sign in sign up
<<
Home , Epidermal growth factor

ICANCER RESEARCH 52, 4242-4247, August I. 1992] Effect of on the Growth Properties and on Epidermal Expression in the Teratoma-derived Tumorigenic Cell Line 1246-3A1

Ginette Serrerò,2Nancy M. Lepak, Jun Hayashi, and Dominic P. Eisinger H'. Alton Jones Cell Science Center, Inc., Lake Placid, New York 12946

ABSTRACT tocrine fashion (5, 6). In addition to producing -related factor, 1246-3A cells could synthesize and secrete in their cul 1246-3A is an insulin-independent tumorigenic cell line isolated from ture medium polypeptide growth factors belonging to the the C3H mouse teratoma-derived adipogenic cell line 1246. In the TGF-04 and TGF-a families which were biologically active, present paper, we have demonstrated that testosterone inhibits the in vivo tumorigenic properties of the 1246-3A cells. Castrated male mice since they stimulated anchorage-independent growth of normal receiving injections of 1246-3A cells developed larger tumors at a higher rat cells (7). Moreover, these factors could inhibit adi frequency than sham-operated animals. Administration of testosterone pose differentiation (8, 9) via binding to TGF-/3 and EGF re to castrated male mice resulted in a dramatic decrease in tumor devel ceptors (IO).5 In order to examine the tumorigenic properties of opment. In vitro studies indicated that testosterone inhibited by 50% the the 1246-3A cells, we routinely used 6-week-old female C3H proliferation of the 1246-3A cells in culture. However, growth inhibition mice. However, when we compared the tumor growth of cells was observed only if the cells had been cultivated in the presence of injected in age-matched female and male C3H mice, it was testosterone for at least 4 days. In contrast, testosterone had little effect found that tumor growth was slower when cells were injected on the proliferation of the parent cell line 1246. Binding of several into male mice as compared to female mice, thus suggesting polypeptide growth factors was examined in cells cultivated in the ab sence and in the presence of testosterone. Testosterone increased I25I- that testosterone may be acting as a negative regulator of the 1246-3A tumor growth in vivo. Sex steroids have been shown to EGF specific binding to 1246-3A cells. Scatchard analysis of EGF binding indicated that testosterone treatment induced a 2.4-fold increase influence the growth of tumors originating from tissues which in the number of cell surface EGF binding sites. This was accompanied are responsive in vivo such as prostate for androgens by an increase in the intensity of cross-linked EGF receptors on the cells and mammary epithelial cells for . In both cases, slow- treated with testosterone. In addition, 1246-3A cells cultivated for 9 growing prostatic tumors and mammary epithelial cell carcino days in the presence of testosterone displayed a 10-fold increase in the mas require the presence of the corresponding steroid hormone level of EGF receptor niR.VA when compared to 1246-3A cells main for their sustained growth. The effect of both androgens and tained in its absence. Similar to its action on cell proliferation, the estrogens on the growth of normal and malignant prostate and increase in EGF receptor number and inRN A expression was observed tissues has been studied (11, 12). However, in mainly if 1246-3A cells had been exposed to testosterone for 9 days. The both cases it has not yet been possible to implicate androgen or data presented in this paper demonstrate that both in vivo and in vitro, testosterone induces in the teratoma-derived 1246-3A cell line pheno- directly in the regulation of tissue growth. This has led typic changes such as growth inhibition and modulation of EGF receptor to the currently formulated hypothesis that polypeptide growth expression. factors and/or their receptors may be the mediator of androgen or estrogen action on in vivo and possibly in vitro. Since we have found out that male mice receiving injections of INTRODUCTION 1246-3A cells develop tumors at a slower rate than female mice, The 1246 cell line is a C3H mouse teratoma-derived adipo we have investigated here whether testosterone inhibits the in genic cell line which can proliferate and differentiate in defined vivo growth of the tumorigenic teratoma-derived 1246-3A cell medium and which stringently requires insulin for both pro line. Experiments were also carried out to determine the effect cesses ( 1). From 1246 cells maintained in the absence of insulin, of testosterone on cell proliferation in vitro and on the binding several insulin-independent cell lines were isolated. One of of growth factors, particularly EGF, on the 1246-3A cells. them, called 1246-3A, was particularly studied. It was shown that 1246-3A cells could proliferate in the absence of insulin, MATERIALS AND METHODS have lost the ability to differentiate, and have become tumori genic when injected into syngeneic hosts C3H mice (2). Materials 1246-3A cells injected s.c. at a density of IO6 cells/animal gave Calf serum and fetal bovine serum were purchased from HyClone rise to tumors within 30 days. These tumors have been analyzed (Logan, UT). Nutrient media DME and Ham's F12 were obtained from as being similar to leiomyosarcomas.3 1246-3A-conditioned Gibco (Grand Island, NY). fibronectin, human recombinant medium contained growth-promoting activity for the parent basic FGF, and mouse submaxillary receptor grade EGF were obtained cell line 1246. Biochemical characterization of the growth fac from Upstate Biotechnology, Inc. (Lake Placid, NY). Porcine receptor tors produced by the 1246-3A cells was undertaken using 1246- grade 125I-insulin (374 nCi/ng) and [«-"PjrUTP (1000 Ci/mmol) were 3A-conditioned medium as starting material. It was found that from New England Nuclear (Wilmington, DE). NaCl and chloroform were from Baker (Philippsburgh, NJ). Disuccinimidyl suberate was the cells produced a factor similar to insulin which was purified to homogeneity (3, 4). This factor could bind to 1246-3A insu from Pierce Chemicals (Rockford, IL). Acyl restriction enzyme was from Promega (Madison, WI). RNase A and RNase TI were from lin receptors and stimulate the producer cell growth in an au- Boehringer Mannheim (Indianapolis, IN). Phenol and formamide were from Fisher Scientific (Pittsburgh, PA). Chemicals and molecular Received 3/3/92; accepted 5/20/92. weight markers for polyacrylamide gel electrophoresis were from Bio- The costs of publication of this article were defrayed in part by the payment of Rad (Richmond, CA). All other chemicals were from Sigma (St. Louis, page charges. This article must therefore be hereby marked advertisement in accord ance with 18 U.S.C. Section 1734 solely to indicate this fact. MO). 1Supported by Grant CA37589 from the NIH and Grant 2003 from the Council for Tobacco Research. 4 The abbreviations used are: TGF, transforming growth factor; EGF, epidermal 2To whom requests for reprints should be addressed, at W. Alton Jones Cell growth factor; FGF, ; DME, Dulbecco's modified Eagle's Science Center, Inc., 10 Old Barn Road. Lake Placid, NY 12946. medium: F12, Ham's F12 medium; ChXCS, charcoal-extracted calf serum. 3 G. Serreròand J. Hayashi. unpublished results. 5 G. Serrerò,unpublished results. 4242 Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1992 American Association for Cancer Research. TESTOSTERONE. TUMOR GROWTH, AND EGF RECEPTOR EXPRESSION

Methods Mgof total RNA and 5 x IO5 cpm of probe were hybridized overnight at 50°Cin50% formamide and lx 1,4-piperazinediethanesulfonic acid In Vivo Tumorigenesis. Six- to seven-week-old C3H mice (Trudeau in a total volume of 40 n\. The next day samples were digested for 30 Institute, Saranac Lake, NY) were used in our experiments. Male mice min at room temperature with 300 M!of an RNase buffer were castrated and received s.c. injections in the flank of IO6 1246-3A containing 10 m\i Tris-HCl, pH 7.5, 1 mM EDTA, 0.3 M NaCl, 12 Mg cells/animal, 1 day after castration. In some experiments, castrated RNase A, and 150 units RNase T|. The samples were treated with males received a pellet containing 5 mg/pellet of testosterone (60-day proteinase K, phenol/chloroform extracted, precipitated with twenty Mg release) (Innovative Research, Toledo, OH) or a placebo pellet which of tRNA, resuspended in 4 M!loading buffer, heated to 92"C for 5 min, was placed under the skin in the back of the animals. Three to five and electrophoresed on a 6% denaturing polyacrylamide gel. Dried gels animals were used per experimental condition. Experiments were re were analyzed by autoradiography. Protected EGF receptor transcripts peated five times. Animals were examined daily for the appearance of are 160 nucleotides in length. tumors. After 30 days animals were sacrificed, and tumors were excised EGF Radioreceptor Assay for Measuring EGF Receptor Competing and weighed. We routinely checked that testicular tissue was absent and Material. The presence of EGF/TGF-a-like polypeptide in 1246-3A that testosterone and placebo pellets were still in place throughout the culture medium was determined by its ability to compete with 125I-EGF experiments. for binding to CCL64 cells (16). 1246-3A cells were cultivated in 2% Proliferation of 1246 and 1246-3A Cells in Vitro. 1246-3A stock ChXCS in the presence or the absence of testosterone (10~8 M) for 8 cells were cultivated in DME-F12 medium supplemented with fibronec- days. Medium was then changed to factor-free, serum-free DME me tin (1 Mg/ml), transferrin (10 Mg/ml), and FGF (1 ng/ml). For experi dium in the presence or absence of testosterone and collected every 3 ments comparing the growth of 1246 and 1246-3A cells, 12,000 cells/ days. Medium was concentrated 10- to 20-fold by ultrafiltration well in a 12-well dish were inoculated in DME-F12 medium containing through a YM 2 membrane (Amicon, Danvers, MA) and used directly 2% ChXCS in the absence or in the presence of increasing concentra tions of testosterone (10~'°M to 10~8 M). Cells were counted with a in the EGF radioreceptor assay using CCL64 cells as described by Bringman et al. (16). Coulter counter following trypsinization. Insulin, FGF, and EGF Binding to 1246-3A Cells in the Presence and Absence of Testosterone. Cells were cultivated in 35-mm dishes in RESULTS DME-F12 medium supplemented with 2% ChXCS in the absence or the presence of 10~8 M testosterone for 8 days. Cells were then washed Tumorigenicity of 3A Cells Injected into Syngeneic Host C3H Mice. We have shown previously that 1246-3A cells form free of serum and hormone and incubated in 1 ml of DM E containing tumors when injected s.c. at a cell density of IO6 cells into adult 1 mg/ml bovine serum albumin for performing binding. Conditions for I25l-insulin binding have been described previously (5). I25I-FGF was female C3H mice (2) and that tumors developed in male mice kindly supplied by Dr. Mikio Kan (W. Alton Jones Cell Science Cen with a slower rate and a decreased frequency than in age- ter). I25I-FGF binding was performed according to the method of Kan matched female mice. Here, we compared the rate of tumor et al. (13). lodination of EGF and I25I-EGF binding conditions were growth in male mice which had been castrated or sham oper described elsewhere (8). ated. Castrated mice received a pellet containing either testos Affinity Labeling of EGF Receptors. For affinity labeling studies, monolayers of 1246-3A cells which had been cultivated in the presence terone or a placebo compound which was implanted s.c. 1 day or the absence of testosterone (10~8 M)were washed twice with DME prior to injecting the 1246-3A cells. The data of Table 1 show medium containing 1 mg/ml bovine serum albumin. Binding was per that all castrated mice developed tumors with an average weight formed for 2 h at 25°Cwith 300,000 cpm/dish of 125I-EGF (375 tiCil of 1.68-2.16 g/tumor after receiving injections of IO6 1246-3A Mg)alone or in the presence of 1 Mg/ml unlabeled EGF. The solution cells. If the castrated mice received testosterone, only 10 of 14 was removed by aspiration, and the cells were washed three times with mice (60%) developed tumors with a lower weight (Table 1) and ice-cold phosphate-buffered saline. Cells were then treated with disuc- size (Fig. 1). In sham-operated animals, 10 of 16 animals de cinimidyl suberate at a final concentration of 1 mM for 30 min at 4°C. veloped tumors with an average weight of 0.34 g/tumor. These The reaction was quenched by the addition of 2 volumes of cold 10 mivi experiments indicate that testosterone inhibits the in vivo Tris-HCl (pH 7.4), 1 mM EDTA, 0.1 HIMphenylmethylsulfonyl fluo growth of 1246-3A cells. ride, 1 Mg/ml pepstatin, and 1 Mg/ml leupeptin. The washed cells were Effect of Testosterone on the Proliferation of 1246-3A Cells scraped into the buffer containing protease inhibitors and centrifuged at 4°Cat 100 x g for 5 min. The cell pellet was extracted twice with 2% in Culture in Vitro. We then investigated whether testosterone Triton X-100 for 15 min at 4°C.Samples were centrifuged at 4°Cat added to the culture medium of the 1246-3A cells would affect 12,000 x g for 10 min. Supernatants were mixed with an equal volume their proliferation in vitro. As shown in Fig. 2A, testosterone of 2x sample buffer (ix sample buffer consists of 62.5 HIMTris-Cl pH inhibited the proliferation of the 1246-3A cells in a dose-de pendent fashion. At a testosterone concentration of 10~8 M, 6.8, supplemented with 10% glycerol, 2% sodium dodecyl sulfate, 5% 2-mercaptoethanol, and 0.0025% bromophenol blue) (14). Samples proliferation was inhibited by 50%. A higher concentration of were heated at 100'C for 5 min and electrophoresed in a 7.5% slab testosterone did not have a higher inhibitory effect. In contrast, polyacrylamide gel. Gel was stained with 0.05% Coomassie blue in 10% acetic acid and 25% methanol, destained, dried, and subjected to auto- radiography using Kodak XAR film and intensifying screen. Table I Effect of castration and testosterone administration on the tumor EGF Receptor mRNA Expression in 1246-3A Cells. Expression of development of mice receiving injections of 1246-3A cells EGF receptor transcript in 1246-3A cells was measured by RNase C3H male mice aged 6-8 weeks were divided into 4 groups. Mice in group I were sham operated. Mice in groups II, III. and IV were castrated. Mice in groups protection assay. The plasmid pMEGFr, which is a 444- poly- III and IV were implanted with a placebo pellet or a pellet containing S mg merase chain reaction fragment of the intracellular domain of the testosterone (60-day release). One day later mice received s.c. injections of 10* mouse EGF receptor subcloned into SK plasmid,'' was linearized by 1246-3A cells. Mice were observed daily to monitor the appearance of tumors. After 30 days mice were sacrificed, and tumors were excised and weighed. Results restriction enzyme Âcyl,digested with proteinase K, phenol/chloroform indicate the number of mice having developed tumors when compared to the total extracted, ethanol precipitated, and resuspended in water at a concen number of mice that received injections. Average weight of tumors is provided tration of 0.5 Mg/Ml.An antisense [a-32P]rUTP-labeled RNA probe 230 with SE. Experiments were repeated four times. nucleotides in length was generated using T3 RNA polymerase from a of mice weight RNA transcription (Stratagene). Total RNA was extracted from MiceI developingtumors10/16 of tumors(g)0.34 1246-3A cells by the method of Chomczynski and Sacchi (15). Twenty Sham ±0.08 II Castrated 6/6 2.16 ±0.36 III Castrated placebo 10/10 1.68 ±0.32 6 D. P. Eisinger and G. Serrerò,manuscript in preparation. IV Castrated testosteroneNo. 10/14Average 0.28 ±0.06 4243 Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1992 American Association for Cancer Research. TESTOSTERONE. TUMOR GROWTH, AND EOF RECEPTOR EXPRESSION long-term exposure of 1246-3A cells to testosterone in condi CM tions that inhibited their proliferation resulted in changes in specific cell surface binding for the polypeptide growth factors !! III! known to stimulate 1246-3A cell growth (i.e., insulin, FGF, and EGF). As shown in Table 2, specific binding of radiolabeled insulin and FGF remained the same whether or not the cells were treated with testosterone. In contrast, EGF binding in creased by 3-fold when the 1246-3 A cells were treated with 10~8 Mtestosterone when compared to the EGF binding on 1246-3A cells cultivated in the absence of the hormone. However, if the cells were treated with testosterone only for 4 h prior to per forming EGF binding, EGF binding remained the same. These results would suggest that the increase in cell surface binding

10'

Castrated + Testosterone Castrated Fig. 1. Comparison of the size of tumors developed after 30 days by injecting IO6 1246-3A cells s.c. into castrated C3H male mice which had received a tes tosterone pellet (left) or a placebo pellet (right).

150•I Cells-:i|1jjaBI;;i;1246I:

(0 10° 5 *ECELL

NUMB IO«

8A

o io-10 io-9 io-* io-7 io-8 o io-10 i

TESTOSTERONE (M) IO' Fig. 2. Effect of increasing concentrations of testosterone on the proliferation of 1246-3A cells (A) and 1246 cells (B). Cells were inoculated at a density of 6 10 14 10.000 cells/well in a 12-well dish containing 2 ml of DME-F12 supplemented DAYS IN CULTURE with 2% ChXCS. One day after inoculation testosterone was added at concentra tions ranging from 10~'°Mto 10~7 M.Cells were counted at day 9 using a Coulter Fig. 3. Growth of 1246-3A cells in the absence O and in the presence •¿of testosterone. 1246-3A cells (20,000 cells/35-mm dish) were plated in DME-F12 counter. Values are expressed as a percentage of control and are given with SD. medium containing 2% ChXCS in the absence or in the presence of 10~8 M One hundred % control corresponds to the number of cells cultivated in the absence of testosterone. Values were 940.000 ±21.300 cells/well for 1246-3A testosterone added I day after inoculation. Duplicate sets of dishes were counted with a Coulter counter. Values are given with SD. cells and 300,000 ±38.000 cells/well for 1246 cells. testosterone did not inhibit the proliferation of the nontumor- Table 2 Comparison of'"l-insulin, '"I-FGF, and '"l-EFG binding to 1246- 3A cells cultivated in the absence or the presence of testosterone igenic adipogenic 1246 cell line (Fig. 2B). The growth curve of 1246 and 1246-3A cells were cultivated in DME-F12 medium supplemented 1246-3A cells in the absence and the presence of 10~8 M tes with 2% ChXCS in the absence or the presence of IO"8 Mtestosterone. At day 8, duplicate dishes were used to measure I25l-insulin, 125I-EGF, or 12SI-FGF bind tosterone showed that the inhibitory effect of testosterone on ing as described in "Materials and Methods." Nonspecific binding was measured cell growth was not immediate and became apparent after the by performing the binding in the presence of excess cold insulin, EGF, or FGF, cells had been in culture for several days (Fig. 3). respectively. Duplicate dishes were used to determine the cell number. Experi Effect of Testosterone on Polypeptide Growth Factor Binding ments were repeated three times. on 1246-3A Cells. We have shown previously that 1246-3A cell cells)Cells1246-3A Specific binding (cpm/106 proliferation is stimulated by several growth factors (7) and that 1246-3A cells present specific cell surface binding sites for sev testosterone1400 testosterone1800 eral polypeptide growth factors, some of which are being pro duced by the cells (3-7, 10). One possible explanation for the Insulin ±150 ±226 delayed effect of testosterone on 1246-3A cell proliferation FGF 6015 ±1650 6403 ±844 EGF1246 3110±7984700 9490 ±25004645 (Fig. 3) is that testosterone acts indirectly via modulation of the production or response to one or several of these growth fac EGFWithout ±2300With ±1855 tors. To investigate this possibility, we determined whether 4244 Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1992 American Association for Cancer Research. TESTOSTERONE, TUMOR GROWTH. AND EOF RECEPTOR EXPRESSION

collected and analyzed for the presence of EGF receptor com 0.2 peting activity as described in "Materials and Methods." Media collected from three separate experiments were assayed. The results indicate that the amount of EGF receptor competing activity when normalized to the cell number was about the same whether the cells had been cultivated in the presence or in the absence of testosterone [0.22 ±0.01 (SE) ng/106 cells in the 0.1 m absence of testosterone and 0.37 ±0.05 ng/106 cells in its presence]. These results would suggest that the increase in EGF binding sites cannot be explained by a decrease in the amount of EGF receptor competing activity produced by the treatment of the 1246-3A cells with testosterone. Scatchard Analysis of EGF Binding to 1246-3 A Cells Treated or Not Treated with Testosterone. Scatchard analysis of ' 25I- EGF binding to 1246-3A cells cultivated in either the presence 0.4 B or the absence of 10~8 Mtestosterone for 8 days was performed. Typically, Scatchard representation of EGF binding was curvi linear for 1246-3 A cells cultivated in the absence of testosterone and linear for 1246-3A cells cultivated in its presence (Fig. 4). In both cases, linear portions of the Scatchard plots were used to calculate binding parameters (Table 3). The results indicate 0.3 a 2.4-fold increase in the number of EGF receptors in the tes tosterone-treated cells when compared to the untreated 1246-3A cells without a change in the value of the dissociation constant. Characterization of EGF Receptor Species in 1246-3A Cells Exposed or Not Exposed to Testosterone. Cross-linking exper 0.2 iments of I25l-labeled EGF to 1246-3A cells cultivated with or without 10~8 M testosterone were performed after 9 days in culture. An autoradiograph of the results (Fig. 5) showed in both cell lines a cross-linked doublet band (Fig. 5, Lanes 1 and 3) which was competed by adding excess cold EGF (Fig. 5, Lanes 2 and 4). This band had the expected molecular weight of 0.1 Table 3 EGF binding parameters for 1246-3A cells cultivated in the absence or the presence of testosterone Values are the mean of five separate experiments using duplicate dishes/ experiment and are given with SE. Cells/dish Sites/cell With testosterone 1.9 ±0.3 X106 5.0± I 1820 ±230 0.97 ±0.19x10« 3.4±0.8X10-"M 8 12 16 Without testosterone 4400 ±956 EGF BOUND pM Fig. 4. Scatchard analysis of '"I-EGF binding to 1246-3A cells cultivated in the absence (A) or in the presence (A) of 10~8 Mtestosterone. 1251-EGF binding conditions are described in "Materials and Methods." kOa may not have been due to a direct effect of testosterone on EGF _200 binding (data not shown). Moreover, long-term exposure of 1246 cells to testosterone did not have any effect on cell surface _ 116 EGF binding (Table 2). Testosterone Does Not Change the Production of EGF Re ceptor Competing Activity by the 1246-3A Cells. We have pre _ 66.2 viously shown that medium conditioned by the 1246-3A cells contains a TGF-a/EGF-like activity which can bind and occupy EGF receptors on the producer cells (7).7 Based on this infor mation, it is possible to assume that the increase in the number of cell surface binding sites observed when the 1246-3 A cells are treated with testosterone is due to the fact that testosterone inhibits the production of EGF receptor competing activity by the 1246-3A cells, resulting in a reduced occupancy of EGF cell Fig. 5. Affinity labeling of EGF binding sites in 1246-3A cells cultivated for 9 surface binding cells. Medium conditioned by 1246-3A cells days in the presence of IO"8 M testosterone (Lanes 1 and 2) or in its absence cultivated in the absence and the presence of testosterone was (Lanes 3 and 4). I25I-EGF binding and cross-linking conditions are described in "Materials and Methods." Lanes 1 and 3, binding is performed in the absence of

" G. Serreròand D. Mills, unpublished results. excess cold EGF; Lanes 2 and 4. binding is performed in the presence of excess cold EGF. 4245 Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1992 American Association for Cancer Research. TESTOSTERONE, TUMOR GROWTH, AND EOF RECEPTOR EXPRESSION

The data presented in this paper demonstrate that the in vivo Nt Nt tumorigenic property of the 1246-3A cells is inhibited by tes 230. probe 230- <—probe tosterone. Castrated male C3H mice developed larger tumors with a higher frequency than sham-operated animals. When castrated animals received testosterone, the frequency, rate of appearance, and size of the tumors decreased. Testosterone 160_ <-EGFr supplementation also inhibited tumor growth in female C3H mice that received injections of 1246-3A cells (data not shown). 160- Examination of the effect of testosterone on the proliferation of the 1246-3A cells in vitro indicates that only long-term expo +T -T C P sure of the cells to testosterone leads to growth inhibition. In contrast, testosterone did not inhibit the proliferation of the nontumorigenic parent cell line 1246 even after 16 days in culture. and added in the same range of +T -T C P Fig. 6. A, effect of testosterone on EGF receptor mRNA expression. 1246-3A concentrations had no effect on 1246-3A and 1246 cell prolif cells were cultivated for 9 days in the presence (Lane +7") or in the absence (Lane eration (data not shown). As shown in Fig. 3, the effect of -T) of 10~8 M testosterone. Cells were collected and total RNA extracted by the testosterone on cell growth in vitro was not immediate but method of Chomczinsky and Sacchi (15). EGF receptor mRNA expression was measured by RNase protection assay as described in "Materials and Methods." became apparent only after the cells had been exposed to the Twenty wg of total RNA were hybridized to 5 x IO5 cpm of probe overnight. hormone for at least 4-5 days. Since 1246-3A cells produce Protected transcripts (160 nucleotides) were run on a 6% polyacrylamide gel containing 8 Murea at 60 W until the bromophenol blue dye ran off the gel. Dried several polypeptide growth factors which act as autocrine gels were autoradiographed at —¿70°Cwith an intensifying screen for 24 h. Lane growth stimulators, these results suggest that the inhibitory C, negative control corresponding to 20 »igoftRNA; Lane P, full-length probe (230 nucleotides). B, effect of short-term exposure of 1246-3A cells to testoster effect of testosterone might not be direct but rather could be via one on EGF receptor mRNA expression. 1246-3A cells were cultivated in DME- changes in the response to growth factors that the cells produce F12 medium supplemented with 2% ChXCS until they reach confluency. Then one-half of the plates were treated with testosterone for 24 h (Lane +7"), whereas or to which they respond. The inhibition of proliferation caused the other half were maintained in its absence (Lane -T) prior to extracting RNA by testosterone could be due to the fact that either the produc as described above. Conditions for RNase protection assay are the same as in A. tion or the response to the polypeptide growth factors found in Dried gels were autoradiographed at —¿70'Cwithintensifying screen for 3 days in 1246-3A-conditioned medium is being altered by long-term cul order to be able to see EGF receptor transcript in 1246-3A cells. tivation with the hormone. In this paper, we focused on study ing the effect that testosterone would have on the expression of cross-linked EGF receptors in other cell types (17) and probably EGF-competing activity and EGF receptor in the 1246-3A corresponds to native cross-linked EGF receptors and to pro- cells. Several observations support this study: TGF-o or EGF- teolytic fragments generated upon lysis of the cells in spite of like polypeptides and EGF receptors are coexpressed in many the presence of protease inhibitors (17). In agreement with the tumors and tumor cell lines (18, 19); another sex steroid, es cell surface binding data, the intensity of the cross-linked band trogen, which stimulates the growth of mammary epithelial was higher in the cells treated with testosterone (Fig. 5, Lane 1) cells, has been found to modulate TGF-a and EGF receptor than in the ones cultivated in its absence (Fig. 5, Lane 3). expression in normal and malignant mammary epithelial cells Expression of EGF Receptor mRNA in 1246-3A Cells Ex (20, 21); we have shown previously that the insulin-independent posed or Not Exposed to Testosterone. EGF receptor mRNA 1246-3A cells produce insulin and EGF-like competing activi expression was measured by RNase protection assay as de ties which stimulate their growth in an autocrine fashion and scribed in "Materials and Methods." Total RNA was collected which occupy cell surface receptors for insulin and EGF, re from 1246-3A cells which had been cultivated in the absence or spectively (4-7). In this paper we examined the binding of in the presence of testosterone as described above. As shown in sulin, EGF, and FGF, which are growth stimulators of 1246 Fig. 6A, a protected fragment of 160 nucleotides, which is the and 1246-3A cell proliferation. Long-term cultivation of the expected size for a protected EGF receptor transcript, was de 1246-3A cells with testosterone resulted in a 3-fold increase in tected in the 1246-3A cells in both culture conditions. Densit- EGF specific binding without changes in insulin and FGF spe ometric analysis of the autoradiograph indicated that the ex cific binding. This increase in EGF binding corresponded to a pression of EGF receptor mRNA was 10 times higher in the 2.4-fold increase in the number of high-affinity EGF receptors 1246-3A cells cultivated in the presence of testosterone than in without a change in the dissociation constant. In addition, long- cells cultivated in its absence. Interestingly, exposure of the term exposure and not short-term exposure to testosterone re cells to testosterone for 24 h stimulated the expression of EGF sulted in a 10-fold increase in EGF receptor mRNA expression receptor mRNA in the 1246-3A cells by 1.8-fold only (Fig. dB). in 1246-3A cells. It is not yet determined at this point why the This indicates that the effect of testosterone on EGF receptor increase in the level of EGF receptor mRNA caused by the mRNA expression also requires long-term exposure of the cells exposure to testosterone is higher than the increase in EGF to the hormone. receptor number on the cell surface. Binding studies have shown that testosterone stimulates EGF receptor number in the DISCUSSION prostate (22, 23). The data presented in this paper indicate that testosterone increases the number of EGF receptors in a cell The 1246-3A cell line is an insulin-independent cell line type that is other than prostate and that is of mesodermal which has been isolated from the C3H mouse teratoma-derived origin. Moreover, the data in this paper also show that testos adipogenic cell line 1246 maintained in the absence of insulin terone stimulates EGF receptor mRNA expression in the (2). Unlike the parent cell line, the 1246-3A cells have lost the 1246-3A cells. In contrast to its stimulatory effect on EGF ability to differentiate and have become tumorigenic. When receptor expression, testosterone did not affect the level of pro injected to syngeneic hosts at a density of IO6 cells/mouse, the duction of EGF receptor competing activity by the 1246-3A 1246-3A cells give rise to tumors similar to leiomyosarcomas. cells. Thus, it is unlikely that the increase in the number of EGF 4246 Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1992 American Association for Cancer Research. TESTOSTERONE. TUMOR GROWTH. AND EOF RECEPTOR EXPRESSION binding sites is also due to a decrease in the occupancy of cell secreted by a teratoma cell line. Biochem. Biophys. Res. Commun., 135: 533-540, 1986. surface EGF receptors by the secreted factor. Moreover, the 4. Yamada, Y., and Serrerò,G. Purification of an insulin related factor secreted addition of EGF to the culture medium of 1246-3A cells main by a teratoma derived mesoderma! cell line. J. Biol. Chem., 262: 209-213, tained in the presence of testosterone for 9 days when EGF 1987. 5. Cazzano, H., and Serrerò,G. Identification of insulin receptors on the insu receptor number had been up-regulated did not stimulate pro lin-independem variant 1246-3A cell line. J. Cell. Physiol., 136: 348-354, liferation and did not overcome the inhibition of proliferation 1988. 6. Yamada, Y., and Serrerò,G. Autocrine growth induced by the insulin related caused by the hormone (data not shown). However, one cannot factor in the insulin-independent teratoma cell line 1246-3A. Proc. Nati. exclude the possibility that testosterone induces changes either Acad. Sci. USA, 85: 5936-5940, 1988. in the production of TGF-a/EGF-like polypeptides or in the 7. Yamada, Y., and Serrerò,G. Characterization of transforming growth factor produced by the insulin-independent teratoma derived cell line 1246-3A. J. response to these factors in the microenvironment of the cells Cell. Physiol., 140: 254-263, 1989. which could lead to growth inhibition but which are too small to 8. Serrerò, G. EGF inhibits the differentiation of adipocyte precursors in pri be detected in our experiments. In fact such minute changes at mary culture. Biochem. Biophys. Res. Commun., ¡46:194-202, 1987. the level of the microenvironment of cells have been hypothe 9. Serrerò,G., and Mills, D. Decrease in transforming growth factor 01 binding during differentiation of rat adipocyte precursors in primary culture. Cell sized to play an important role in the activation of TGF-0 latent Growth Differ., 2: 173-178, 1991. precursor and in its action (24). Moreover, it has been shown in 10. Serrerò, G.. Zhou, J., Mills, D., and Lepak, N. Decreased transforming growth factor-/? response and binding in insulin-independent teratoma de human breast cancer cells that although estrogen modulates rived cell lines with increased tumorigenic properties. J. Cell. Physiol., 149: TGF-«and EGF receptor expression, blockade of EGF recep 503-511, 1991. tor inhibits TGF-a-induced but not estrogen-induced growth 11. McKeehan, W. L. Growth factor receptors and prostate cell growth. Cancer Surv. Prostate Cancer, //: 165-175, 1991. (25). Based on these results, it is possible to assume that the 12. Lippman, M. E., and Dickson, R. B. Mechanisms of growth control in modulation of EGF receptor expression by testosterone is not normal and malignant breast epithelium. Recent Prog. Horm. Res., 45: 383- necessarily the cause of its growth-inhibitory effect on the 440, 1989. 13. Kan, M., DiSorbo, D., Hou, J., Hoshi, H., Mansson, P. E., and McKeehan, 1246-3A cells either in vivo or in vitro. Since modulation of W. L. High and low affinity binding of heparin-binding growth factor to a 130 EGF receptor number and mRNA expression only occurred kDa receptor correlates with stimulation and inhibition of growth of a dif after long-term exposure of the 1246-3A cells to testosterone, ferentiated human hepatoma cell. J. Biol. Chem., 263: 11306-11313, 1988. 14. Laemmli, U. K. Cleavage of structural during the assembly of the one can assume that this change in EGF receptor is one of the head of bacteriophage T4. Nature (Lond.), 227: 680-685. 1970. 15. Chomczynski, P., and Sacchi, P. Single step method of RNA isolation by acid phenotypic changes that the hormone has induced in the cells. guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., It has been shown in the prostate that programmed cell death 162: 156-159, 1987. induced by castration was accompanied by an increase in the 16. Bringman, T. S., Lindquist, P. H.. and Derynck, R. Different transforming expression of TGF-ßwhich is a growth inhibitor for the pros growth factor-«species are derived from a glycosylated and palmitoylated transmembrane precursor. Cell, 48: 429-440, 1987. tate epithelial cells (11, 26). In the case of the 1246-3A cells, 17. Cohen, S., Ushiro, H., Stoscheck, C., and Chinkers, M. A native 170,000 long-term exposure to testosterone did not cause any change in epidermal kinase complex from shed plasma residues. the level of secretion of TGF-/3 (data not shown). We have J. Biol. Chem., 257: 1523-1531, 1982. 18. DiMarco, E., Pierce, J. H., Fleming, T. P., Kraus, M. H., Molloy, C. J., shown previously that the totality of TGF-/3 produced by the Aaronson, S. A., and DiFiore, P. P. Aulocrine interaction between TGF-a 1246-3A cells is secreted as a latent precursor which is known and the EGF receptor quantitative requirements for induction of the malig to be biologically inactive (27) and that TGF-/3 added exoge- nant phenotype. Oncogene, 4: 831-838, 1989. 19. Salomon, D. S., Kim, N., Saeki, T., and Ciardello, F. Transforming growth nously is a growth simulator for the cells (10). Therefore, it is factor-a An oncodevelopmental growth factor. Cancer Cells (Cold Spring unlikely that modulation in TGF-0 level or response could ex Harbor), 2: 389-397, 1990. 20. Bates, S. E., Davidson, N. E., Valverius, E. M., Fréter,C.E., Dickson, R. B., plain the growth-inhibitory effect of testosterone for the tum- Tarn, J. P., Kudlow, J. E., Lippman. M. E., and Salomon, D. E. Expression origenic 1246-3A cells in vivo and in vitro. However, the pos of transforming growth factor a and its messenger ribonucleic acid in human breast cancer: its regulation by estrogen and its possible functional signifi sibility is not excluded that testosterone affects the level of cance. Mol. Endocrino!., 2: 543-555, 1988. response or production by the cells of yet unidentified growth 21. Salomon, D. S., Kidwell, W. R., Kin. N., Ciardello, F., Bates, S. E., Valver stimulators or growth inhibitors in vitro and in vivo. In addition, ius, E. M., Lippman, M. E., Dickson, R. B., and Stampfer, M. R. Modulation by estrogen and growth factors of transforming growth factor a and epider it is possible that testosterone inhibits tumor growth in vivo not mal growth factor receptor expression in normal and malignant human mam only via changes in the response to local and/or circulating mary epithelial cells. Cancer Res., 113: 57-69, 1989. growth regulators but also by modifying other biological phe 22. St. Arnaud, R., Poyet, P., Walker, P., and Labrie, F. Androgens modulate epidermal growth factor receptor levels in the rat ventral prostate. Mol. Cell. nomena, such as vascularization, which are related and neces Endocrino!., 56: 21-27, 1988. sary for tumor growth in vivo (28). These various possibilities 23. Mulder, E., Van Loon, W., De Boer, A. L., Schuurmans. A. L. G., Bolt, J., Voorhorst, M. M., Kuiper, G. G. J. M.. and Brinkman, A. O. Mechanism of are currecntly under investigation in our laboratory. androgen action: recent observations on the domain structure of androgen receptors and the induction of EGF receptors by androgens in prostate tumor cells. J. Steroid Biochem.. 32: 151-156, 1989. ACKNOWLEDGMENTS 24. Lyons. R. M., Keski-Oja, J.. and Moses, H. L. Proteolytic activation of latent transforming growth factor-/? from fibroblast conditioned medium. J. Cell The authors wish to thank Dr. Mikio Kan for the gift of '"I-FGF, Biol., 106: 1659-1665, 1988. Marina LaDuke and Carol Baine for the illustrations, and Valerie 25. Arteaga, C. L., Coronado, E., and Osborne, C. K. Blockade of the epidermal growth factor receptor inhibits transforming growth factor «-inducedbut not Oliver for preparation of the manuscript. estrogen-induced growth of hormone dependent human breast cancer. Mol. Endocrino!., 2: 1064-1069, 1988. REFERENCES 26. Kyprianou, N., and Isaacs, J. T. Expression of transforming growth factor-/? in the rat ventral prostate during castration-induced programmed cell death. 1. Serrerò, G., and Khoo, J. C. An in vitro model to study adipose differentia Mol. Endocnnol., 3: 1515-1522, 1989. tion in serum free medium. Anal. Biochem., 720: 351-359, 1982. 27. Pircher, R„Lawrence, D. A., and Jullien, P. Latent ¿(-transforminggrowth 2. Serrerò, G. Tumorigenicity associated with loss of differentiation and of factor in nontransformed and kirstein sarcoma virus transformed normal rat response to insulin in the adipogenic cell line 1246. In Vitro Cell. Devel. kidney cells clone 49F. Cancer Res., 44: 5538-5543, 1984. Biol., 21: 537-540. 1985. 28. Cross, M., and Dexter, T. M. Growth factors in development, transformation 3. Yamada, Y., and Serrerò, G. Characterization of an insulin related factor and tumorigenesis. Cell, 64: 271-280, I991.T

4247 Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1992 American Association for Cancer Research. Effect of Testosterone on the Growth Properties and on Epidermal Growth Factor Receptor Expression in the Teratoma-derived Tumorigenic Cell Line 1246-3A

Ginette Serrero, Nancy M. Lepak, Jun Hayashi, et al.

Cancer Res 1992;52:4242-4247.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/52/15/4242

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/52/15/4242. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1992 American Association for Cancer Research.