Gen. Pharrnac. Vol. 22, No. 3, pp, 491~,93, 1991 0306-3623/91 $3.00 + 0.00 Printed in Great Britain. All rights reserved Copyright © 1991 Pergamon Press pie

ESTRIOL BINDING IN UTERINE CORPUS CANCER AND IN NORMAL UTERINE TISSUES

KoJI hDA, ATSUSHI IMAI* and TERUHIKO TAMAYA Department of Obstetrics and Gynecology, Gifu University School of Medicine, Gifu 500, Japan

(Tel. 582-65-1241; Fax 582-65-9006)

(Received 8 August 1990)

Abstract--1. The specific bindings of estriol (E3) and estradiol-17/~ (E2) to their specific receptors were investigated in endometrial carcinoma from 7 patients and normal tissues from their respective organs or from other patients.

2. In both cytosolic and KCl-extracted fractions from them, specific binding sites for E 3 and E 2 were detected, demonstrating the presence of their separate receptors in human uterus-associated tissues.

3. In certain cases (6 cases) of well-differentiated adenoearcinoma, the ratio of concentration of E3 receptor to that of E2 receptor was almost equal to or higher than in other normal tissues.

4. These findings of unique localization of E 3 receptor distribution may offer new insight into identification of endometrial carcinoma more likely to respond to hormonal influence or therapy.

INTRODUCTION

Target tissues differ widely in their response and sensitivity to estrogens (Luine et al., 1974; Bergman et al., 1987). The effects of estrogens on the reproduc- tive tract have been demonstrated that estradiol-17fl (E2) is a full agonist, while estriol (E3) plays as a partial agonist as well as antagonist of E 2 (Hisaw et al., 1954; Huggins and Jensen, 1955; Miiller et al., 1985). The principal action of E2 on their target tissues is tissue growth following binding of the steroid hormone to its receptor (Anderson et al., 1975). E2 increases uterine size and stimulates endo- metrial proliferation. Some uterine leimyomata or endometrial carcinoma are known to be particulary sensitive to estrogen stimulation, and the former increases in size during reproductive periods or in responses to exogenous estrogens and decreases in size after menopause, although tumor growth is usually autonomous (Wilson et al., 1962; Jensen and Jacobson, 1962). On the other hand, E3 has a biphasic action on estrogenic action; the stimulation of uterine growth elicited by E2 is partially inhibited by simul- taneous administration of E3, but E 3 produces full estrogenic responses upon administration by repeti- tive injections or of long-acting derivatives (Hisaw et al., 1954; Miller, 1969; Lan and Katzenellenbogen, 1976). The rationale for agonistic and antagonistic action of E3 has been explained on the basis of E 3 competition with E2 for binding to the estrogen receptors. Also E;receptor complex dissociate and are lost from the nuclear compartment more rapidly than E2-receptor complex; therefore, repetitive injec- tions of E3 are required to keep a critical number of E3-receptor complex in the nucleus for a time suffi- ciently long to induce true uterine growth (Clark et al., 1977). An alternative explanation has recently been put forward that in a cell-free system E2 binding

*To whom all correspondence should be addressed.

to the receptor exhibits positive cooperativity and that E 3 can inhibit this cooperativity (Notides et al., 1981; Sasson and Notides, 1982). More recently, we have proposed that E3 and E2 have their own specific receptor to reveal their respective biological responses (Tamaya et al., 1990).

The ability of steroid receptor content to predict biological behavior in patients with cancer has been well established for hormonal manipulation of the steroid-dependent carcinomas. The evaluation, there- fore, of interaction of E 2 and E 3 with their receptors may give new insight to understand the regulation of endometrial carcinoma. In this communication, the E2 and E3 receptors location was examined in endometrial carcinoma and normal tissues.

MATERIALS AND METHODS

The 12 women included in this study were evaluated in the Obstetric and Gynecologic Clinic for various gynecologic diseases (Table 1). The patients ranged in age from 35 to 74 yr, with no history of exogenous hormone usage. Each of these women underwent hysterectomy.

Within several minutes from the time the uterus was removed from the patient, portions of tissue were obtained under direct visualization. Approximately 1 g each of the myometrium, endometrium (area selected to exclude the lesion) and lesion was removed from each surgical specimen, washed with Tris-EDTA-DTT buffer described below and immediately frozen in liquid nitrogen. The remaining tissue was prepared for routine histological examination. The concentration of estrogen receptors for E 3 and E 2 was measured by dextran-coated charcoal (DCC) assay tech- niques, as previously described (Tamaya et al., 1990).

Preparation of cytosol and nuclear extracts

All procedures were carried out at ice-bath temperature. The frozen tissues were excised with scissors, and homogen- ized in 4-5 ml of the buffer using a polytron homogenizer for myometrium, or a teflon homogenizer for carcinoma or endometrium. The buffer used consisted of 10mM Tris-HCl, 1.5 mM ethylendiamine tetraacetic acid (EDTA) and 0.5 mM dithiothreitol (DTT), pH 7.4. The homogenate

491

492 KoJI hDA et al.

Table 1. Estradiol receptor (Ez R) and estriol receptor (E 3 R) distributions in uterus from patients with endometrial carcinoma

Cytosol KCl-extract

Case No. (age) E 3 R E 2 R Ratio E 3 R E 2 R Ratio

1 (60) Carcinoma < 3 32 < 0.09 < 3 44 < 0.08 2(53) Carcinoma 28 39 0.71 30 <3 10< 3 (53) Carcinoma 70 43 1.93 159 < 3 53 <

Corpus 111 46 2.41 438 <3 146 < Cervix 83 27 3.07 83 < 3 28 < Vagina 112 31 3.61 235 44 5.34

4 ( 5 8 ) Carcinoma 62 74 0.84 308 176 1.75 5 ( 5 9 ) Carcinoma 44 <3 14.8< 151 54 2.80

Corpus 36 21 1.71 309 <3 103< Cervix 46 17 2.71 237 < 3 79 < Vagina 67 50 1.34 58 147 0.39

6 ( 7 0 ) Carcinoma 220 16 13.8 35 <3 11.7< 7 ( 6 0 ) Carcinoma 39 40 0.98 568 < 3 189 <

8(74) Corpus <3 310 <0.01 40 158 0.25 Cervix <3 149 <0.02 81 41 1.98 Vagina 95 150 0.63 - - - - - -

9 (45) Corpus 42 33 1.27 116 271 0.43 Cervix 63 < 3 21 < 246 114 2.16 Vagina < 3 < 3 - - 220 < 3 73.3 <

10(47) Corpus 56 78 0.71 102 157 0.64 Cervix 115 209 0.55 96 123 0.78

! 1 (58) Corpus 38 89 0.43 75 96 0.78 Cervix 204 288 0.71 157 189 0.83 Vagina 88 96 0.91 36 143 0.25

12(52) Corpus <3 123 <0.02 32 65 0.49

Cases 8, 9, 10, 11 and 12 were diagnosed as cervical cancer, cervical cancer, adenomyosis and adenomyosis, respectively. Each data was expressed as fmol/mg protein.

was centrifuged at 800g for 10min. The resulting super- natant was centrifuged at 100,000g for 1 hr to separate cytosolic fraction. For preparation of nuclear extraction, the 800 g pellet was filtrated through nylon mesh to remove cell debris. The filtrate was then centrifuged at 800g for I0 min, washed with the buffer and resuspended in 5 ml of 0.6 M KC1 prepared in the above buffer, pH 8.5. After a 30-min extraction period with frequent vortexing, the mixtures were centrifuged at 20,000g for 10rain. The supernatant was used as KCl-extracts. Protein concentration was quantitated by the method of Lowry et al. (1951) using bovine serum albumin as standard.

Assay for estradiol (Ee) receptor and estriol (E3) receptor

The affinity constants of [3H]E2 and [3 H]E3 binding in the cytosol and KCl-extract were determined according to the method of Scatchard (1949). Aliquots (0.1 ml) containing approx. 1 mg protein were incubated in triplicate with [3H]E2 or [3H]E3 in the concentration range of 0.1 5 × 10 -9 M. Parallel incubation were performed in the presence of 200-fold excess of the nonradioactive estrogens. The binding of estrogens to plasma steroid-binding globulin was eliminated by adding a 100-fold molar excess of un- labeled dihydroxytestosterone (DHT) to each tube. Incu- bations were carried out at 4°C for 16 hr, after which bound and unbound steroids were separated with DCC (Norit "A", 0.3%, Dextran T70, 0.03%, in phosphate buffer, pH 7.4), as described previously (Tamaya et al., 1990). Specifically bound [3H]E 2 or [3H]E3 was calculated by sub- tracting binding in the presence of excess of unlabeled E2 and E 3 from those in the presence of excess of unlabeled E3 and E2, respectively. Nonspecific binding did not exceed 5% of total binding.

The measurement of specific [3H]E2 or [3H]Ea binding in the cytosolic and nuclear extracts was consequently exam- ined. The samples were incubated with 8 × 10 -9 M [3H]E 2 or [3H]E3 in the absence (total binding) or presence (nonspecific binding) of 50-100-fold molar excess of unlabeled estrogens and DHT. [3H]E2 or [3H]E 3 specifically bound was obtained as above and their binding capacities were calculated from the maximal binding sites of Scatchard analysis.

Materials

[6,7JH]estradiol (60Ci/mmol) and [2,4,5,9-3H]estriol (1 I0 Ci/mmol) were purchased from Amersham. Estradiol, estriol and dihydrotestosterone were obtained from Sigma. All other chemicals were of reagent grade.

RESULTS

E 2 R or E 3 R was determined by specifically bound [3H]E2 and [3H]E3, which were evaluated by subtract- ing the binding in the presence of excess of unlabeled E2 and E 3 from those in the presence of excess of unlabeled E 3 and E2, respectively. Cytosolic fractions from endometrial carcinoma tissues contained both E2 and E 3 receptors. The concentrat ions of receptors of carcinoma from individual patients compared with normal tissues from the same organ and with tissues from other patients are presented in Table 1. The concentrat ions of E 3 R in carcinoma were observed to be almost equal to or greater than that o f E 2 R, except case No. 1. Also in KCl-extract fractions, specific binding sites for E2 and E 3 were detected. Al though the concentrat ion of E 3 receptor and the ratio of E3R to E 2 R in the KCl-extract were higher than those in cytosotic fraction, no difference between the value from carcinoma tissues and other port ions appeared to be observed. These findings were essentially confirmed in total (cytosolic plus KCl-extract) recep- tor concentrations.

DISCUSSION

In cytosolic and KCl-extract fractions from well- differentiated adenocarcinoma, specific receptors f rom E2 and E 3 were detected. In 6 cases among 7, the ratio of concentrat ion of E 3R to that o f E 2R was found to be almost equal to or higher than in other normal tissue from their respective organs.

Estriol receptor in human uterus and corpus cancer 493

E 2 and E 3 seem to have different binding sites for and effects on various tissues (Hisaw et al., 1954; Jensen and Jacobson, 1962). The efficacy of E3 in human was postulated to be selective in its activity, in particular, on the lower genital tract (Miller, 1969; Lan and Katzenellenbogen, 1976). In addition, this hormone plays a role in the etiology of breast cancer (Wotiz et al., 1978), is active as a postcoital contra- ceptive (Miiller and Wotiz, 1981), is used in the treatment of menopousal syndrome, and is the prin- cipal estrogen in human pregnancy. In spite of the unique actions of E3, it has been reported that no significant difference in receptors for E2 and E3 across tissue types exists with regards to biochemical behavior, so far. E3 has also been shown to act as an E2 antagonist. The anti-E2 action of E3 has been explained mainly on the basis of E 3 and E2-receptor interaction, e.g. the ability of E 3 to compete with E2 for receptor binding in the target cell (Clark et al., 1977; Notides et al., 1981; Sasson and Notides, 1982).

Hormonal manipulation is effective in the treat- ment of many neoplasms derived from steroid-depen- dent tissues including endometrial carcinoma or breast cancer. In addition to progesterone receptors, the uterus with endometrial carcinoma often contains appreciable amounts of estrogen receptors (Ehlich and Yound, 1981; Mutch et al,, 1987). However, the relative amount that the estrogen receptor contents of myometrium, stromal and benign glandular epithelial elements contribute to the total estrogen receptor of tissue homogenates of individual endometrial carcinoma specimens is not clear. Cytosol fractions of adenocarcinoma from certain patients contained significantly more receptor proteins for E3 than did cytosol fractions of normal tissues from their respective organs or from other patients. The specific localization of E3 receptor within uterus from patients with endometrial adenocarcinoma might suggest the estrogen receptor contribution of noncancerous com- ponents. Although biological properties of E 3 are yet in blackbox, the specific localization of E3 receptor in endometrial carcinoma, therefore, identify the cases more likely to respond to hormonal influences or therapy.

Acknowledgement--This study was supported in part by the research grant from Ministry of Education, Culture and Science, Japan (No. 1480390).

REFERENCES

Anderson J. N., Peck E. J. and Clark J. H. (1975) Estrogen- induced uterine responses and growth: relationship to receptor-estrogen binding by uterine nuclei. Endocrin- ology 96, 160-164.

Bergman M. D., Karelus K., Felicio L. S. and Nelson J. F. (1987) Tissue differences in estrogen receptor dyanmics: nuclear retention, rate of replenishment, and transient receptor loss vary in hypothalamus, pituitary, and uterus of C57 BL/6J mice. Endocrinology 121, 2065-2074.

Clark J. H., Paszko Z. and Peck E. J. Jr (1977) Nuclear binding and the retention of the estrogen complex: re- lation to the agonist and antagonistic properties of estriol. Endocrinology 100, 91-98.

Ehlich C. E. and Young P. C. M. (1981) Cytoplasmic progesterone and estradiol receptors in normal, hyper- plastic and carcinomatous endometria: therapeutic impli- cations. Am. J. Obstet. Gynec. 141, 539-546.

Hisaw F. L., Velardo J. T. and Goolsby C. M. (1954) Interaction of estrogens on uterine growth. J. clin. Endoc. Metab. 14, 1134-1140.

Huggins C. and Jensen E. V. (1955) The depression of oestrone-induced uterine growth by phenolic oestrogens. J. exp. Med. 102, 335-340.

Jensen E. V. and Jacobson H. I. (1962) Basic guides to the mechanism of estrogen action. Recent Prog. Horm. Res. 18, 387-412.

Lan N. C, and Katzenellenbogen B. S. (1976) Temporal relationship between hormone receptor binding and bio- logical responses in the uterus: studies with short- and long-acting derivatives of estriol. Endocrinology 98, 220-227.

Lowry O. H., Rosebrough N. J., Farr A. L. and Randall R. J. (1951) Protein measurement with the Folin phenol reagent. J. biol. Chem. 193, 265-271.

Luine V. N., Khyle, hevskaya R. I. and McEwen B. S. (1974) Oestrogen effects on brain and pituitary enzyme activities. J. Neurochem. 23, 925-932.

Miller B. G. (1969) The relative potentices of oestriol, oestradiol and ~strone on the uterus and vagina of the mouse. J. Endocr. 43, 563-572.

Miiller R. E. and Wotiz H. H. (1981) Post-coital contracep- tive activity and estrogen receptor binding affinity of phenolic steroids. Endocrinology 96, 160-166.

Miiller R. E., Traish A. M. and Wotiz H. H. (1985) Interaction of estradiol and estriol with uterine estrogen receptor in vivo and in excised uteri or cell suspensions at 37°C: noncooperative estradiol binding and absence of estriol inhibition of estradiol-induced receptor activation and transformation. Endocrinology 117, 1839-1847.

Mutch D. C., Soper J. T., Budwit-Novotny D. A., Cox E. B., Creasman W. T., McCarty K. S. and McCarty K. S. Jr (1987) Endometrial adenocarcinoma estrogen receptor content: association of clinicopathologic features with immunohistochemical analysis compared with standard biochemical methods. Am. J. Obstet. Gynec. 157, 924-931.

Notides A. C., Lerner N. and Hamilton D. E. (1981) Positive cooperativity of the estrogen receptor. Proc. natn. Acad. Sci. U.S.A. 78, 4926-2930.

Sasson S. and Notides A. D. (1982) Estriol and estrone interaction with the estrogen receptor II: estriol and estrone-induced inhibition of the cooperative binding of [3H] estradiol to the estrogen receptor. J. biol. Chem. 258, 818-823.

Scatchard G. (1949) The attraction of protein for small molecules and ions. Ann. N.Y. Acad. Sci. 51, 660-667.

Tamaya T., Kawabata I., Iida K. and Imai A. (1990) Effects of estradiol-17fl and estriol on their binding sites in the rabbit uterus. Comp. Biochem. Biophys. 95B, 415-418.

Wilson E. A., Yang F, and Rees E. D. (1962) Estradiol and progesterone binding in uterine leiomyomata and in normal uterine tissues. Obstet. Gynec. 55, 20-24.

Wotiz H. H., Chattoraj S. C., Kudisch M. and Miiller R. E. (1978) Impending estrogens and the etiology of breast cancer. Cancer Res, 38, 4012-4018.