D., Rosenfeld, R. G., and Hintz, R. L. Insulin-like EUF-like molecules such as amphiregulin play an important role in growth factor binding proteins from cultured human fibroblasts. Characterization and the process of normal cellular proliferation and differentiation hormonal regulation. J. Clin. Invest., 83: 852â€”859,1989. (26, 48). Members of the EGF family have been implicated as agents 9. Martin, J. L., and Baxter, R. C. Production of an insulin-like growth factor (IGF)-inducible IGF-binding protein by human skin fibroblasts. Endocrinology, 127: that cooperate to cause the malignant transformation of epithelial cells 781â€”788,1990. (49). In addition, carcinoma cells derived from squamous epithelia 10. Dc Mellow, J. S. M., and Baxter, R. C. Growth hormone-dependent insulin-like growth factor (IGF) binding protein both inhibits and potentiates IGF-I-stimulated constitutively produce TGFa (50) and TGFa overexpression has been DNA synthesis in human skin fibroblasts. Biochem. Biophys. Res. Commun., 156: associated with formation of hyperplastic epidermis (51). In addition 199â€”204,1988. to overexpression of members of the EGF family, overexpression of I I. Blum, W. F., Jenne, E. W., Reppin, F., Kietzmann, K., Ranke, M. B., and Bierich, J. R. Insulin-like growth factor I (IGF-l)-binding protein complex is a better mitogen the EGF receptor has been observed in a variety of squamous carci than free IGF-I. Endocrinology, 125: 766â€”772,1989. noma cell lines along with apparent inability to down-regulate these 12. Ernst, M., and Rodan, G. A. Increased activity of insulin-like growth factor (IGF) in 3165

osteoblastic cells in the presence of growth hormone (GH): positive correlation with hormone-dependent insulin-like growth factor-binding protein. Mol. Endocrinol., the presence of the GH-induced IGF-binding protein BP-3. Endocrinology, 127: 2: 1176â€”1185,1988. 807â€”814, 1990. 34. Tso, J. Y., Sun, X. H., Kao, T. H., Reece, K. S., and Wu, R. Isolation and 13. Conover, C. A., and Powell, D. R. Insulin-like growth factor (IGF)-binding protein-3 characterization of rat and human glyceraldehyde-3-phosphate dehydrogenase blocks IGF-I-induced receptor down- regulation and cell desensitization in cultured cDNAs: genomic complexity and molecular evolution of the gene. Nucleic Acids bovine fibroblasts. Endocrinology, 129: 710â€”716,1991. Rca., 13: 2485â€”2502,1985. 14. Gopinath, R., Walton, P. E., and Etherton, 1. D. An acid-stable insulin-like growth 35. Pattillo, R. A., Hussa, R. D., Story, M. J., Ruckert, A. C. F., Shalaby, M. R., factor (IGF)-binding protein from pig serum inhibits binding of IGF-l and IGF-II to and Mattingly, R. F. Tumor antigen and human chorionic gonadotropin in CaSki vascular endothelial cells. J. Endocrinol., 120: 231â€”236,1989. cells: a new epidermoid cervical cancer cell line. Science (Washington DC), 196: 15. Bicsak, T. A., Shimonaka, M., Malkowski, M., and Ling, N. Insulin-like growth 1456â€”1458,1977. factor-binding protein(IGF-BP) inhibition ofgranulosa cell function: Effect on cyclic 36. Friedl, F., Kimura, I., and Osato, T. Studies on a new human cell line (SiHa) drived adenosine 3',S'-monophosphate, deoxyribonucleic acid synthesis, and comparison from carcinoma of the uterus. I. Its establishment and morphology. Proc. Soc. Exp. with the effect of an IGF-I antibody. Endocrinology, 126: 2184â€”2189, 1990. Biol. Med., 135: 543â€”545,1979. 16. Conover, C. A., Ronk, M., Lombana, F., and Powell, D. R. Structural and biological 37. Hossenlopp, P., Scum, D., Segovia-Quinson, B., Hardouin, S., and Binoux, M. characterizationofbovineinsulin-likegrowthfactorbindingprotein-3.Endocrinol Analysis of serum insulin-like growth factor binding proteins using western blotting: ogy, 127: 2795â€”2803,1990. use ofthe method for titration ofthe binding proteins and competitive binding studies. 17. Conover, C. A. Potentiation ofinsulin-like growth factor(IGF) action by IGF-binding Anal. Biochem., 154: 130â€”143,1986. protein-3: studies of underlying mechanism. Endocrinology, 130: 3191â€”3199,1992. 38. Chomczynski, P., and Sacchi, N. Single-step method of RNA isolation by acid 18. Broker, T. R. Structural and genetic expression of papillomaviruses. Obstet. Gynecol. Gin. NorthAm.,14: 329â€”348,1987. guanidinium thiocyanate-phenol-cMoroform extraction. Anal. Biochem., 162: 156â€”159,1987. 19. Durst, M., Giasmann, L, lkenberg, H., and zur Hausen, H. A papillomavirus DNA from a cervical carcinoma and its prevalence in cancer biopsy samples from different 39. Boedtker, H. Conformation independent molecular weight determination of RNA by geographic regions. Proc. Nail. Acad. Sci. USA, 80: 3812â€”3815,1983. gel electrophoresis. Biochim. Biophys. Acts, 240: 448â€”453,1971. 20. zur Hausen, H. The role of papillomaviruses in human anogenital cancer, In: N. P. 40. Krane, J. F., Murphy, D. P., Carter, C. M., and Krueger, J. 0. Synergistic effects of Salaman and P. M. Howley (eds.), The Papovaviridae, Vol. 2, pp. 245-263. New epidermal growth factor (EGF) and insulin-like growth factor 1/somatomedin C York: Plenum Publishing Corp., 1987. (IGF-I) on keratinocyte proliferation may be mediated by IGF-I transmodulation of 21. Lang, R. A., and Burgess, A. W. Autocrine growth factors and tumongenic transfor the EGF receptor. J. Invest. Dermatol., 96: 419â€”424, 1991. mation. Immunol. Today, 11: 244-249, 1990. 41. Barrandon, Y., and Green, H. Cell migration is essential for sustained growth of 22. Goustin, A. S., Leof, E. B., Shipley, G. D., and Moses, H. L Growth factors and keratinocyte colonies: the roles of transforming growth factor-a and epidermal cancer. Cancer Res., 46: 1015-1029, 1986. growth factor. Cell, 50: 1131â€”1137,1987. 23. Heldin, C-H., and Westermark, B. Growth factors as transforming proteins. Eur. J. 42. Ristow, H-J., and Messmer, T. 0. Basic fibroblast growth factor and insulin-like Biochem., 184: 487â€”496,1989. growth factor I are strong mitogens for cultured mouse keratinocytes. J. Cell. Physiol., 24. Yee, D., Paik, S., Lebovic, 0. S., Marcus, R. R., Favoni, R. E., Cullen, K. J., 137: 277â€”284,1988. Lippman, M. E., and Rosen, N. Analysis of insulin-like growth factor I gene 43. Han, V. K. M., Hill, D. J., Strain, A. J., Towle, A. C., Lauder, J. M., Underwood, expression in malignancy: evidence for a paracrine role in human breast cancer. Mol. L E., and D'Ercole, A. J. Identification of somatomedin/insulin-like growth factor Endocrinol., 3: 509â€”517,1989. immunoreactive cells in the human fetus. Pediatr. Rca., 22: 245-249, 1987. 25. Coffey, R. J., Jr., Derynck, R., Wilcox, J. N., Bringman, T. S., Goustin, A. S., Moses, 44. Cohen, P., Lamson, 0., Okajima, T., and Rosenfeld, R. 0. Transfection of the human H. L, and Pittelkow, M. R. Production and auto-induction of transforming growth insulin-like growth factor binding protein-3 gene into Balb/c fibroblasts inhibits factor-a in human keratinocytes. Nature (Lond.), 328: 817â€”820,1987. cellular growth. Mol. Endocrinol, 7: 380-386, 1993. 26. Cook, P. W., Mattox, P. A., Keeble, W. W., Pittelkow, M. R., Plowman, G. D., 45. Mondschein, J. S., Smith, S. A., and Hammond, J. M. Production of insulin-like Shoyab, M., Adelman, J. P., and Shipley, G. D. A heparmnsulfate-regulated human growth factor binding proteins (IGFBP5) by porcine granulosa cells: identification of keratinocyte growth factor is similar or identical to amphiregulin. Mol. Cell. Biol., 11: IGFBP-2 and -3 and regulation by hormones and growth factors. Endocrinology, 127: 2547â€”2557,1991. 2298â€”2306,1990. 27. Barreca, A., Dc Luca, M., Del Monte, P., Bondanza, S., Damonte, G., Cariola, 0., 46. Smith, E. P., Dickson, B. A., and Chernauseck, S. D. Insulin-like growth Di Marco, E., Giordano, 0., CanCedda, R., and Minuto, F. In vitro paracrine factor binding protein-3 secretion from cultured rat sertoli cells: dual regulation by regulation of human keratinocyte growth by fibroblast-derived insulin-like growth folliclestimulatinghormoneandinsulin-likegrowthfactor1.Endocrinology,127: factors. J. Cell. Physiol., 151: 262â€”268,1992. 2744â€”2751,1990. 28. Park, J. H., McCusker, R. H., Vanderhoof, J. A., Mohammadpour, H., Harty, R. F., 47. Yateman, M. E., Claffey, D. C., Cwyfan-Hughes, S. C., Frost, V. S., Wass, J. k, and andMacDonald,R.0. Secretionofinsulin-likegrowthfactorII (IGF-Il)andIGF Holly, J. M. Cytokines modulate the sensitivity of human fibroblasts to stimulation binding protein-2 by intestinal epithelial (IEC-6) cells: implications for autocrine with insulin-like growth factor-I (IGF-I) by altering endogenous IGF binding protein growth regulation. Endocrinology, 131: 1359â€”1368,1992. production. J. Endocrinol., 137: 151â€”154,1993. 29. Fontana, J. A., Burrows-Mezu, A., Clemmons, D. R., and LeRoith, D. Retinoid 48. Li, S., Plowman, 0. D., Buckley, S. D., and Shipley, 0. D. Heparin inhibition of modulation of insulin-like growth factor-binding proteins and inhibition of breast autonomous growth implicates amphiregulin as an autocnne growth factor for normal carcinoma proliferation. Endocrinology, 128: 1115â€”1122,1991. human mammary epithelial cells. J. Cell. Physiol., 153: 103-111, 1992. 30. ae@@ons, D. R., Camacho-Hubner, C., Coronado, E., and Osborne, C. K. Insulin 49. Cross, M., and Dexter, T. M. Growth factors in development, transformation, and like growth factor binding protein secretion by breast carcinoma cell lines: correlation tumorigenesis. Cell, 64: 271â€”280,1991. with estrogen receptor status. Endocrinology, 127: 2679â€”2686,1990. 50. Reiss, M., Stash, E. B., Vellucci, V. F., and Thou, Z. L. Activation of the autocrine 31. Agarwal, C., Rorke, E. A., Irwin, J. C., and Eckert, R. L Immortalization by human transforming growth factor a pathway in human squamous carcinoma cells. Cancer papillomavirus type 16 alters retinoid regulation of human ectocervical epithelial cell Rca., 51: 6254â€”6262,1991. differentiation. Cancer Rca., 51: 3982â€”3989,1991. 51. Final, E., Kilkenny, A., Strickland, J. E., Balaschak, M., Bringman, T., Derynck, R., 32. Choo, C. K@,Rorke,E. A., and Eckert, R. L. Calcium regulates the differentiation of Aaronson, S., and Yuspa, S. H. TGF a stimulates growth of skin papillomas by human papillomavirus type 16 (HPV16) immortalized ectocervical epitheial cells, autocrine and paracrine mechanisms but does not cause neoplastic progression. Mol. but not the expression of the papillomavirus E6 and E7 oncogenes. Exp. Cell Rca., Carcinog.,1:7â€”12,1988. 208: 161â€”169,1993. 52. Sizemore, N., and Rorke, E. A. Human papillomavirus 16 immortalization of normal 33. Wood, W. I., Cachianes, 0., Henzel, W. J., Winslow, 0. A., Spencer, S. A., Heilmiss, human ectocervical epithelial cells alters retinoic acid regulation of cell growth and R., Martin, J. L, and Baxter, R. C. aoobig and expression of the growth EGF receptor expression. Cancer Rca., 53: 4511â€”4517,1993.

3166

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1994 American Association for Cancer Research. Epidermal Growth Factor Suppresses Insulin-like Growth Factor Binding Protein 3 Levels in Human Papillomavirus Type 16-immortalized Cervical Epithelial Cells and Thereby Potentiates the Effects of Insulin-like Growth Factor 1

Joan R. Hembree, Chapla Agarwal and Richard L. Eckert

Cancer Res 1994;54:3160-3166.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/54/12/3160

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/54/12/3160. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.