Androgen Effects Mediated by Estrogen Receptor in 7,12- Dimethylbenz(A)Anthracene-Induced Rat Mammary Tumors1

Home , Estradiol, Estrogen

[CANCER RESEARCH 38, 3922-3929. November 1978] 0008-5472/78/0038-0000$02.00 Androgen Effects Mediated by Estrogen Receptor in 7,12- Dimethylbenz(a)anthracene-induced Rat Mammary Tumors1

Marcel Garcia and Henri Rochefort2

INSERM U 148, 60, rue de Navacelles, 34100 Montpellier, France

ABSTRACT On the basis of previous studies of this (4, 19) and other (21, 22, 26) laboratories, we have instead considered the To improve our understanding of the effects of andro- possibility of a direct interaction of androgens on the ER gens on the induction and growth of mammary tumors, localized in the mammary tumor. In the rat uterus, DHT has the interaction of androgen on the estrogen receptor (ER) actually been shown to bind to ER with a low affinity to has been evaluated in the 7,12-dimethylbenz(a)anthra- translocate it into the nucleus and to induce specific estro- cene-induced rat mammary tumor. In vitro competitive genie effects (4,19). The high doses of androgens currently experiments have shown that androgens interact on ER used in the treatment of human breast cancer then appear with a low affinity but have a good stereospecificity for C,9 to be sufficient to occupy and eventually to activate the ER steroids bearing a 170- and/or 3/3-hydroxyl group. The localized in the tumor cells. administration of one dose ( -20 mg) of 5a-dihydrotestos- We now report the effect of a single high-dose injection terone to 1-day ovariectomized rats induced ER nuclear of DHT on ER in rat mammary tumors induced by DMBA. translocation in 92% of the mammary tumors. In 60% of With the use of this experimental system, the effect of these tumors, this translocation was followed by a signifi androgens appears to be biphasic. At moderate doses (0.5 cant stimulation of the rate of [3H]leucine incorporation to 4 mg/rat), androgens induce tumor regression in 60% of into cytosoluble proteins by 5a-dihydrotestosterone or the cases (7) while at higher doses they either have no estradici. When injected together, 5a-dihydrotestosterone effect or they stimulate tumor growth (5). did not antagonize the effect of estradiol. We conclude that one very high dose of 5a-dihydrotes- MATERIALS AND METHODS tosterone that stimulates leucine incorporation into pro teins is also able to occupy the cytosol ER in mammary Steroids. [6,7-3H]Estradiol (specific activity, 58 Ci/mmol) tumors and to translocate it in vivo into the nucleus. We and [1,2-3H]dihydrotestosterone (specific activity, 30 Ci/ suggest that a direct interaction of androgens with the ER mmol) were obtained from CEA (Gif-sur-Yvette, 91, France). located in mammary tumors is responsible, as in rat uteri, Non radioactive steroids were kindly given by Roussel for the stimulating effect of very high doses of androgens. Uclaf (Romainville, 93, France, Dr. Martel). Dromostano- We cannot ascertain that the same mechanism is involved lone (2a-methyl-17/3-hydroxy-5a-androstane-3-one) and in the androgen-induced regression of the mammary tu dromostanolone propionate were kindly supplied by Syntex mors. Laboratories, Inc. (Palo Alto, Calif., Dr. Dorfman). After injection of [3H]DHT (specific activity, 0.6 mCi/ INTRODUCTION mmol), radioactive steroids were extracted twice from tu mor homogenate with ethanol. Extracts were analyzed on It is generally agreed that estrogens via their receptors Merck F 254 thin-layer silica gel sheets with chloroform: are involved in the induction and growth of breast cancer. ethanol (98:2). In some cases, DHT was purified twice by The role of androgens is less clearly established. These the same system before use. hormones can stimulate growth of androgen-responsive Tumor Induction. Virgin female Sprague-Dawley rats (Iffa mammary tumors such as the Shionogi 115 (1) and prevent Credo, Saint-Germain-sur-l'Arbresle, 69, France) were growth of human breast cancer, for which the androgens housed in a light (14 hr/day)- and temperature (22Â°)-con- give a 30% remission rate (12). trolled environment and were fed a diet of A 04 (UAR S.A., The mechanism by which androgens induce this mam Villemoisson, France) and tap water ad libitum. Mammary mary tumor regression is unknown, and several hypotheses tumors were induced in 50-day-old rats by administration of have been proposed (13, 23) which deal either with an 15 mg DMBA in 1 ml corn oil by gastric intubation under indirect effect of androgen on the hypothalamopituitary light ether anesthesia. Two to 3 months later, rats that axis or with a direct effect on the mammary tumor itself. produced at least one mammary tumor of 1 cm diameter The possibility of androgens having an indirect effect were ovariectomized under light ether anesthesia. through their aromatization into estrogens has also been Biopsy Experiments. One or 2 days after the ovariectomy, considered; however, this is very unlikely since androgens a tumor biopsy (approximately 400 mg of tissue) was per such as DHT3cannot be aromatized (11). formed under light ether anesthesia, and the remaining tumor was collected 1 to 24 hr after s.c. injection of a single Received February 17, 1978; accepted July 26, 1978. dose of 50 mg DHT dissolved in corn oil or after injection of 1This work was supported by the Institut National de la Sante et de la Recherche Medicale and the Fondation pour la Recherche MÃ©dicale. 1 To whom requests for reprints should be addressed. 3The abbreviations used are: DHT, 5a-dihydrotestosterone; ER, estradiol EDTA, 12 mu monothioglycerol HCI butter (pH 7.4); DCC, dextran-coated charcoal; TCA, trichloroacetic acid; Re, cytosolic estrogen receptor; Rn, receptor; DMBA, 7,12-dimethylbenz(a)anthracene; TET, 10 mm Tris, 1.5 mm; nuclear estrogen receptor.

3922 CANCER RESEARCH VOL. 38

oil as a control. In one case, a 2-day ovariectomized rat was were precipitated with cold TCA and hydrolyzed in 0.1 N anesthetized with Nembutal (20 mg/kg), and several biop NaOH at 80Â°as previously described (4). sies were collected either before (time 0) or during the 5 hr Miscellaneous Techniques. Protein concentrations were following the injection of DHT or oil. The biopsy tissue was evaluated according to their absorption at 280 and 260 nm dissected free of fat and necrotic tissue and was used (9). The radioactivity was counted in a SL 30 scintillating immediately or frozen in liquid nitrogen and stored for 1 to counter (Intertechnique, Plaisir, 78, France) in 10 ml PPO 3 days at -20Â°. (0.3%), POPOP (0.01%), toluene scintillation mixture with a Preparation of Extract. The tumor tissue was washed in 20 to 25% efficiency for tritiated steroids as evaluated by TET buffer, and homogenized in 10 volumes of TET buffer external standard. The hydrolyzed TCA-insoluble material with a glass:glass Potter-Elvehjem homogenizer. The ho- containing [3H]leucine was counted in 5 ml Triton X-100 mogenate was centrifuged at 250,000 x g for 60 min, and and 10 ml of the toluene scintillation mixture with a 20 to the supernatant was defined as the cytosol. The pellet was 22% efficiency. homogenized in TET buffer plus 0.4 M KCI (pH 8.5), ex tracted 40 min at 2Â°,and then ultracentrifuged at the same RESULTS speed. The KCI paniculate extract obtained was shown to contain the majority of the extractable nuclear receptors Binding Specificity of Androgens on the Estrogen Re (8). ceptor. Androgens have been shown to compete specifi Estrogen Receptor Assays. The ER sites of cytosol and cally with estradiol on the uterine ER (19). Similar experi paniculate KCI extract were generally assayed after incu ments were done with the cytosol prepared from the DMBA bation with DCC (charcoal, 0.5%:dextran, 0.05%) for 2 hr at rat mammary tumors and by use of the same "sensitized" 2Â°with 2 saturating concentrations (5 and 10 nM) of competitive experiments in which the nonradioactive an [3H]estradiol and a 200-fold excess of diethylstilbestrol for drogens were preincubated with cytosol before addition of nonspecific binding. The details and validity of this in vitro [3H]estradiol for a short period of time (see "Materials and Methods"). This procedure allowed us to detect a signifi assay have already been shown in the DMBA mammary tumor (24). After DHT injection, the estrogen receptors cant inhibition of estradiol binding by weak affinity ligand, were labeled in vitro by [3H]estradiol with an exchange such as testosterone, which could not be shown by the technique previously tested on the rat uterine extracts (19). classical competitive experiments performed at equilib The free steroids were removed by incubation with DCC for rium. 15 min at 2Â°.The extracts were then incubated in the Table 1 shows that the highest competitive efficiency was presence of 10 nM [3H]estradiol plus 2 /Â¿Mdiethylstilbestrol found for androgens containing 2 hydroxyl groups at the for 15 hr at 2Â°.The total ER sites (accessibles plus ex 3/3 and 17/3 positions such as A5-androstene-3/3,17/3-diol changed at 2Â°)of the extracts were then assayed in dupli and 5a-androstane-3/3,17/3-diol. The removal of one of cate with DCC. All values were corrected for nonspecific these groups decreased the affinity for ER, whereas 17a- binding determined by parallel incubation with 2 /Â¿Mnon- testosterone and the C21 steroids, like progesterone, dis radioactive diethylstilbestrol. played no affinity. Dromostanolone propionate was com Competitive Experiments. Nonradioactive androgens (5 pletely inefficient to compete on ER contrary to the free MM) were incubated in tumor cytosol for 1 hr at 2Â°.Then [3H]estradiol (1 nw) was added for 5 min, and the associa Table 1 tion was stopped by adding non radioactive estradiol (1 P.M). The [3H]estradiol-receptor complexes were then assayed by Binding specificity of androgens on the estrogen receptors in the DMBA-induced rat mammary tumors DCC adsorption. After a 5-min incubation, the proportion of Steroid"NoneA5-Androstene-3/3 of [3H]estradiolbound*10021 the nonspecific binding was negligible (1 to 3%) and was therefore not corrected. ,17/3-diolSa-And Â±2e25 Sucrose Gradient Ultracentrifugation. Aliquots of cytosol and nuclear extracts labeled with [3H]estradiol were ana 7/3-diol5a-DihydrotestosteroneDromostanoloneDehydroepiandrosteroneTestosterone1rostane-3/3 ,1 Â±535 lyzed in a 5 to 20% linear sucrose gradient at 206,000 x g for 13 to 14 hr at 2Â°in an SW 50 rotor with a Beckman L2 65 Â±1036 B ultracentrifuge. Two-drop fractions were collected after Â±545 Â±1262 puncturing the tube and were counted. Â±3106 In Vitro [3H]Leucine Incorporation. Immediately after re moval, the tumor tissue was cut into 1-cu mm pieces and 7a-Testosterone5/3-Di Â±8101 rinsed in incubation medium. Two lots of 50 mg of the hydrotestosteroneDromostanolone Â±798 sliced tumor were incubated separately in a 20-ml flask propio Â±16103 containing 4 ml minimal essential medium/Earle's salt so nateProgesterone% Â±13 lution with 5 Â¿tCi/ml [3H]leucine (specific activity, 1 Ci/ " Cytosol aliquots from mammary tumors collected 2 days after mmol). The incubation was carried out in a shaker-incuba castration were preincubated for 1 hr with 5 UM of the indicated tor at 37Â°under an O2:CO2 (95:5) atmosphere. Tumor pieces radioinert steroids and then for 5 min with 1 nw [3H]estradiol as were then rinsed 3 times with 10 ml ice-cold TET buffer and described in "Materialsand Methods." '' Results are expressed as percentage of bound [3H]estradiol homogenized in 2 ml of the same solution. Homogenates measured without competing steroid. were centrifuged at 27,000 x g for 50 min at 2Â°.The soluble c Mean Â±S.E. of the values determined from at least 3 different proteins in 2-aliquot fractions (200 /Â¿I)of the supernatant tumors.

NOVEMBER 1978 3923

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1978 American Association for Cancer Research. M. Garcia and H. Rochefort dromostanolone, indicating that the 17/3-hydroxyl must be free. We conclude that the binding specificity of estrogen receptors for androgens is similar in mammary tumors to CYTOSOL that observed in uteri. In Vivo Nuclear Translocation of ER Induced by DHT. Twenty four hr after a bilateral ovariectomy, the estrogen receptors were assayed in the cytosol and in the KCI particulate extract before and 2 to 5 hr after DHT treatment with the DCC exchange assay at 2Â°as described in "Materi als and Methods." We controlled the validity of this ex change technique on mammary tumor cytosol after in vitro occupation of ER sites by 5 /XM DHT, 5a-androstane- 3/3,17/3-diol, or 5 nM estradiol. As shown in Chart 1, after 15 hr incubation at 2Â°the bound DHT or 5a-androstane- LU 3/3,17/3-diol could be totally displaced by [3H]estradiol, h* ÃœÃ– while the bound estradiol was only partially exchanged. The fact that we recovered the majority of the solubilizable (Re plus Rn) ER sites after in vivo DHT injection in most of the tumors also validated this exchange procedure. As shown in Chart 2, the level of ER in the cytosol of biopsies removed 1 day following ovariectomy varied widely from 1 to 9 pmol/ g of tumor (or 30 to 300 fmol/mg cytosol protein). The Biopsy DHT Biopsy DHT injection of 20 to 50 mg DHT in the rat induced a decrease Chart 2. Effect of DHT on the repartition of the estrogen receptors in of ER concentration in the cytosol and an increase in the cytosol and particulate extract. O, total (accessible plus exchanged at 2Â°) ER concentration in cytosol (a) or KCI particulate extract (Â£>)asdetermined particulate extract in 75 and 67% of the tumors, respec in duplicate with the exchange technique described in "Materials and tively. After DHT treatment, the average decrease of ER in Methods." Unes connect the values of the biopsy and of the same tumor the cytosol was 1.26 Â±0.47 (S.E.) pmol/g tumor (p < 0.02), collected 2 to 6 hr after injection of 20 or 50 mg DHT. The number of each tumor is represented. â€¢¿,meanvalues of the 12 tumors studied. Bars, S.E. and its average increase in the particulate extract was 0.54 *, same experiment with purified DHT (as described in the text). Â±0.2 pmol/g tumor (p < 0.02) (Chart 3). In spite of a large variation according to the tumor, it generally appeared that the totality of the solubilizable ER (Re plus Rn) was not â€¢¿2â€¢ recovered after DHT treatment. In these experiments, we 1--2--i-O7 did not specify whether this apparent loss of sites was due C Â«- o to some artifact or whether it was biologically significant. In â€”¿ OT OÃ•":Ã•'Â°"CYTOSOL,â€žCONItÃ¬eO'{ Tumor 3 no translocation was observed because the Re and Rn sites remained constant. However, the DHT-induced nuclear translocation of ER was significant in 92% of the tumors and constant in all tumors tested 21/2 hr after DHT i!O9

Qx lÃ•J*

O"O'ROL

PARTICULATEEXTRACT

Chart 3. Variation of ER concentration in cytosol and particulate extract after DHT treatment. Results of Chart 1 are plotted according to the variation of the nuclear and cytosol ER concentrations in the DHT-treated tumor as compared to the biopsy value taken as 0. â€¢¿,meanÂ±S.E. for the 12 tumors studied. The significance for the mean of biopsy content versus the mean of DHT-treated tumors is p < 0.02 both for cytosol and particulate extract.

hours exchange at 2Â° Chart 1. Differential exchange at 2Â°ofestradiol and of androgen. Aliquots injection. of mammary tumor cytosol were incubated for 45 min with 5 nM estradiol With DHT that had been purified further by 2 successive (O). 5 tiU DHT (â€¢),5MM 5a-androstane-3/3,17/3-diol (Ad/o/; A) or without silica gel chromatographies as described in "Materials and hormone (100%). The free hormone in excess was then removed by adding a Methods," the nuclear translocation of ER was also dem pellet of DCC for 15 min. After centrifugation of DCC each sample was further incubated at 2Â°for the indicated periods of time with 10 nM onstrated (Chart 2,0). |'H]estradiol alone or together with a 200-fold excess of nonradioactive The time course of the DHT-induced nuclear transloca estradiol for nonspecific binding. The specifically bound ['HJestradiol (8s) was obtained with DCC assay as described in "Materials and Methods." The tion was followed with successive biopsies in one rat time 0 values were obtained by competition between 2 nM [JH|estradiol and anesthetized with Nembutal and bearing 3 different tumors the same concentrations of nonradioactive hormones incubated simultane ously for 1 hr. Results are expressed as the percentage of the specific (Chart 4). The lack of effect of the anesthetic alone on the |'H]estradiol binding measured at each time without nonradioactive steroid. localization of ER already described by Nicholson eia/. (16)

3924 CANCER RESEARCH VOL. 38

Table 2 Absence of depletion of Re 24 hr after DHT The estradiol cytosol receptor sites were assayed with the ex change technique before (biopsy) and 24 hr after DHT (50 mg) injection. The percentage of variation of the tumor area at biopsy as compared to the tumor area at ovariectomy was evaluated 70 according to the product of the 2 major diameters. [3H]Estradiol cytosol E sites (fmol/mg protein) CL Days after % of change Tumor castration in tumor area Biopsy DHT-treated I/) LU 1314152660-30+516728361492540 (51 (6l Rn l/l it) (M 20- "â€¢$â€¢ ID 5). i 2 Â¿ DHT 100mg hours Nembutal Chart 4. Time course of the ER nuclear translocation induced by DHT. Re and Rn were assayed in biopsies taken from 3 separate tumors (Tumors 7, 8, and 9) developed in the same rat either before or at the indicated times following DHT injection. Results are expressed as a percentage of the cytosol receptor level of each tumor before DHT injection. Points, mean; oars.S.E. E CL was confirmed in a parallel experiment (not shown). DHT was able to decrease partially the Re concentration and to increase the Rn concentration. A maximal ER translocation 2 2- was observed 2 to 3 hr after DHT as already shown in the rat uterus (19), and the Re sites appeared to be subsequently ÃœJ replenished. In fact, in 3 other tumors analyzed separately, I a good recovery of the Re binding sites was noticed 24 hr after DHT injection, suggesting an actual replenishment of the cytosol receptor (Table 2). The nuclear estrogen-binding proteins present in the KCI particulate extract before and after DHT treatment were analyzed in a sucrose gradient after their in vitro labeling with [3H]estradiol (Chart 5). After DHT treatment and a subsequent exchange for 20 hr with [3H]estradiol, a well- defined saturable 3 to 4S binding peak was observed. On the other hand, in the absence of DHT (Chart 5, 20-hr control) or without the 20-hr exchange procedure (DHT 1 Chart 5. Sucrose gradient analysis of the nuclear ER before and after DHT hr), the amount of 3 to 4S binding peak was much less treatment. In a 1-day castrated rat, a biopsy (control) was taken before important, which suggests that additional nuclear ER sites injection of DHT (50 mg). After 3 hr the remaining tumor was collected. The particulate KCI extracts containing the nuclear ER were then prepared and had been translocated to nuclei by DHT. The significance incubated in vitro with radioactive estradiol at 0 to 2Â°for1 or 20 hr and finally of the saturable binding peak observed after 1-hr (DHT) or analyzed in a sucrose gradient containing KCI as described in "Materials 20-hr (control) labeling with [3H]estradiol was not specified. and Methods." A, 20-hr control; O, DHT-treated and exchanged for 20 hr by 10 HM ['Hlestradiol; â€¢¿.DHT-treatedand exchanged for 20 hr by 10 DM It could be due to accessible or very easily exchangeable [3H]estradiol plus a 100-fold excess of unlabeled DES; x, treated and ER sites. incubated for 1 hr with 5 nM [3H]estradiol to give the accessible ER sites. The nature of the ligand responsible for the nuclear translocation of ER after administration of DHT was not account for the weak in vivo ability of DHT to translocate specified. However, thin-layer chromatography of the tumor ER in mammary tumors. These results indicated that DHT extract obtained 3 hr after injection of 50 mg [3H]DHT was able to translocate ER to the nucleus in 92% of the (specific activity, 0.6 mCi/mmol) showed that =80% of the tumors studied. However, the degree of the nuclear trans- radioactivity migrated as nonmetabolized DHT, 10% as a location varied markedly according to the experiment and 5a-androstanediol and 10% as 2 more polar compounds. was more visible for the ER-rich tumors. The evaluated DHT uptake in the tumor (=1.7 /xg/g tissue) Effect of DHT on the Rate of [3H]Leucine Incorporation gave an approximate final concentration of 0.4 V.M in a into Soluble Proteins. The incorporation of [3H]leucine into cytosol of 2 mg protein per ml. This relatively low concen soluble proteins was assayed after in vitro incubation of tration of DHT as compared with the high dose injected may pieces of tumor collected either before or 24 hr after an in

NOVEMBER 1978 3925

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1978 American Association for Cancer Research. M. Garcia and H. Rochefort We concluded that a single high dose of DHT was able to Â° Biopsy o Biopsy stimulate leucine incorporation into tumoral proteins of â€¢¿DHT - trtoud â€¢¿Oil- trroted castrated rats but was unable to prevent estradiol action on the same parameter.

DISCUSSION When rats bearing DMBA-induced mammary tumors were treated by a high pharmacological dose of DHT, we ob served 2 kinds of responses. On the one hand, the estrogen receptor sites are occupied in vivo by weak affinity ligand(s) and are subsequently transferred to the nucleus. On the Chart 6. Time course of the incorporation of [3H]leucine into protein other hand, the general protein synthesis is increased in before and after DHT or oil injection. Two days after ovanectomy and approximately 60% of the tumors. The interaction of andro- immediately after biopsy, 2 rats were treated with 50 mg DHT (Tumor 17) or gens with the ER sites has already been shown in the rat with oil as a control (Tumor 18). The biopsy and remaining tumors collected 24 hr after treatment were then incubated in vitro with [3H]leucine for the uterus (4, 19). Here, however, the degree of nuclear trans- indicated periods of time. The [JH]leucine incorporation into protein was location varied markedly according to the tumor. The rea determined as described in "Materials and Methods." son for this variation has not been specified. It could be due to a different degree of DHT uptake or metabolism or to vivo injection of oil or 50 mg DHT (Chart 6). The in vitro different hormone responsiveness of the tumor. In addition, [3H]leucine incorporation into TCA-precipitable material the proportion of ER transferred to the nucleus was not as was checked to be linear with time during a 2-hr incubation high as that obtained with estradiol in the same type of period. It was also homogeneous for different localizations tumors (20) or with DHT in rat uterus (19). Competitive in the same tumor since it varied less than 8% according to experiments performed with the cytosol from mammary the tissue samples. However, a large variability was noticed tumors indicate that the binding specificity of androgens from one tumor to another since the [3H]leucine incorpo for ER in the rat mammary tumor is similar to those rated into protein varied between 2,000 and 15,000 cpm/mg observed in the rat uterus (19) and in human uterus and protein per hr. breast cancer (18). Recently, Daviesef al. (2) independently As shown in Table 3, a significant stimulation of reached the same conclusion on the rat mammary tumors. [3H]leucine incorporation into protein was induced by DHT Dromostanolone propionate which is currently used in vivo in 60% of the treated tumors. In Tumors 7,12, and 19 where to treat breast cancer was unable to compete with ER, the ER nuclear translocation induced by DHT was verified, contrary to the free steroid which is probably liberated in the [3H]leucine incorporation was also stimulated by DHT. vivo by hydrolysis. The ER nuclear translocation observed In addition, the lack of response in some tumors seemed to in vivo after DHT injection is in agreement with the binding be independent of the time following DHT injection (Tumor affinity of DHT for ER as determined in vitro and strongly 20). A significant stimulation of [3H]leucine incorporation suggests a direct interaction of androgens on the estrogen was also obtained with estradici (10 /u.g) in =50% of the receptor under in vivo conditions. A decrease of the in vivo tumors. It is probable that the proportion of responsive uptake of [3H]estradiol had been suggested in mammary tumors is underestimated. Actually, in Tumor 26, only tumors after injection of up to 1 mg testosterone (14) and in the second injection appeared to be efficient. When the rats rat uterus after injection of 3 mg DHT (M. Garcia, unpub received both DHT and estradici, a similar [3H]leucine lished observations). In our case, the nature of the andro- incorporation was observed as with estradici alone, sug gen responsible for the ER nuclear translocation, either gesting that DHT did not antagonize the estradici effect. untransformed DHT and/or DHT metabolite(s), cannot be Actually, the means of the stimulation of leucine incorpo ascertained, since in addition to a majority of DHT other ration by DHT, estradici, and DHT plus estradici were 140 compounds like androstanediol have been extracted from Â±16, 188 Â±30, and 203 Â±50%, respectively, and were not the tumor. The responsibility of a high-affinity metabolite of statistically different. Furtherexperiments would be needed DHT present in small amounts could not be excluded by to ascertain an additive effect of estradiol and DHT. When thin-layer chromatography; however, it is unlikely on the rats were treated by oil in control experiments, the basis of the rapid exchangeability of ER sites obtained at [3H]leucine incorporation decreased from Days 2 to 3 after 2Â°.Similarly, the involvement of a minute amount of con castration. This decrease, already described by others (6, taminating estrogens in the DHT preparations is very un 15) was probably due to ovariectomy and could be respon likely for the following reasons, (a) As shown in Chart 1 sible for an underestimation of the stimulating effect of DHT after estradiol binding, it would not be possible to exchange and estradiol. totally the sites at 2Â°;if so, the recovery of ER sites would On the basis of the studies performed on the uterus (4), it have been very low. The same difference in exchangeability is most likely that the observed increase of leucine incor was also found after in vivo occupation of ER sites by poration into protein was mainly due to the stimulation of estradiol or DHT (not shown), (b) A contamination by general protein synthesis. However, other possibilities were estrogens would not be in agreement with the binding not excluded since the TCA-soluble [3H]leucine was also stereospecificity of the different androgens shown in Table slightly increased by DHT (not shown). 1. (c) We have recently demonstrated the direct binding of

3926 CANCER RESEARCH VOL. 38

Table 3 In vivo effects of DHT and estradiol on the in vitro [3H]leucine incorporation into soluble proteins Two pieces of the same tumor of a 2-day ovariectomized rat were removed either before or at the indicated time following the injection of oil, DHT (50 mg), estradiol (10 ju.g),or DHT (50 mg) plus estradiol (10 /Â¿g).Thetissue samples were then immediately incubated at 37Â°in minimal essential medium containing [3H]leucine for 1 hr, and the incorporation of [3H]leucine into protein was assayed as described in "Materials and Methods." The [3H]leucine incorporation in the hormone-treated tumor is expressed as the percentage of the values obtained in the biopsy of the same tumor collected before treatment. For each biopsy 2 to 4 separate incubations were performed, and the mean Â± S.D. was taken as the 100% value, p values express the statistical significance of the variation of [3H]leucine incorporation observed in the treated tumor as compared to the biopsy. [3H]Leucine incorporation in % of biopsy Time following Control bi- opsyDHT Tumor nÂ° injection (hr) tumor171 (50mg)712192021221723Oil1824Estradiol Â±2a100 Â±5129 Â±2100 Â±3170 Â±6100 Â±975 Â±976 Â±17100 Â±961 Â±1132 Â±14100 Â±14171 Â±4100 Â±16196 Â±7100 Â±1182 Â±5100 Â±86t

Â±15100 Â±268 Â±13100 Â±2339 Â±2224 Â¿Â¿g)252626e27e28e29eEstradiol(10 Â±10100 Â±2857 Â±7100 Â±7208 Â±33100 Â±32225 Â±6100Â± Â±19263 2100 Â±16155 Â±4100 Â±21253 mg)30313233e34e4444112417242428244242424d242424242424ee100(10 /Â¿g)+DHT (50 Â±12100 Â±16333 Â±16100 Â±1638 Â±12100 Â±7247 Â±10100 Â±10147 Â± 5Treated Â± 5P<0.01<0.01<0.01NS6NSNSNS<0.05<0.001NS<0.05NS<0.05<0.05<0.05NS<0.05<0.01NS<0.02<0.01<0.05<0.01<0.02 " Mean Â±S.D. for 2 to 4 separate incubations. 6 NS, not significant. c This experiment was performed 6 days after castration. A'e A second injection of estradiol (d) or of estradiol plus DHT (e) was administered 72 hr after a first injection of estradiol (10 /Â¿g).The effect was observed 24 hr after the second injection. androgens to ER with tritiated androgens.4 Finally, the DHT ble that the stimulation of protein synthesis observed after used had been crystallized twice. In control experiments, DHT administration is due to the ER nuclear translocation. DHT purified by 2 successive silica gel chromatographies In addition to ER the progesterone receptor and the andro- displayed the same nuclear transfer ability as did ER (Chart gen receptor, which both bind DHT, could also be respon 2,*). We have checked by adding [3H]estradiol to DHT that sible a priori for the stimulating effect of DHT observed. The the contamination of DHT by estradiol or estriol was less androgen receptor sites are present in the DMBA tumor, than 10~5after these purification steps. although in small concentrations (0 to 30 fmol/mg protein) The metabolic response to DHT reported in this paper is (F. Vignon and M. Garcia, unpublished observations). How the stimulation of leucine incorporation into protein. This ever, the growth of the DMBA rat mammary tumors was effect agrees with the increase of tumor size also observed never stimulated by doses of androgens that are able to in some hormone-dependent DMBA mammary tumors after occupy the androgen receptor sites but not the progester injection of DHT (50 mg) daily for 5 days (unpublished one or the estrogen receptor sites. An interaction of andro observations) or of testosterone propionate (5). It is proba- gens on the progesterone receptor cannot be excluded here since progesterone has been shown to stimulate DMA 4 M. Garcia and H. Rochefort, manuscript in preparation. synthesis in rat mammary tumors (17).

NOVEMBER 1978 3927

However, evidence suggests that the ER-androgen com the cytosol ER in mammary tumors and to translocate it in plexes are in fact responsible for the stimulating effect of vivo into the nucleus. We hypothesize that this androgen- very high doses of DHT. The time- and dose-response induced translocation of ER is responsible for the stimulat studies of the ER nuclear translocation and of the ing effect of very high doses of androgens on protein [3H]leucine incorporation are similar to those observed in synthesis and tumor growth. However, we cannot now the rat uterus (4). Moreover, preliminary experiments have ascertain whether the same receptor mechanism is involved shown that DHT as estradiol induces the progesterone in the regression of mammary tumors provoked by lower receptor in these DMBA rat tumors." A similar induction of doses of androgens. the progesterone receptor has been shown in the MCF7 human breast cancer cell line (26). Finally, as in the case of estrogens, not all tumors were responsive to DHT. Actually, ACKNOWLEDGMENTS in this study, only 60% of the tumors were stimulated by DHT for leucine incorporation, while most of the tumors We are grateful to J. Vanbiervliet for her excellent technical assistance. were responsive to ER nuclear translocation. However, we We thank E. Barrie and H. Spowart for their help in the preparation of the could not ascertain that the DHT-resistant tumors were also manuscript. estradiol insensitive. The relationship between the effects of DHT reported here and the mammary tumor regression observed after REFERENCES repeated injections of lower doses of androgens in the 1. Bruchovsky, N., Sutherland, D. J. A., Meakin, J. W.. and Minesita T. noncastrated rat, known to provoke a decrease of tumor Androgen Receptors: Relationship to Growth Response and to Intracel- speculate that the DHT-induced regression of tumor is lular Androgen Transport in Nine Variant Lines of the Shionogi Mouse Mammary Tumors. Biochim. Biophys. Acta, 387: 61-71, 1975. mediated by ER. Statistical analysis of the rate of response 2. Davies, P., Powell-Jones, W., Nicholson, R. I., and Griffiths, K. The of breast cancer to androgen therapy indicates a correlation Specificity of the Oestrogen Receptor of DMBA-lnduced Mammary with the ER content (12) but not with the androgen receptor Tumours of the Rat. European J. Cancer, 13:1421-1427,1977. content (3). More precisely, 90% of human breast cancer 3. Engelsman, E., Persijn, J. P., and Korsten, C. B. Experience with Oestrogen and Androgen Receptors in Breast Cancer and Correlation containing no ER did not respond to androgen therapy. with Clinical Data. Rev. Endocrine Related Cancer, in press, 1978. However, after repeated injections of 1 to 4 mg DHT per 4. Garcia, M., and Rochefort, H. Androgens on the Estrogen Receptor. II- Correlation between Nuclear Translocation and Uterine Protein Synthe noncastrated rat, known to provide a decrease of tumor sis. Steroids, 29: 11-126, 1977. growth, the ER nuclear translocation could not be ascer 5. Meise, E., and Gorlich, M. Growth and Therapy of Mammary Tumors Induced by 7,12-Dimethylbenzanthracene in Rats. Brit. J. Cancer, 20: tained 24 hr after the last injection (25). Preliminary experi 539-545,1966. ments indicated to us that these low doses of DHT were 6. Hilf, R., Battaglini, J. W., Delmez, J. A., Cohen, N., and Rector, W. D. also insufficient 2 to 3 hr after DHT injection, which are the Some Biochemical Changes Preceding Regression of 7,12-Dimethyl- optimal conditions for evaluation of the ER nuclear trans- benz(a)anthracene-induced Mammary Tumors following Oophorectomy. Cancer Res., 3Ã•:1195-1200,1971. location (unpublished observations). Conversely, at higher 7. Muggins, C., Briziarelli, G., and Button, H., Jr. Rapid Induction of doses giving the nuclear translocation of ER as reported Mammary Carcinoma in the Rat and the Influence of Hormones on the Tumors. J. Exptl. Med., 709: 25-42, 1959. previously, tumor growth was stimulated rather than in 8. King, R. J. B., Smith, J. A., and Steggles, A. W. Oestrogen Binding and hibited. For explanation of these discrepancies, we propose the Hormone Responsiveness of Tumors. Steroidologia, 7: 73-88, 1970. that DHT at very high doses is an agonist and that the DHT- 9. Layne, E. Spectrophotometric and Turbimetric Methods for Measuring Proteins. Methods Enzymol. 3: 447-454, 1957. ER interaction reported in this paper is responsible for the 10. Maynard, P. V., Pike, A. W., Weston, A., and Griffiths, K. Analysis of stimulation of protein synthesis and tumor growth subse Dehydroepiandrosterone and Androstenediol in Human Breast Tissue Using High Resolution Gas Chromatography-Mass Spectrometry. Euro quently observed. For lower doses of androgens that are pean J. Cancer, 73: 971-975, 1977. known to induce tumor regression but do not seem to 11. McGuire, J. S.. Hollis, V. M., and Tomkins, G. M. Some Characteristics provoke ER nuclear translocation, other mechanisms must of the Microsomal Steroid Reductases (5a) of Rat Liver. J. Biol. Chem., 235:3112-3117, 1960. be invoked. In this case, DHT could still be acting via ER as 12. McGuire, W. L. Current Status of Estrogen Receptors in Human Breast an estrogen antagonist by binding to Re without activating Cancer. Cancer, 36: 638-644, 1975. 13. McGuire, W. L. Physiological Principles Underlying Endocrine Therapy it, or the effect of DHT could be mediated by a completely of Breast Cancer. In: W. L. McGuire (ed.), Breast Cancer, Vol. 1, p. 217- different receptor mechanism. 262. New York: Plenum Publishing Corp., 1977. From a practical point of view, the effect of androgen on 14. Mobbs, B. G. The Effect of Testosterone Treatment on the Uptake of 3H- Oestradiol-17/3 by Dimethylbenzanthracene-lnduced Rat Mammary Tu the estrogen receptors of tumors might ameliorate our mors. J. Endocrinol., 48: 293-294,1970. understanding of the mechanism of action of additive 15. Nicholson, R. I. Influence of Altered Lysosomal Enzyme Activities on the androgen therapy and/or suppressive adrenalectomy on Regression of DMBA-lnduced Rat Mammary Tumours. European J. Cancer, 73: 1225-1230,1977. breast cancer. A direct interaction of endogenous andro 16. Nicholson, R. l., Golder, M. P., Davies, P., and Griffiths, K. Effects of gens with ER could also be involved in the induction and/or Oestradiol-17/3 and Tamoxifen on Total and Accessible Cytoplasmic Oestradiol-17,3 Receptors in DMBA-lnduced Rat Mammary Tumours. promotion of mammary cancer since the concentrations of European J. Cancer, 72: 711-717, 1976. adrenal androgens such as A5-androstenediol both in fe 17. PasteÃ©is,J-L., Heuson, J-C., Heuson-Stiennon, J., and Legros, N. male plasma (2 to 3 nM) (18) and in human breast cancer Effects of Insulin, Prolactin, Progesterone, and Estradiol on DNA Syn thesis in Organ Culture of 7,12-Dimethylbenz(a)anthracene-induced Rat (10 to 200 ng/g tumor) (10) appear sufficient to interact Mammary Tumors. Cancer Res., 36: 2162-2170, 1976. efficiently with ER if one assumes the relative high affinity 18. Poortman, J., Vroegindewey-Jie, D., Thijssen, J. H. H., and Schwartz, F. of A5-androstenediol (Kd =4.5 nM).4 Relative Binding Affinity of Androstane and C19-nor Androstane-Ste- roids for the Estradiol-Receptor in Human Myometrial and Mammary In conclusion, a very high dose of DHT which stimulates Cancer Tissue. Mol. Cellular Endocrinol. 8: 27-34, 1977. leucine incorporation into proteins is also able to occupy 19. Rochefort, H., and Garcia, M. Androgen on the Estrogen Receptor: I-

3928 CANCER RESEARCH VOL. 38

Binding and in vivo Nuclear Translocation. Steroids, 28: 549-560, 1976. 98: 702-716, 1976. 20. Rochefort, H., Garcia, M., and Vignon, F. Regulation of Estrogen 23. Segaloff, A. The Use of Androgens in the Treatment of Neoplastic Receptor in Experimental Mammary Tumors. Rev. Endocrine Related Disease. Pharmacol. Therap. C.,2: 33-37, 1977. Cancer, in press, 1978. 24. Vignon, F., and Rochefort, H. Regulation of Estrogen Receptors in 21. Ruh, T. S., and Ruh, M. F. Androgen Induction of a Specific Uterine Ovarian-Dependent Rat Mammary Tumors. I. Effects of Castration and Protein. Endocrinology, 97: 1144-1150,1975. Prolactin. Endocrinology, 98: 722-729,1976. 22. Schmidt, W. M., Sadler, M. A., and Katzenellenbogen, B. S. Androgen- 25 Zava, D. T., and McGuire, W. L. Estrogen Receptors in Androgen Uterine Interaction: Nuclear Translocation of the Estrogen Receptor and induced Breast Tumor Regression. Cancer Res., 37: 1608-1610, 1977. Induction of the Synthesis of the Uterine-Induced Protein (IP) by High 26 Zava, D. T., and McGuire, W. L. Human Breast Cancer: Androgen Action Concentrations of Androgens in Vitro but not in Vivo. Endocrinology, Mediated by Estrogen Receptor. Science, J99: 787-788,1978.

NOVEMBER 1978 3929

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1978 American Association for Cancer Research. Androgen Effects Mediated by Estrogen Receptor in 7,12-Dimethylbenz( a)anthracene-induced Rat Mammary Tumors

Marcel Garcia and Henri Rochefort

Cancer Res 1978;38:3922-3929.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/38/11_Part_1/3922

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/38/11_Part_1/3922. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.