Contrasting effects of estradiol-17β and 17α-ethinyl estradiol-17β on cultured whole embryos

J. steroid Biochem. Vol. 29, NO. 6, pp. 629-634 Printed in Great Britain. All rights reserved

0022-4731/88 $3.00 + 0.00

1988

Copyright 0 1988Pergamon Press plc

CONTRASTING EFFECTS OF ESTRADIOL-178 AND 17a -ETHINYL ESTRADIOL- 178 ON CULTURED WHOLE EMBRYOS BRUCE K. BEYER and MONT R. JUCHAU* Department of Pharmacology, School of Medicine, University of Washington, Seattle, WA 98195, U.S.A. (Received

I May 1987)

Summary-Estradiol178 (Er) and 17a -ethinyl estradiol- 178 (EE) were compared in terms of their relative capacities to alter growth and developmental patterns of cultured whole embryos during the early stages of organogenesis. Embryos exhibited a notable differential susceptibility to the embryotoxic effects of parents Er vs EE when these estrogens were added directly to the media at the onset of the culture period. At initial concentrations of 0.1 mM, Er failed to produce statistically significant effects whereas EE elicited marked embryotoxicity. Inclusion of a P-450-dependent biotransformation system in the culture media resulted in a significant attenuation of the embryotoxic effects of parent & vs EE when these estrogens were added directly to the media at the onset of the culture period. At initial concentrations of 0.1 mM, l$ failed to produce statistically embryotoxicity by hepatic S9. The divergent results produced by the two steroids could not be attributed to differences in rates of catecholestrogen generation in the culture medium or by the conceptuses. The results demonstrate definitive dissimilarities between the effects of two steroidal estrogens on developmental parameters and document marked differences in the effects of biotransformation on their embryotoxic potential. The data strongly suggest that the embryotoxicity of these steroids is not mediated via interactions with estrogen receptors. Additionally, the data show that the differential capacity of these two steroids to produce embryotoxic effects is diametrically opposite to earlier reported patterns of their carcinogenic potential in the Syrian hamster kidney.

INTRODUCI’ION Estrogenic chemicals have exhibited the capacity to elicit a large number of serious and potentially serious toxic effects on biological systems. Such effects

include carcinogenesis (renal, genital, mammary, hepatocellular, pituitary, cervical, testicular, ovarian, uterine, osseous) [l-3], transformation of cultured fibroblasts and Syrian hamster embryo cells [4], numerical chromosomal changes (aneuploidy) [5], disruption of microtubule organization [6], unscheduled DNA synthesis [7], sister chromatid exchange [8], chromosomal aberrations (91, gene mutations [lo], carcinogenesis [l 1] and dystransplacental morphogenesis of the genitourinary tract [12]. It has been reasonably well established that at least some of these effects result from the conversion of estrogens to reactive intermediary metabolites [l-12]. Steroidal as well as nonsteroidal estrogens are subject to cytochrome P-450-dependent conversion to catechol derivatives which may subsequently be converted to reactive 0-semiquinones and quinones. Diethylstilbestrol and certain other nonsteroidal estrogens can be converted directly to reactive p-semiquinones via peroxidative mechanisms. These and/or other *To whom correspondence

should be addressed.

intermediary metabolites may be responsible for or contribute significantly to the observed toxic effects. In order to more fully understand the potential of estrogen chemicals to alter normal morphogenesis, we have investigated the role of biotransformational factors as modulators of the embryotoxic properties of two structurally similar steroidal estrogens, estradiol- 178 (Er ) and 17a -ethinyl estradiol- 178 (EE). The former is the most potent naturallyoccurring estrogen in mammals and the latter is the most heavily utilized steroidal estrogen in medical practice, differing in structure from Er only by the presence of a 17a -ethinyl group. To avoid the complications of superimposed maternal biotransformation, we have chosen the whole embryo culture system to investigate these parameters. Utilization of this system also permits an assessment of embryotoxic potential during early organogenesis, a stage of development widely considered to represent the period most sensitive to chemical/environmental insult. The results obtained indicate profound differences not only in the direct effects of these two steroids, but also in the capacity of biotransforming systems to modufare their embryotoxicity. The evidence obtained points to subtle differences in routes of biotransformation as significant factors in the chemical dysmorphogenic effects of steroidal estrogens.

629

BRUCE K. BEYER and MONT R. JUCHAU

630 EXPERIMENTAL

Materials EE was purchased from Steraloids, Inc. Wilton, NH and exhibited a melting point of 183-l 84°C. E,, NADPH, catechol-O-methyltransferase, S-adenosylmethionine and glucose 6-phosphate were obtained from Sigma Chemical Co., St Louis, MI. [3H]Sadenosylmethionine (99% pure, 10.5 Ci/nmol) was obtained from New England Nuclear Corp., Boston, MA. Hepatic postmitochondrial supernatant (S9) fractions utilized in routine experiments were prepared by homogenizing the livers of adult (20&250 g) male Sprague-Dawley rats that were previously given a single intraperitoneal injection of a mixture of polychlorinated biphenyls (Aroclor 1254, 500 mg/kg 5 days prior to euthanasia) dissolved in corn oil. Fresh livers were homogenized in 2 vol of 1.15% cold KC1 solution and homogenates were centrifuged at 9000gfor 20 min. The supernatant fractions were utilized as enzyme sources in the reported experiments. Carbon monoxide (99.4% pure) was purchased from the Matheson Co., St Louis, MI. Embryo culture system The postimplantation culture system originally developed by New [13] was utilized as described in detail by Faustman-Watts et al. [14]. Conceptuses were explanted on day 10 (10 f 2 somites) and were exposed to varying concentrations of E, or EE added directly to the culture medium at the beginning of the 24 h culture period. Estrogens were dissolved in dimethyl sulfoxide (DMSO) and were added such that final DMSO concentrations did not exceed 25 mM, previously shown [15] to have no detectable effects on cultured whole embryos. Cultures were gassed with either 02: N,: CO2 (20: 75 : 5, by vol) or O,:CO:N,:CO, (20:20:45:5, by vol) for the initial 20 h of culturing after which the mixture was changed to 02:N2:COZ (75:20:5, by vol) or O,:CO:CO, (75 : 20: 5, by vol). It was previously [15] demonstrated that gassing with CO produces no detectable effects on cultured whole embryos. After the 24 h culture period, embryos were removed from culture flasks and evaluated, without knowledge of treatment, for several indices of embryotoxicity. Evaluation of embryos Microscopic examinations of cultured embryos were utilized to evaluate viability, axial rotation (flexure), neural tube closure, maximal embryonic length, somite numbers, prosencephalic index, limb bud size, and other microscopically detectable indices of normal development. Conceptuses were scored as viable only if circulation in the vitelline yolk sac and active heartbeat were both detectable. Only viable embryos were further evaluated. Complete axial rotation (ventral) was scored as 0, partial rotation was scored as 1.0 and embryos exhibiting dorsal flexure were scored at 2.0. Neural tube closure was scored

strictly on a quanta1 basis as either normally closed or abnormally open. Maximal embryonic length, maximal distance (mm) from the tip of the prosencephalon to the center of the optic field (procencephalic index) and limb bud size were measured with a calibrated optical micrometer attached to the dissecting microscope. Following evaluations by light microscopy, embryos were sonicated and assessed for content of protein and DNA by methods described by Bradford [16] and Labarca and Paigen [ 171 respectively. Representative embryos were photographed and fixed in glutaraldehyde (2.5% in 0.1 M sodium phosphate buffer, pH 7.4) for histological examinations. They were then dehydrated, embedded in hydroxyethyl methacrylate, sectioned with glass knives at 2-3 PM, and stained with toluidine blue. Assays of monooxygenase activity Conversion of estrogens to corresponding catechol metabolites was assessed as described by Hoffman et al. [ 181. Monooxygenase activities were measured under conditions described previously [19] as well as under conditions existing in the culture system. Reactions were linear with respect to time and protein concentration under the reaction conditions utilized. For certain experiments, tritium displacement analyses of estrogen hydroxylase activity as described by Namkung et al. [20] were employed to provide corroborative data for the routinely utilized assay. Statistics Student’s t-test was used to analyze differences in parameters measured on a continuous (graded) scale. Included were embryonic protein and DNA, measurements of length, limb bud size, prosencephalic index, somite numbers, and enzymatic activities. The x2 test was utilized for analyses of differences in parameters measured on a quanta1 scale (viability, abnormally open neural tubes, malformation incidence). Both the t-test and Wilcoxon’s Signed-Rank test were employed for analyses of differences in axial rotation. Statistical procedures are described by Steele and Torrie [21]. RESULTS

Initial experiments were designed to ascertain the concentrations of steroidal estrogens required to elicit morphologic abnormalities detectable in cultured embryos with light microscopy and/or biochemical analyses. Range-finding initial concentrations extended from 1.OnM to 10 mM. No gross morphologic effects could be observed at concentrations between 1.0 nM and 10 pM although EE produced significant (P < 0.05) decreases in protein content at the lowest concentrations. At 93 PM, EE produced profound morphologic changes on a variety of developmental parameters (Tables 1 and 2) whereas effects elicited by E, were statistically insignificant at the chosen probability level (P < 0.05). All steroidal concen-

Embryotoxicity

of steroidal

estrogens

631

Table 1. Growth and developmental parameters measured in embryos unexposed to estrogens Parameter* % Viable % Abnormalities flexure neural tube prosencephalon other Somite number Embryo length (mm) Limb index Prosencephalic index Protein @g/embryo) DNA (ue/embrvo)

None (Ill)t

+ s9t (22)

91.3 6.1 5.4 0.9 3.6 3.9 21.00~0.17 3.02 + 0.03 2.30 k 0.1 0.89 k 0.01 212.00 + 5.8 16.00 + 0.4

100.0

Variation in culture conditions + s9/coft (8 1) 98.7 6.2 I .2 0.7 6.2 4.9 20.70 + I .6 2.98 +_0.2 2.30 f 0.5 0.87 * 0.13 199.00 + 62.0 13.80 f 3.6

9.1 4.6 0.0 9.1 0.0 21.30* 1.3 3.06 +_0.2 2.40 f 0.5 0.91 * 0.10 219.00 +_40.0 16.40 f 4.7

+ Cof (28)

99.3 3.6 3.6 0.0 3.6 3.6 21.30 f 1.4 3.02 + 0.2 2.40 + 0.5 0.84 f 0.03 218.00 f 52.0 17.00 + 3.0

CO (25) 96.1 8.0 8.0 4.0 4.0 8.0 20.70 f I .8 2.94 f 0.3 2.20 f 0.4 0.85 f 0.12 205.00 f 49.0 17.60 + 3.9

*Incidences of viability and flexure, neural tube, prosencephalic and other abnormalities are provided as quanta1 (all-or-none) data. All other indices of embryotoxicity are provided as quantitative measurements (see Experimental) on a continuous (graded) scale. These data represent the accumulation of all control values (*SD) acquired during the course of these investigations. tNumbers in parentheses are the numbers of embryos in each category. S9 is the hepatic postmitochondrial (9OoOg 20 min) supernatant fraction; 16 5 3 mg protein/culture flask; Cof refers to NADPH (0.5 mM) and glucose 6-phosphate (5.0mM) added as final concentrations to the culture medium at the beginning of the 24 h culture period.

of 1.OmM were 100% embryolethal a biotransforming system was added to the culture media. Analyses of control data indicated that hepatic S9, NADPH (0.5 mM), glucose 6-phosphate (5.0 mM) or carbon monoxide each failed to produce detectable embryotoxicity under the conditions employed in these experiments (Table 1). Supplementation of the culture medium with a hepatic, P-450-dependent monooxygenation system consisting of S9, NADPH and glucose 6-phosphate resulted in marked attenuation of EE-elicited embryotoxicity (Table 2) but produced a profound exacerbation of the effects of Ez. Each of these effects were observable in a concentration-dependent manner within the range of 18-130 PM (Fig. 1). Partial replacement of nitrogen gas with carbon monoxide (see Experimental) in the head space of the culture bottles resulted in a diminished bioactivation of E, by the added biotrations

in excess

irrespective

of whether

transforming systems but did not appear capable of altering the attenuating effect of the biotransforming system on EE-elicited embryotoxicity (Fig. 2). Because catecholestrogen formation has been heavily implicated in the toxic effects of estrogens, we compared EE with E, in terms of their conversion to catecholestrogens by enzyme sources present in the culture medium (Table 3) as well as with additional enzyme sources that were analyzed for comparative purposes. The results indicated that both adult hepatic tissues as well as embryonic tissues would convert EE and E, to the corresponding catecholestrogens and that the two steroids were converted at comparable rates in the respective tissues studied. In spite of profound differences in the capacity to produce embryonic abnormalities, the nature of the defects elicited by EE and E, was remarkably similar. Each steroid was capable of producing marked

Table 2. Effects of estradiol-17/I and I’la-ethinyl estradiol-17b on cultured whole rat embryos: Effects of an added biotransforming system*

Parametert

Addition to culture medium 17a-Ethinyl estradiol-l’lfi (95 PM) Estradiol-17fl (95 PM) -SB/cof (26) +S9/cof (51) -S9/cof (34) +S9/cof (52)

% Viable % Malformed flexure neural tube prosencephalon other Somite number Embryo length (mm) Limb index Prosencephalic index Protein @g/embryo) DNA @g/embryo)

69.2 lOO.O$ lOO.O$ 33.3: lOO.O$ 88.9$ 15.4 f. 0.5 2.3 k 0.1 0.6 f 0.2: 0.58 f 0.04$ 106+25f 7.1 + 1.8:

100.0 39.2 35.2 0.0 62.8 19.6 19.0 f 0.7 2.7 + 0.08 1.6 + 0.4 0.8 +_O.l 154*31 10.2 f 2.5

94. I 9.4 6.3 0.0 3.1 9.4 20.8 f 0.4 2.9 + 0.06f 2.0 f 0.1x 0.82 k 0.03$ 190+34 14.8 * 0.7f

84.6 1oo.q 65.9 11.4 97.75 lOO.@ 15.0 f 0.4$ 2.42 f 0.04 0.8 + 0.1g 0.48 f 0.04$ 91.1 f 27.w 4.6 f OS5

*The added biotransforming system (S9jcofJ consisted of a hepatic postmitochondrial (9OOOg 20 min) supernatant fraction from PCB-induced rats (see Experimental) NADPH (0.5 mM) and glucose 6-phosphate (5.0mM) added as final concentrations to the culture medium at the beginning of the 24-h culture period. tSee footnote *, Table 1. @ignificantly different (P < 0.05) from corresponding values appearing in columns 2 and 3 of this table and from all columns of Table I. ISignificantly different (P < 0.05) from corresponding values appearing in column 3 of this table and from all columns of Table I.

BRUCE

K. BEYER and MONT R. JUCHAIJ

0

n n

CONTROL + s-S/COf + co + S-S/Co‘ +co

.6

l

.3.

E2

0 E2+S-9Kof .

EE

0 EE

.2

+S-9Kof

-

Oo

Ethyl

Ertradiol-17 13 K35p.4)

.I 20

40

60

80

Estrogen Concentration

00

120

140

(,uM)

Fig. 1.Concentration-dependent effects of estradiol- 178 and 17a-ethinyl estradiol-17p on the prosencephalic index (see Experimental) of cultured whole rat embryos. Effects were measured in both the presence (open symbols) and absence (closed symbols) of an added hepatic biotransforming system as described in Experimental. Qualitatively similar effects were observed with other indices of embryotoxicity (see Tables 1 and 2).

decreases in protein and DNA content (Table 2), suggestive of a more generalized growth deficit. However, abnormalities of the neural tube and its closure were notably infrequent relative to other morphologic changes, indicating specificity in the causation of embryonic defects. The most consistent effect observed was on the development of the prosencephalon (prosencephalic hypoplasia). Thus, the prosencephalic index was utilized as the most reliable marker of estrogen-induced embryotoxicity in these experiments (Figs 1 and 2). In addition to the prosencephalic hypoplasia observable with the dissecting microscope, histologic examinations indicated that both steroids produced regions of cellular necrosis in the prosencephalon. Detailed morphologic analyses of the observed defects will be reported in a separate publication. DISCUSSION

Catecholestrogen metabolic pathway

formation represents a major for steroidal estrogens [22,23]

IEl rodiol-17 D l95pt4)

Fig. 2. Effects of substitution of N, with CO in the head space gas (see Experimental) on the capacity of an added biotransforming system to modulate the embryotoxic effects of steroidal estrogens. The prosencephalic index was utilized as the representative measure of toxicity; qualitatively similar effects were observed with other indices (see Tables 1 and 2).

and has also been heavily implicated in the capacity of estrogens to elicit a variety of toxic effects including tumorigenesis, mutagenesis and other manifestations of genotoxicity including unscheduled DNA synthesis, sister chromatid exchange, aneuploidy, etc. [24,25]. The investigations reported in this paper demonstrate that at least two steroidal estrogens, Ez and EE, also are capable of profoundly altering normal embryonic development during the early stages of organogenesis. However, in view of the comparable rates of conversion of E, vs EE to catecholestrogens in a variety of metabolic systems (including embryonic homogenates), it was very surprising to observe marked differences between these two steroids-not only in embryotoxic potential per se-but also in the modulating effects on embryonic potential via biotransformation. Whereas supplementation of an hepatic biotransforming system markedly reduced the embryotoxicity of EE, addition of the same biotransforming system under the same conditions produced major increases in the embryotoxicity of E,. The only structural difference between the two steroids is the presence of the 17cr-ethinyl group on EE, making it evident that this group can play a crucial role as a determinant of embryotoxic potential. The presented data would

Table 3. Mean estrogen 2/4 hydroxylase activities* with l’la-ethinyl estradiol-l7a

and estradiol-178 as substrates Substrate

Enzyme source

Male rat hepatic S9 Aroclor 1254-induced male rat hepatic S9 Pregnant female rat hepatic S9 Embryo whole homogenate

17a-Ethinyl estradiol-I’ll 656 f 549 f 736 f 0.051 +

186 (5) 121(7) I34 (7) 0.021 (3 pools)t

Estradiol- I78 742k 723 + 237 + 0.043 f

112(5) 94 (5) 40 (7) 0.017 (3 pools)

*Activities are given as nmol/mg/min f standard deviations. Numbers in parentheses are the number of separate enzyme sources assayed. tA pool of embryonic homogenate was prepared by sonicating 12-15 embryos for each pool.

Embryotoxi~ty of steroidal estrogens tend to infer that the ethinyl group was not only responsible for a much greater inherent embryotoxic potency of the parent compound (EE) but that biotransformation of the same group led to a marked diminution in its embryotoxic potential. However, the greater embryotoxicity of bioactivated E, than of parent EE at initial equimolar estrogen concentrations is difficult to explain at present and will probably require considerable research effort for clarification. Nevertheless, it seems clear that biotransformational factors play a major role in the embryotoxic potential of steroidal estrogens, and the results appeal for further detailed investigations of the bioactivation as well as bioinactivation of these chemicals and the factors regulating such processes. The observation that in spite of the marked differences noted above, the effects of bioactivated Ez and parent EE on cultured embryos were qualitatively very similar was also of considerable interest and suggested a common mechanism or common final pathway leading to the observed malformations. It has long been known that estrogens play a vital role in the growth and differentiation of the uterus 126,271 and that estrogen receptors are of prime improtance in mediating various changes affected by estrogens at the cellular level. It seemed possible that estrogen receptors could be responsible for the common effects of EE and E, by directing them (or respective active metabolites) to common critical sites in embryonic tissues. In this regard, it is noteworthy that estrogen receptors have been detected in prenatal tissues at relatively early stages of gestation (281. However, in these experiments, the likelihood that either E2 or EE caused embryotoxicity via interaction with estrogen receptors appears very remote in view of the high, unphysiologic concentrations required to elicit the observed effects. The nature of the elicited effects appeared to result from a nonspecific cytotoxicity, also arguing against receptor-mediated effects. It should also be borne in mind that considerable evidence has accumulated in recent years to show that estrogens can produce a number of important biological effects that are independent of estrogen receptor binding [29,30]. Of particular interest in this regard are the comparative effects of EE vs E2 on renal tumorigenesis reported by Li and Li [31] in which EE, as potent an estrogen in the hamster as either Ez or diethylstilbestrol, induced only a very low tumor incidence in compa~~n to the high incidence produced by the latter two estrogens. These investigators su~essfully demonstrate a clear separation of estrogenic vs carcinogenic effects and sug gested that while ho~onal activity could be necessary for tumorigenic activity, it did not appear to be suflicient, at least in the model investigated. Their results, coupled with our observations, make it clear that biotransformational determinants of estrogenic carcinogenicity cannot be directly extrapolated to estrogenic embryotoxicity. From this perspective, it was of interest that embryos at days 10

633

and 11 of gestation exhibited the capacity to convert (albeit at extremely low rates) both E2 and EE to ~t~hol~trogens. Maydl et al. [32] have shown that the fetal genital tract will convert diethylstilbestrol to ZJ-dienestrol and that intermediates in the conversion may be important in the causation of genital tract defects. Such observations point to the necessity for a detailed investigation of the capacities of prenatal target tissues in terms of their biotransformation of embryotoxins. The resultant knowledge is of great potential value in ensuring protection against the inappropriate use of estrogenic chemicals. Acknowledgements-The authors wish to thank Elizabeth Walker and Julie Pascoe for technical assistance and Joan Russell for manuscript preparation. Research was supported by NIH grants ES-04041 and ES-04342. REFERENCES 1. Liehr J. G., Fang W. F., Sirbasku D. A. and Ulubelen A. A.: Carcinogenicity of catecholestrogens in Syrian hamsters. J. steroid Biochem. 24 (1986) 353-357. 2. Liehr J. G., Avitts T. A., Randerath E. and Randerath K.: Estrogen-induced endogenous DNA adduction: possible mechanisms of hormonal cancer. Proc. nazn. Acad. Sci. U.S.A. 83 (1986) 5301-5305. 3. Adams S. P. and Notides A. C.: Metabolism and irreversible binding of diethylstilbestrol in the kidney of the Syrian golden hamster. Biochem. Pharmac. 35 (1986) 2171-2178. 4. Tsutsui T., Suzuki N., Maizumi H., McLachlan J. A. and Barrett J. C.: Alteration in diethylstilbestrolinduced mutagenicity and cell transformation by exogenous metabolic activation. Curcinogenesis 7 (1986) 1415-1418. 5. Tsutsui T., Maizumi H., McLachlan J. A. and Barrett J. C.: Aneuploidy induction and cell transformation by diethylstil~trol: A possible chromosome mechanism in carcinogenesis. Cancer Res. 43 (1983) 38143821. 6. Tucker R. W. and Barrett J. C.: Decreased numbers of spindle and cytoplasmic microtubules in hamster embryo cells treated with a carcinogen, diethylstilbcstrol. Cancer Eles. 46 (1986) 2088-2095. I . Martin C. N., McDermid A. C. and Garner R. C.: Testing of km&n carcinogens and noncarcinogens for their ability to induce unscheduled DNA synrhesis in HeLa cells. Cancer Rex 38 11978) 2621-2627. 8. Hill A. and Wolff S.: Increased induction of sister chromatid exchange by diethylstilbestrol in lymphocytes from pregnant and premenopausal women. Cancer Res. 42 (1982) 893-896. 9. Bishun N., Forster S., Valera N. and Williams D. S.: The clastogenic effects of diethylstil~trol on ascitic tumor cells in v&o. Microbios Lett. 213 (1980) 27-31. 10. Clive D., Johnson K. O., Spector J. F. S., Batson A. G. and Brown M. M. M.: Validation and characterization of the L5178Y TK + l-mouse lymphoma mutagen assay system. h&far. Res. 59 (1979) 61-108. II. Herbst A. L., Ulfelder H. and Poskanzer D.: Adenocarcinoma of the vagina: Association of maternal stilbestrol therapy with tumor appearances in young women. N. Engl. J. Med. 284 (1971) 878-881. 12. Metzler M. and McLachlan J. A.: Oxidative metabolism of diethylstilbestrol and steroidal estrogens as a potential factor in their fetotoxicity. In Role of Phurmacokinetics in PrenataI mid ~erjnataf Toxicology (Edited by D. Neubert, H. J. Merker, H. Nau and J. Langman). Georg Thieme Verlag, Stuttgart, Federal Republic of Germany (1978) p. 157.

BRUCE K. BEYERand MONT R. JUCHAU

634

13. New D. A. T.: Whole embryo culture and the study of mammalian embryos during organogenesis. Eiol. Rev. 53 (1978) 81-122. 14. Faustman-Watts E. M., Greenaway J. C., Namkung M. J., Fantel A. G. and Juchau M. R.: Teratogenicity in vitro of 2-acetylaminofluorene. Teratology 27 (1983) 19-28. 15. Greenaway J. C., Fantel A. G., Shepard T. H. and Juchau M. R.: The in vitro teratogenicity of cyclophosphamide in rat embryos. Terufology 25 (1982) 3355343. 16. Bradford M. M.: A rapid, sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein dye binding. Analyt. Biothem. 72 (1976) 248-254.

17. Labarca C. and Paigen K.: A simple, rapid and sensitive DNA assay procedure. Analyt Biochem. 102 (1980) 344352. 18. Hoffman A. R., Paul S. M. and Axelrod J.: Estrogen 2-hydroxylase in the rat: Distribution and response to hormonal manipulation. Biochem. Pharmac. 29 (1980) 83-87.

19. Chao S. T., Omiecinski C. J., Namkung M. J., Nelson S., D., Dvorchik B. H. and Juchau M. R.: Catechol estrogen formation in placental and fetal tissues of humans, macaques, rats and rabbits. Dev. Pharmac. Ther. 2 (1981) i-17. 20. Namkune M. J.. Porubek D., Nelson S. D. and Juchau M. R.: &gulation of aromatic oxidation of estradiol17p in maternal hepatic, fetal hepatic and placental tissues: Comparative effects of a series of inducing agents. J. steroid Biochem. 22 (1985) 563-567. 71 &I. Steel R. G. D. and Torries J. H.: Principles and Procedures of Statistics. McGraw-Hill, New York (1960) pp. 67-88, 352-366. 22. Bolt H. M.: In Pharmacology of Estrogens (Edited by R. R. Chaudhury). Pergamon Press, Oxford (1981) pp. 1948.

23. Fishman J.: In Catechol Estrogens (Edited by G. R.

Merriam and M. B. Lipsett). Raven Press, New York (1983) pp. 1-3. 24. Jellinck P. H., Hahn E. F. and Fishman J.: Absence of reactive intermediates in the formation of catecho) .__._. estrogens by rat liver microsomes. J. biol. Chem. 261 (1986) 7729-7732. 25. LeQuesne P. W., Abdel-Baky

S., Durga A. V. and Purdy R. H.: Active nonaromatic intermediates in the conversion of steroidal estrogens into catechol estrogens. Biochemistry 25 (1986) 2065-2073. 26. Clark J. H., Peck E. J. Jr, Hardin J. W. and Ericksoson H.: The biology and pharmacology of estrogen receptor binding: Relationship to uterine growth. In Recenfnrr and Hirmone Action-(Edited by BTW. O’Malley aid-L. Birnbaumer). Academic Press, New York, Vol. 2 (1978). 27. Katzenellenbogen B. S., Bhakoo H. S., Lan N. C., Tatee T., Tsai T. L. and Katzenellenbogen J. A.: Estrogen and antiestrogen action in reproductive tissues and tumors, Recent Prog. Horm. Res. 35 (1979) 259-283. 28. Holdregger C. and Keefer D.: The ontogeny of the mouse estrogen receptor: the pelvic region. Am. J. Anal. 177 (1986) 285-297. 29. Liehr J. G., Stance1 G. M., Chorich L. P., Bousfield

G. R. and Ulubelen A. A.: Hormonal carginogenesis: Separation of estrogenicity from carcinogenicity. Chem. Biol. Interacf. 59 (1986) 173-184. 30. Epe B., Schiffmann D. and Metzler M.: Possible role of

oxygen radicals in cell transformation by diethylstilbestrol and related compounds. Curcinogenesis 7 (1986) 132991334. 31. Li J. J. and Li S. A.: Estrogen-induced tumorigenesis in hamsters: Roles for hormonal and carcinogenic activities. Arch. Tox. 55 (1984) 110-118. 32. Maydl R., Newbold R. R., Metzler M. and McLachlan J. A.: Diethylstilbestrol metabolism by the fetal genital tract. Endocrinology 113 (1983) 14fS151.