The E-SCREEN Assay as a Tool to Identify : An Update on Estrogenic Environmental Pollutants Ana M. Soto,' Carlos Sonnenschein,1 Kerrie L. Chung,' Mariana F. Fernandez,1 Nicolas Olea,2 and Fatima Olea Serrano3 1Tufts University School of Medicine, Department of Anatomy and Cellular Biology, Boston, Massachusetts; 2School of Medicine, University of Granada, Granada, Spain; 3School of Pharmacy, University of Granada, Granada, Spain Estrogens are defined by their ability to induce the proliferation of cells of the female genital tract. The wide chemical diversity of estrogenic compounds precludes an accurate prediction of estrogenic activity on the basis of chemical structure. Rodent bioassays are not suited for the large-scale screening of chemicals before their release into the environment because of their cost, complexity, and ethical concerns. The E-SCREEN assay was developed to assess the estrogenicity of environmental chemicals using the proliferative effect of estrogens on their target cells as an end point. This quantitative assay compares the cell number achieved by similar inocula of MCF-7 cells in the absence of estrogens (negative control) and in the presence of 1 7,B- (positive control) and a range of concentrations of chemicals suspected to be estrogenic. Among the compounds tested, several "new" estrogens were found; alkylphenols, , some PCB congeners and hydroxylated PCBs, and the insecticides , , and toxaphene were estrogenic by the E-SCREEN assay. In addition, these compounds competed with estradiol for binding to the receptor and increased the levels of progesterone receptor and pS2 in MCF-7 cells, as expected from estrogen mimics. Recombinant human growth factors (bFGF, EGF, IGF-1) and insulin did not increase cell yields. The aims of the work summarized in this paper were a) to validate the E-SCREEN assay; b) to screen a variety of chemicals present in the environment to identify those that may be causing reproductive effects in wildlife and humans; c) to assess whether environmental estrogens may act cumulatively; and finally d) to discuss the reliability of this and other assays to screen chemicals for their estrogenicity before they are released into the environment. - Environ Health Perspect 1 03(Suppl 7):113-122 (1995) Key words: xenobiotics, cell proliferation, endosulfan, phthalates, antioxidants, alkylphenols

Introduction of natural hormones; and d) modifying This diversity makes it difficult to predict For the last 40 years, substantial evidence hormone receptor levels. the estrogenicity of xenobiotics solely on has surfaced on the hormonelike effects of Among environmental chemicals found structural bases. To overcome this short- many xenobiotics in , wildlife, and to cause reproductive impairment in coming, their identification as estrogens has humans (1). The endocrine and reproduc- wildlife and humans there are estrogen relied on rodent bioassays. These assays tive effects of xenobiotics are believed to be mimics (). Natural estrogens measure either vaginal cornification or the due to their a) mimicking effects of promote cell proliferation and hypertrophy increase in uterine wet weight; however, the endogenous hormones such as estrogens of female secondary sex organs and induce latter is not a specific estrogen response (5). and androgens; b) antagonizing the effects the synthesis of cell type-specific proteins To obviate problems inherent to animal of normal, endogenous hormones; c) alter- (2). Xenobiotics of widely diverse chemical testing, quantitative bioassays using cells in ing the pattern ofsynthesis and metabolism structure have estrogenic properties (3,4). culture have been developed. For example, the induction of prolactin in primary sheep pituitary cell culture has been proposed as a This paper was presented at the Symposium on Estrogens in the Environment, III: Global Health Implications measure of estrogen action (6); in this held 9-11 January 1994 in Washington, DC. Manuscript received: March 15, 1995; manuscript accepted: model, estrogens induce protein synthesis April 4, 1995. This work was partially supported by grants from the W. Alton Jones Foundation, EPA-CR 820301, NIH-CA- but are ineffective at inducing cell prolifera- 13410, and NSF-DCB-9105594. The assistance of The Center for Reproductive Research at Tufts University tion. The limitations of this assay are that (P30 HD 28897) is gratefully acknowledged. some estrogen-inducible genes could also be Address correspondence to Dr. Ana M. Soto, Tufts University School of Medicine, Department of Anatomy and Cellular Biology, 136 Harrison Avenue, Boston, MA 02111. Telephone: (617) 636-6954. Fax: induced by nonestrogenic substances. For (617) 636-6536. E-mail: [email protected] example, prolactin synthesis may be Abbreviations used: PCB, polychlorinated biphenyls; bFGF, basic fibroblast growth factor; EGF, epidermal induced by EGF, thyrotropin releasing fac- growth factor; IGF-1, insulinlike growth factor 1; DME, Dulbecco's modification of Eagle's medium; FBS, fetal tor, and phorbol esters (7). Another estro- bovine serum; E2, 17f-estradiol; DDT, dichlorodiphenyltrichloroethane; DDD, dichlorodiphenyidichloroethane; DDE, dichlorodiphenyidichloroethylene; NIEHS, National Institute of Environmental Health Sciences; DES, gen-inducible marker, ovalbumin synthesis, ; DMSO, dimethyl sulfoxide; CD, charcoaldextran; CDHuS, CD human serum in phenol red- is stimulated by other such as prog- free DME; td, doubling time; PE, proliferative effect; RPP, relative proliferative potency; RPE, relative prolifera- esterone and tive effect; PBS, phosphate-buffered saline; BHT, butylated hydroxytoluene; BHA, t-butylhydroxyanisole; PR, glucocorticoids (8). Also, progesterone receptor; IC50, concentration that inhibits 50%; APEs, alkylphenol polyethoxylates; DCB, dichloro- induction of reporter genes under control biphenyl; RBA, relative binding affinity; TCB, tri- or tetrachlorobiphenyl. by estrogen-responsive elements has been

Environmental Health Perspectives 113 SOTO ETAL.

proposed to assess estrogenicity; however, Triangle Park, NC [NIEHS]). DES metab- year in the freezer maintained their elevated basal expression in the absence of olites were a gift of K.S. Korach (NIEHS). inhibitory properties on the proliferation estrogen often occurs, and this may raise Estradiol was stored as a 1-mM stock solu- of human estrogen-sensitive breast tumor concern about the reliability of these assays. tion in at -20°C. were MCF-7 cells; plasma- and blood-derived Therefore, the proliferative effect of natural dissolved in ethanol to a final concentra- sera were equally effective. estrogens on the female genital tract tion of 10 mM, except endosulfan mixed remains the hallmark ofestrogen action. isomers, dieldrin, and toxaphene, which The E-SCREEN Test Hertz argued convincingly that this were dissolved in dimethyl sulfoxide The E-SCREEN assay was developed based proliferative property should be adopted as (DMSO); they were all diluted to desired on the following premises: a) a human the one method to determine whether or concentrations in phenol red-free DME serum-borne molecule specifically inhibits not a chemical is an estrogen (2). This immediately before using. The final solvent the proliferation of human estrogen-sensi- requires measuring the increase of mitotic concentration in culture medium did not tive cells (12-16); and b) estrogens induce activity in tissues of the female genital tract exceed 0.1%; this concentration did not cell proliferation by canceling this after estrogen administration. However, affect cell yields. Human recombinant inhibitory effect (12,13,16). Nonestrogenic this method is not suitable for large-scale EGF, basic FGF, and IGF-1 were pur- steroids and growth factors did not abolish screening of suspected chemicals and an chased from Collaborative Research the proliferative inhibition by mammalian equally reliable, easy, and rapid-to-perform (Lexington, MA). serum (12,13). method would be preferable. The novel Cloned MCF-7 cells were trypsinized E-SCREEN assay fulfills these require- Plasma-derived and Blood-derived and plated into 12-well plates (Costar, ments (9). This assay measures estrogen- Human Serum Cambridge, MA) at initial concentrations induced increase of the number of human Plasma-derived human serum was prepared of 20,000 cells per well (9,10). Cells were breast MCF-7 cells and is recognized as from outdated plasma supplied by the New allowed to attach for 24 hr; then, the seeding biologically equivalent to the increase of England Medical Center Blood Bank, medium (5% FBS in DME) was removed mitotic activity in the rodent endometrium (Boston, MA). Calcium chloride was added and replaced by the experimental medium (9,10). The objectives of this study were to to a final concentration of 30 mM to facili- [5% CD human serum supplemented to validate the E-SCREEN assay and to test tate clot formation. Blood-derived serum phenol red-free DME (CDHuS)]. A range the estrogenicity of chemicals released into was obtained using blood from healthy ofconcentrations ofthe test compounds was the environment in large volumes. adult volunteers; blood was allowed to clot added to this medium. The bioassay was in glass centrifuge tubes for 2 to 4 hr to terminated on day 6 (late exponential Materials and Methods obtain serum. Plasma- and blood-derived phase) by removing the media from the serum were clarified by centrifugation wells, adding a cell lysing solution (10% Cell Line and Cell Culture Conditions (2000xg for 10 min), heat-inactivated ethylhexadecyl-dimethylammonium bro- Human breast cancer estrogen-sensitive (56°C for 30 min), centrifuged, charcoal- mide [Eastman Kodak Co., Rochester, MCF-7 cells were obtained from the dextran stripped, and stored in glass tubes NY] in 0.5% Triton X-100, 2 mM MgCl2, Michigan Cancer Foundation (Detroit, at -20°C until use. 15 mM NaCl, 5 mM phosphate buffer, MI) (11). For routine maintenance, cells pH 7.4) and counting the nuclei in a were grown in Dulbecco's modification of Removal ofSex Steroids by Coulter Counter Apparatus, Model ZM Eagle's medium (DME) (GIBCo, Grand Charcoal-Dextran Treatment ofSerum (Coulter Electronics, Hialeah, FL). Island, NY) supplemented with 5% fetal Charcoal (Norit A, acid washed; Sigma The best estimate of the proliferative bovine serum (FBS) (Hyclone, Logan, UT) Chemical Co, St. Louis, MO) was washed behavior of a cell population is td or dou- at 37°C in an atmosphere of 5% C02/95% twice with cold sterile water immediately bling time. td is the time interval in which air under saturating humidity. before using. A 5% charcoal-0.5% dextran an exponentially growing culture doubles T70 (Pharmacia-LKB, Uppsala, Sweden) its cell number. Determining td requires Steroids, Xenobiotics, and suspension was prepared. Charcoal-dex- measuring cell yields at several time inter- Growth Factors Tested tran (CD) suspension aliquots of a volume vals during the exponential proliferation 17,-Estradiol (E2) was obtained from similar to the serum aliquots to be phase. A less cumbersome alternative to Calbiochem (Richmond, CA). Other processed were centrifuged at 2500 rpm for measuring proliferation rates is comparing steroids were purchased from Steraloids 10 min. Supernatants were aspirated and the cell yield achieved by similar cell inocu- (Keene, NH). R26008 (allenolic acid) was serum aliquots were mixed with the char- la harvested simultaneously during the late supplied by Roussel-UCLAF, Romainville, coal pellets. This charcoal-serum mixture exponential phase of proliferation. The France. Toxaphene (technical grade) and was maintained in suspension by rolling at proliferative effect (PE) is measured as the endosulfan (technical grade) were obtained 4 cycles/min at 370C for 1 hr. This suspen- ratio between the highest cell yield from Chem Services (West Chester, PA). sion was centrifuged at 2000xg for 20 min. obtained with the test chemical and with Endosulfan a and P isomers, o,p'-DDT, The supernatant was then filtered through the hormone-free control. Under these p,p'-DDT, p,p'-DDD, p,p'-DDE, PCB a 0.45-pm Nalgene filter. Over 99% of experimental conditions, cell yield repre- congeners, , dieldrin, phtha- serum sex steroids were removed by this sents a reliable estimate of the relative pro- late esters, and antioxidants were from treatment when determined by removal of liferation rate achieved by similar inocula Ultra Scientific (North Kingstown, RI). 3H-E2 (12); E2 levels after CD treatment exposed to different proliferation regula- Hydroxylated biphenyls were a gift from J. were less than 0.01 pg/ml when measured tors. In our experimental design, MCF-7 A. McLachlan (National Institute of by radioimmunoassay. CD sera were stored cell yields were measured 6 days after to; Environmental Health Sciences, Research at -20°C until needed. Samples kept for 1 however, significant differences between

114 Environmental Health Perspectives THE E-SCREEN TESTAND NOVEL XENOESTROGENS

control and estrogen-treated cultures are N2. To extract receptor molecules, cells Shaffe's test (19). A p value of < 0.05 was apparent after 4 days (16). were incubated with 1 ml of extraction regarded as significant. The estrogenic activity of xenobiotics buffer (0.5 M KCI, 10 mM potassium was assessed a) by determining their relative phosphate, 1.5 mM EDTA, and 1 mM Results proliferative potency (RPP), which mea- monothioglycerol, pH 7.4) at 40C for Proliferative Effect ofCompounds sures the ratio between the minimal con- 30 min (17). After centrifugation to pellet Known To Be Estrogenic centration of estradiol needed for maximal the cell debris, receptor levels were mea- in Animal Models cell yield and the minimal dose of the test sured in 100--pl extract aliquots by enzyme compound needed to achieve a similar immunoassay using the Abbott estrogen E2 induced maximal cell yields at 10 to effect; and b) by measuring their relative and progesterone receptors kits (Abbott 100 pM using the E-SCREEN assay. proliferative effect (RPE), which is 100 Diagnostics, Chicago, IL) according to the Twenty-two compounds reported to have times the ratio between the highest cell manufacturer's instructions. estrogenic activity were also tested. Their yield obtained with the chemical and with RPP is listed in Table 1. Their relative E2. RPE is calculated as 100 x (PE-1) of the pS2 Assay potency measured by the E-SCREEN assay test compound/(PE-1) of E2. Thus, the Culture media were harvested after 72 hr correlated with their relative binding RPE indicates whether the compound of exposure to the test chemicals and cen- affinity to the and with being tested induces a proliferative response trifuged to eliminate floating and detached their biological effect in uterotropic assays. quantitatively similar to the one obtained cells; samples were kept frozen at -80°C Exceptions to these correlations have been with E2, that is, a full agonist (RPE = 100), until the immunoradiometric assay was reported in the literature; they reflect rates or a proliferative yield significantly lower performed following the manufacturer's of clearance and metabolization of estro- than the one obtained with E2, that is, a protocol (ELSA-PS2, CIS Bio Inter- gens (20). The E-SCREEN assay mimics partial agonist (9). Figure 1 displays a national, Gif-sur-Yvette, France). exposure to a constant level of hormone, schematic representation of these concepts. much like that achieved in animals by For screening purposes, the range ofxenobi- Determination ofRelative using estrogen-filled silastic implants. otic concentrations was from 1 nM to 10 Binding Affinities behaved as a full agonist in the E- pM, and for E2 from 0.1 pM to 1 nM, mea- MCF-7 cells were grown in 150-cm2 flasks SCREEN assay as it did when adminis- sured at intervals of one order ofmagnitude. in 5% FBS; they were harvested during the tered to animals in multiple doses (21). late exponential phase after 24 hr of expo- Similarly, the proliferative potency of DES Progesterone Receptor Assay and sure to 5% CDHuS. Cells were rinsed with metabolites measured by the E-SCREEN Estrogen Receptor Processing PBS, and a suspension of 20 x 106 cells/ml assay paralleled that in the uterotropic MCF-7 cells were seeded in 25-cm2 flasks of buffer (500 mM KCl, 1.5 mM EDTA, assay; however, pseudo-DES and indane- in 5% FBS-supplemented DME. Twenty- 10 mM Tris-HCl, pH 7.4, at 40C) was strol had poor uterotropic activity but were four hours later, the medium was changed sonified at 40C (5-sec pulses with 30-sec full agonists when assayed by the to 5% CDHuS, and the chemicals to be intervals). The cell homogenate was cen- E-SCREEN test (RPP = 10). The lowered tested were added. Control flasks were trifuged at 100,OOOxg for 40 min, and treated with vehicle. After 72 hr of expo- supernatant aliquots were incubated with 2 Table 1. Proliferative effect of diverse estrogens mea- sure to the test xenoestrogens, medium was nM 3H-E2 alone and in combination with sured by the E-SCREEN assay. aspirated, the cell layer was rinsed with unlabeled competitors at concentrations PBS, and the cells were frozen in liquid ranging from 1 pM to 1 pM E2 or 1 nM to Compound RPP, % 1 mM xenoestrogens for 16 hr at 40C. The 17j-Estradiol 100 reaction mixture contained 15% DMSO 17a-Estradiol 10 RPP 1 I0.1 to solubilize hydrophobic xenoestrogens. Estriol 10 600 - 100 This treatment did not alter the shape of 1000 the competition curve for E2 and nonyl- 16-Hydroxyestrone 0.1 = 400- E phenol, the only two compounds from Diethylstilbestrol (DES) 1000 37.5 which a competition curve could be cis- 0.001 c-a 200 - -{RPE obtained in the absence of DMSO. Metabolite E (from tamoxifen) 0.001 R26008 0.1 Separation of bound and free hormone was 11 3-Chloromethylestradiol 1000 done by CD adsorption (18). 0 0.001 0.1 10 1000 Indanestrol 1 0 Concentration Indenestrol-A 100 Statistical Aaysis Indenestrol-B 1 0 Figure 1. Schematic representation of the dose-response Results were expressed as the mean ± SE. Pseudo-DES 0.1 curve to E2 (-*-), a full agonist (-O-), and a par- Proliferation yield experiments conducted Pseudo-DES-e 10 tial agonist (-A-). The horizontal bars indicate that in duplicate wells were repeated at least a Pseudo-DES-z 10 RPP is a comparison between effective concentrations of Zearalenol 1 minimum of 5 times. Mean cell numbers Zearalenone 1 the agonist and E2. The vertical bars at the right of the from each experiment were normalized to graph box illustrate that RPE compares the ability of E2 d- 1 the -free control (100%) to correct 0.001 and of agonists to increase the cell yield over the values Ethynyl-estradiol 100 obtained in untreated controls. The RPE of the full agonist for differences in the initial plating den- in this figure is 100, that is, lOOx(500/100)- sity. Differences between the diverse Relative proliferative potency (RPP) is the ratio 1(500/100)-l. The RPE of the partial agonist is 37.5, steroid treatment groups were assessed by between 170-estradiol and doses that is, 100 x(250/1 00)- 1/(500/100)-1. analysis of variance and the a posteriori needed to produce maximal cell yields x 100.

Volume 103, Supplement 7, October 1995 1 15 SOTO ETAL.

estrogenic potency of these two compounds polyalkylated, hindered phenols such as among the 209 PCB congeners and their was attributed to slow processing or clear- butylated hydroxytoluene (BHT) and metabolites; the E-SCREEN assay should ance of the ligand-bound receptor (20). Irganox 1640 were not estrogenic, whereas facilitate this undertaking. This notion is at variance with the data dis- t-butylhydroxyanisole (BHA) was estro- cussed below regarding alkylphenols, genic. Among esters used as plas- Identification ofEstrgenic Pesticides which are active both in uterotropic and E- ticizers, those derived from alkylalcohols Table 5 lists a number of pesticides and SCREEN assays, while displaying dimin- such as dibutylphthalate and diamylphtha- industrial chemicals assayed by the E- ished processing ofestrogen receptors. late were not estrogenic, whereas butylben- SCREEN test. None of the herbicides and Nonestrogenic compounds (natural and zylphthalate was estrogenic. fungicides tested were estrogenic. The synthetic progestagens, glucocorticoids, and estrogenicity of DDT, its metabolites, derivatives such as mirex, chlor- Identification ofEstrogenic and was confirmed by the dane-a isomer, chlordane, and ) PCB Congeners E-SCREEN test. Among DDT isomers, did not affect the proliferation of MCF-7 Aroclor 1221 (9), 18 PCB congeners, and o,p'-DDT was slightly more potent than cells. No false positives were observed. 10 hydroxylated PCBs were tested (Table p,p'-DDT, showing significant activity at Moreover, insulin, transferrin, and EGF 4). Five PCB congeners were estrogenic in 1 pM, albeit lower than at 10 pM; p,p'- did not reverse the inhibitory effect of the E-SCREEN assay indicating that they DDT showed estrogenic activity only at serum (12,13,16). This conclusion was val- are estrogenic per se or they have under- 10 pM. Methoxychlor was expected to be idated further by using human recombinant gone hydroxylation by MCF-7 cells; their inactive in the E-SCREEN assay because it bFGF, EGF, and IGF-1 (Table 2). These RPP was 0.0001. None of the estrogenic requires metabolic activation, probably in results strengthen the reliability and PCBs was coplanar. Among the hydroxy- the liver. However, methoxychlor (98% specificity ofthis "in culture" assay. lated PCBs assayed, the most potent were pure, U.S. EPA standard) and p,p'-meth- 2', 5'-dichloro-4-hydroxybiphenyl, oxychlor (99.6% pure) induced the prolif- Identification ofNewXenoestrogens (RPP=0.0 1-0.001), 2',4',6'-trichloro-4- eration of MCF-7 cells. From these data amongAntioxidants and Plasticizers hydroxybiphenyl (RPP=0.01), and we infer that MCF-7 cells have the enzy- An estrogenic contaminant was isolated 2',3',4',5'-tetrachloro-4-hydroxybiphenyl matic complement necessary to activate from modified polystyrene centrifuge tubes (RPP=0.001). 2',5'-dichloro-3-hydroxy- proestrogens. Comparable results were (Corning Glass Co., Corning, NY, Cat. biphenyl and 2',3',4',5'-tetrachloro-3- reported by White et al. (22) regarding No. 25310-15). After purification by flash hydroxybiphenyl were also estrogenic, alkylphenol diethoxylates. A heptachlor chromatography and reverse-phase HPLC, albeit 10-fold less potent than their 4- derivative, 1-hydroxychlordene, showed the estrogenic compound was identified by hydroxy isomers (Figure 2). More con- submaximal estrogenic activity (RPE= 40) gas chromatography-mass spectrometry as geners should be assayed to derive at 10 pM. Technical grade endosulfan and a p- isomer (10). This nonyl- meaningful structure-activity relationships a and ,B endosulfan isomers were estrogenic phenol was a full estrogen for MCF-7 cells (RPE=100; RPP=0.0003 %, Table 3). Table 3. Estrogenic effect of industrial chemicals measured by the E-SCREEN assay. Nonylphenol also increased the mitotic index of the endometrial epithelium in Compound Concentrationa RPE, %C RPP, %d adult ovariectomized rats. As expected Estradiol 30 pM 100 100 from a genuine estrogen, it also induced Phenol 10 PMb 0 progesterone receptor in MCF-7 cells. 4-Ethylphenol 10 pMb 5 4-Propylphenol 10 pM 17 p-Nonylphenol is 10 to 50 times more 4-sec-Butylphenol 10 pM 76 0.0003 potent an estrogen than kepone and 4-tert-Butylphenol 10 pM 71 0.0003 o,p'-DDT, and it mimics both the prolifer- 4-tert-Pentylphenol 10 pM 105 0.0003 ative and inductive properties of natural 4-Isopentylphenol 10 pM 93 0.0003 estrogens. Alkylphenols with at least a 4-Butoxyphenol 10 pMb 0 three-carbon alkyl chain were also found to 4-Hexyloxyphenol 10 pMb 0 4-Hydroxybiphenyl 10 pM 87 0.0003 be estrogenic; p-octylphenol was the more 4,4'-Dihydroxybiphenyl 10 pM 84 0.0003 potent one (RPP = 0.03). Other phenolic 1-Naphthol 10 pMb 0 antioxidants were tested; among them 2-Naphthol 10 pMb 0 5,6,7,8-Tetrahydronaphthol-2 10 pMb 0 10 38 Table 2. Effect of recombinant human growth factors 6-Bromonaphthol-2 PM on the proliferation of MCF-7 cells grown in medium 5-Octylphenol 100 nM 100 0.03 containing 5% charcoal-dextran stripped human serum 4-Nonylphenol 1 pM 100 0.003 (CDHuS). Nonylphenol, technical grade 10 pM 102 0.0003 t-Butylhydroxyanisole 50 pM 30 0.00006 Compound Concentration Cell number Benzylbutylphthalate 10 pM 90 0.0003 Control, 5% CDHuS - 38,140 ± 5,900 'The lowest concentration needed for maximal cell yield. bIndicates the highest concentration tested in culture. 5% CDHuS + estradiol 10 pM 214,120 ±3,400 cThe relative proliferative effect is calculated as 100x(PE-1) of the test compound/(PE-1) of E2; a value of 100 5% CDHuS + bFGF 10 ng/ml 27,310 ±285 indicates that the compound tested is a full agonist, a value of 0 indicates that the compound lacks estrogenicity 5% CDHuS + EGF 10 ng/ml 29,820 ±2,844 at the doses tested, and intermediate values suggest that the xenobiotic is a partial agonist. dRelative prolifera- 5% CDHuS + IGF-1 10 ng/ml 49,140 ±1,519 tive potency is the ratio between E2 and xenobiotic doses needed to produce maximal cell yields x 100. All com- Cells were harvested after 6 days of exposure to the pounds designated as full or partial agonists increased cell yields significantly over the hormoneless control experimental media. (p< 0.05).

116 Environmental Health Perspectives THE E-SCREEN TESTAND NOVEL XENOESTROGENS

Table 4. Estrogenic effect of (PCB) congeners and hydroxylated PCBs measured by the Nonylphenol (J0) and octylphenol (22) E-SCREEN assay. also induced PR. Compound Name Concentrationa RPE, %C RPP, %d Xenoestrogens and the Prssing 1 2-Monochlorobiphenyl lo pMb 4.4 of Estrogen Receptors 2 4-Monochlorobiphenyl 10 PMb 2.1 - 3 2,5-Dichlorobiphenyl 10 pMb 3.7 - Estradiol treatment of MCF-7 cells for a 4 2,6-Dichlorobiphenyl 10 pMb 3.4 - period of 72 hr decreased the level of estro- 5 3,5-Dichlorobiphenyl 10 pMb 2.7 - gen receptors by 50% (receptor process- 6 2,3,4-Trichlorobiphenyl 10 pM 77.0 0.0001 ing); this effect is interpreted by some as an 7 2,3',5-Trichlorobiphenyl 10 pMb 2.2 - important step on estrogen action (24). 8 2,3,6-Trichlorobiphenyl 10 PMb 5.8 - 9 2,2',4,5-Tetrachlorobiphenyl 10 pM 61.6 0.0001 Others have reported that the proliferative 10 2,3,4,4'-Tetrachlorobiphenyl 10 pMb 4.7 - effect of estradiol also occurs in the absence 1 1 2,3,4,5-Tetrachlorobiphenyl 10 pM 39.2 0.0001 of processing (25). Treatment with xeno- 12 2,3,5,6-Tetrachlorobiphenyl 10 pMb 3.1 - estrogens such as endosulfan, toxaphene, 13 2,4,4',6-Tetrachlorobiphenyl 10 PM 75.7 0.0001 dieldrin (Table 8), and nonylphenol (not 14 2,3,3',4,5-Pentachlorobiphenyl 10 pMb 1.0 - shown) did not decrease estrogen receptor 15 2,3,4,5,6-Pentachlorobiphenyl 10 pMb 4.4 levels in MCF-7 cells; to the contrary, a 16 2,2',3,3',5,5'-Hexachlorobiphenyl 10 pMb 1.0 - 17 2,2',3,3',6,6'-Hexachlorobiphenyl 10 PM 61.6 0.0001 slight increase of up to 57% was recorded. 18 Decachlorobiphenyl 10 pMb 3.5 - 19 2',5'-Dichloro-2-hydroxybiphenyl 10 pM 13.0 0.0001 Competitive Binding to the 20 2',5'-Dichloro-3-hydroxybiphenyl 10 PM 69.9 0.0001 Estrgn Receptor 21 2',5'-Dichloro-4-hydroxybiphenyl 1 pMe 71.2 0.001 o,p'-DDT, endosulfan, toxaphene, and 22 3,5-Dichloro-2-hydroxybiphenyl 10 pMb 5.4 - nonylphenol competed with 3H-E2 for 23 3,5-Dichloro-4-hydroxybiphenyl 10 pMb 1.5 - 24 2,2', 5-Trichloro-4-hydroxybiphenyl 1 pM 37.8 0.001 binding to the receptor; their relative 25 2',4',6'-Trichloro-4-hydroxybiphenyl 100 nM 99.8 0.01 binding affinities correlated with their rel- 26 2',3',4',5'-Tetrachloro-3-hydroxybiphenyl 10 PM 35.3 0.0001 ative proliferative potency. Table 9 com- 27 2',3',4',5'-Tetrachloro-4-hydroxybiphenyl 1 pM 92.0 0.001 pares the relative binding affinities of these 28 2',3',4',5,5'-Pentachloro-2-hydroxybiphenyl 10 pMb 4.8 - xenoestrogens, their IC50 values and their aIndicates the lowest concentration needed for maximal cell yield. bIndicates the highest concentration tested in relative proliferative potency. The concen- culture. CRelative proliferative effect is calculated as 100 x (PE-1) of the test compound/(PE-1) of E2; a value of 100 tration of estradiol and xenoestrogens nec- indicates that the compound tested is a full agonist, a value of 0 indicates that the compound lacks estrogenicity essary to decrease 3H-E2 binding by 50% at the doses tested, and intermediate values suggest that the xenobiotic is a partial agonist. dRelative prolifera- were about one order of magnitude higher tive potency is the ratio between E2 and xenobiotic doses needed to produce maximal cell yields x 100. All com- than the concentration needed to achieve pounds designated as full or partial agonists increased cell yields significantly over the hormoneless control maximal cell proliferation yields, PR and (p<0.05). 'This compound exhibited submaximal proliferative activity at 100 nM (RPE:62%). pS2 induction. found to be estrogenic at 10 pM. The Cumulative Effect ofXnoestogens RPEs of these compounds were lower than Analysis of fat or serum from wildlife and those of endosulfan (dieldrin, RPE= 55%; humans often reveals the simultaneous 450- toxaphene, RPE = 52%) (23). The fact that presence of several xenoestrogens (1); these E30001 these compounds have toxic effects at con- findings suggest that xenoestrogens may act centrations one order of magnitude higher cumulatively. By using the E-SCREEN than those needed to evoke a proliferative assay we verified this suggestion. Figure 3 response precluded assessing whether shows the additive effect of 10 xenoestro- HC A B C D E F higher concentrations would attain full gens, each administered at one tenth of the W NC M- 10plM 1ija estrogenic activity in this bioassay. minimal effective dose that produces a WIOnlJm 011p Induction ofProgserone Receptor proliferative effect. Figure 2. Proliferative activity of HC, hormoneless and the control; A, estradiol; B, 2',5',2-hydroxy-DCB; C, 2',5',3- pS2 by Newly Identified Discussion hydroxy-DCB; D, 2',5',4-hydroxy-DCB; E, 2',3',4',5',3- X oestogns The deleterious impact ofxenoestrogens on hydroxy-TCB; and F, 2',3',4',5',4-hydroxy-TCB. Pesticides found to be estrogenic by the the reproductive success, development, and Asterisks (*) indicate significant differences with hor- E-SCREEN assay were also effective induc- health of animals is well documented; the moneless control (p<0.05). ers of markers of estrogen action such as realization that humans are also exposed progesterone receptors (PR) and pS2 in and at risk has become increasingly obvious MCF-7 cells (Table 7). Dieldrin signifi- (26). Data showing a lowering of sperm at concentrations of 10 to 25 pM. It cantly increased pS2 levels; PR levels quality and quantity, increased , should be noted that the RPE of all these increased slightly at the maximal dose and spontaneous rates in humans chemicals is lower than that of estradiol tested (10 IM). Dieldrin is toxic at concen- suggest that environmental estrogens play a (endosulfanmixed isomers RPE 81 %, trations higher than 10 pM; this preduded role in the toxicology of human reproduc- a isomer RPE = 77%, 3 isomer RPE 78%) testing whether higher concentrations tion and development (27,28). An objec- (Table 6). Dieldrin and toxaphene were would result in full induction of PR. tive causal relationship between detrimental

Volume 103, Supplement 7, October 1995 11 7 SOTO ETAL.

Table 5. Xenobiotics tested by the E-SCREEN assay. Table 6. Estrogenic effect of pesticides according to the E-SCREEN assay. Estrogenic xenobiotics Nonestrogenic xenobiotics Compound Concentration RPE, % RPP, % Herbicides None 2,4-D 2,4-DB Estradiol 10 pM 100 100 Alachlor DDT* 10 pM 79.61 0.0001 Butylate Cyanazine o,p'-DDT 10 pM 86.14 0.0001 Dacthal Dinoseb p,p'-DDT 10 pM 71.00 0.0001 Hexazinone Metolachlor Dieldrin 10 pM 54.89 0.0001 Propazine Picloram Endosulfan* 10 pM 81.25 0.0001 Trifluralin Simazine ,B-Endosulfan 10 pM 78.26 0.0001 Insecticides ax-Endosulfan 10 pM 77.17 0.0001 p,p'-DDT Bendiocarb Carbofuran 1-Hydroxychlordene 10 pM 40.00 0.0001 o,p'-DDT Chlordane Chlordimeform Kepone 10 pM 84.00 0.0001 o,p'-DDE Diazinon Chlorpyrifos Toxaphene 10 pM 51.90 0.0001 p,p'-DDE Heptachlor Carbaryl Methoxychlor 10 pM 57.00 0.0001 DDTa Kelthane Concentration describes the dose at which an estro- Dieldrin Malathion Mirex genic effect is detected; maximal cell yield is obtained Chlordecone (kepone) Methoprene Pyrethrum at concentrations between 10 and 100 pM estradiol. Endosulfana Parathion Most xenobiotics are active at 10 pM. The RPE mea- a-Endosulfan sures the ratio between the maximal cell yield P-Endosulfan achieved by the xenobiotic and that of estradiol. The Methoxychlor RPP measures the ratio between the dose of estradiol Toxaphene and that of xenobiotic needed to achieve a prolifera- Fungicides tive effect. Asterisks (*) denote that the compound None Chlorothalonil Thiram referred to was of technical grade. Hexachlorobenzene Zineb Maneb Ziram Metiram Table 7. Effect of xenoestrogens on progesterone Industrial chemicals receptor and pS2 levels. 2,3,4-TCB Butylated hydroxytoluene 2-CB 2,2',4,5-TCB 2,5-DCB 4-CB Concen- 2,3,4,5-TCB 2,6-DCB 2,3,6-TCB Compound tration PRa pS2b 2,4,4',6-TCB 3,5-DCB 2,3,5,6-TCB Control - 7.9 ± 5.4 53.9± 16.7 2,2',3,3',6,6'-HCB 2,3',5-TCB 2,3,3'4,5'-PCB Estradiol 1 nM 153.1 +49.5* 179.6±50.2 2-OH-2',5'-DCB 2,3,4,4'-TCB 2,2',3,3',5,5'-HCB Endosulfan 1 pM 71.0 ± 13.1 * 129.2 ± 6.2* 3-OH-2',5'-DCB 2,3,4,5,6-PCB Endosulfan 10 pM 136.0 ± 27.2* 172.1 ± 61.6* 4-OH-2',5'-DCB 2-OH-2',3',4',5,5'-PCB DecaCB Toxaphene 1 pM 18.1 ±9.2 59.9±6.2 4-OH-2,2',5-TCB 2-OH-3,5-DCB Diamyl phthalate Toxaphene 10 pM 128.7 ± 19.1* 108.0 ± 17.6* 4-OH-2',4',6'-TCB 4-OH-3,5-DCB Dibutyl phthalate Dieldrin 1 pM 8.2 ± 1.7 210.3 ± 29.9* 3-OH-2',3',4',5'-TCB 1 ,2-Dichloropropane Dimethyl isophthalate Dieldrin 10 pM 32.3 ± 11.7 232.7 ± 32.6* 4-OH-2',3',4',5'-TCB Irganox 1640 Dimethyl terephthalate 4-OH-alkyl-phenols 2,3,7,8-TCDD Dinonyl phthalate aProgesterone receptor (PR) levels are expressed as Bisphenol-A Tetrachloroethylene Octachlorostyrene femtomoles per mg cellular protein. bpS2 levels in the 4-OH-biphenyl Styrene culture medium are expressed as ng/106 cells. Each t-Butylhydroxyanisole result represents the mean ± SD of three experiments. Benzylbutylphthalate An asterisk (*) denotes a significant difference with the a Denotes a technical grade isomer mixture. hormoneless control (p<0.05).

Table 8. Effect of xenoestrogens on estrogen receptor health effects and their presumed causation exposure have been verified. The finding of processing. by xenoestrogens is tempered however by reproductive effects caused by xenobiotics the lack of appropriate technology to has largely been accidental. For example, Compound Concentration Estrogen receptora explore this subject on a large scale. The workers at a kepone-producing devel- Control - 183.9 ± 59 first obstacle encountered is that the estro- oped azoospermia and impotence (29); this Estradiol 1 nM 92.3 ± 23.1 genicity of chemicals cannot be predicted was the first observation of reproductive Endosulfan 1 pM 205.1 ± 21.7c it is Because of the occupa- Endosulfan 10 pM 238.3 ± 17.6c solely on structural bases; therefore, toxicity by kepone. Toxaphene 1 pM 261.9±1521,c unknown how many xenoestrogens are pre- tional nature of this case, the culprit kepone Toxaphene 10 pM 288.7±31.7bc sent in the ecosystem. The E-SCREEN became readily apparent. In contrast, Dieldrin 1 pM 236.3±21.2bc bioassay is a reliable tool to rapidly assess wildlife are exposed to a combination of Dieldrin 10 pM 252.2±17.1 ,c com- xenobiotics. It became clear in the Great estrogenicity on a large number of aEstrogen receptor levels are expressed as femto- pounds; in this paper we describe how its Lakes studies that it was difficult to sort out moles/mg cellular protein. Each result represents the use helped to identify estrogens among which one of the xenobiotics played a mean ±SD of at least three experiments. bDenotes a environmental pollutants. The second causal role or whether the signs of intoxica- statistically significant difference with the hormone- obstacle to overcome is how to identify tion were due to cumulative interaction less control. cDenotes a statistically significant differ- causal agents when signs of estrogen among the chemicals present in affected ence with the estradiol-treated group (p< 0.05).

118 Environmental Health Perspectives THE E-SCREEN TESTAND NOVEL XENOESTROGENS

Table 9. Relative binding affinities (RBAs) and IC50 values of xenoestrogens. 400-

Compound RBA RPP IC50 Concentration* e 300- * Estradiol 100 100 1.5±0.4 nM 100 pM ,B-Endosulfan 0.00024 0.0001 631.0±88 pM 10 pM Toxaphene 0.00032 0.0001 470.0±38 pM 10 pM o,p'-DDT 0.00031 0.0001 485.0±42 pM 10 pM p-Nonylphenol 0.021 0.001 7.2±3 pM 1 pM 100 *Concentration needed for maximal proliferative effect. All xenoestrogen IC50 values were significantly different A B C D E F G H I J K L M from that of estradiol. Figure 3. Proliferative activity of A, hormoneless con- trol; B, 10 pM 1-endosulfan; C, lpM 3-endosulfan; D, animals (1). Exposure to environmental levels are mostly constant, much like the 1 pM a-endosulfan; E, 1 pM toxaphene; F, 1 pM dield- estrogens, singly or in combination, may ones achieved when animals are treated rin; G, 1 pM 2,3,4,5-tetrachlorobiphenyl; H, 1 pM p,p'- be easily assessed in male fish, , or with estrogen-filled silastic implants. This DDT; I, 1 pM 2,2',3,3',6,6'-hexachlorobiphenyl; J, 1 pM birds used as sentinels by measuring their approach is more relevant to chronic envi- p,p'-DDD; K, 1 pM p,p'-DDE; L, 1 pM methoxychlor; vitellogenin plasma levels. Instead, expo- ronmental exposure than that of measuring and M, mixture of the 10 chemicals indicated as C to L, sure of females to each at 1 pM. Asterisks (*) indicate significant differ- xenoestrogens is more acute effects after a single dose. In both ences with hormoneless control (p< 0.05). difficult to ascertain through a marker such types of assays, metabolism of the suspected as vitellogenin because the serum levels of xenoestrogen into more or less active com- this protein are high in animals laying eggs pounds is uncertain and should be defined New Estrogns Identfied by the (30,31). In addition, there is no compara- individually for each compound. For exam- E-SCREENAssay ble marker to ascertain exposure in ple, nonylphenol diethoxylate was estro- Novel xenoestrogens were found among mammals. Again, the E-SCREEN assay genic for MCF-7 cells; since it does not antioxidants, plasticizers, polychlorinated represents the best alternative to resolve compete for estradiol binding to the estro- biphenyl (PCB) congeners, and pesticides. this second obstacle. gen receptor, it is likely that estrogenic Alkylphenols are used as antioxidants activity results from nonylphenol diethoxy- and in the synthesis of detergents Animal Bioassays and the E-SCREEN late metabolism to the free phenol (22). Differences [alkylphenol polyethoxylates, (APEs)]. Assay: and Similarities Methoxychlor was also believed to be inac- APEs are used as industrial detergents in Diverse animal models and assays have tive until metabolized to free phenols, pre- the textile and paper industries, in toi- been used to measure estrogenicity. Allen sumably in the liver; again, methoxychlor letries, and as spermicides. Four hun- and Doisy (32) and other pioneers of tested positive when assayed by the E- dred and fifty million pounds of APEs estrogen research used mouse and rat SCREEN test. Therefore, even though the were sold in the United States in 1990. activity units to follow their estrogen putative proestrogens tested so far were APEs are not estrogenic per se; however, purification protocol; the end point of estrogenic when assayed by the E- they are degraded during sewage treat- their assay was vaginal cornification. SCREEN test, an added step in the quest ment. The polyethoxylate chain is Dodds and Lawson(33) used both the for identifying all xenoestrogens may shortened, and free alkylphenols as well Allen and Doisy assay (32) and the femi- include their metabolic activation by liver as mono and diethoxylates are pro- nization of the feather pattern in brown microsome extracts prior to their testing by duced. The free phenols are estrogenic leghorn capons (33). Others adopted the the E-SCREEN assay. (9,10). Recently, White et al. (22) have uterotropic assay using single or multiple Regarding quantitative effects, while shown that the diethoxylates are also doses of estrogens over 24- to 72-hr peri- kepone is 100,000 to 1,000,000 times less estrogenic. These APE degradation ods in immature or ovariectomized mice potent than estradiol according to the products have been detected in drinking and rats. This diversity of end points indi- E-SCREEN assay, an increase of the rat water (34). Nonylphenol has been cates that there is no universal "gold stan- uterine wet weight comparable to that of reported to leach from PVC tubing for dard" of estrogen action among animal estradiol occurred with a 1000- to 5000-fold milk processing (35) and plastics used bioassays. The E-SCREEN assay appears to higher dose of chlordecone than that of in food packaging (36). APEs such as be the best candidate for establishing a estradiol (3). This discrepancy may be due those used as spermicides are degraded quantitative standard of estrogenic activity to rapid metabolism of estradiol and persis- to free nonylphenol when administered at the target organ level. As shown in tence and bioaccumulation of chlordecone to rodents (37). The contribution of Tables 1 through 6, no false positives or in animals. Differences between results in APEs and alkylphenols to the xenoestro- negatives were observed among the estro- culture and in live animals reflect the dif- gen burden of humans is unknown; gens and nonestrogens tested. ferent parameters used as a measure of however, it has been reported that these No qualitative differences could be estrogenicity. On one hand, the rodent chemicals are present in sewage outlets found when comparing animal assays and assay measures the increase of uterine wet in concentrations sufficient to feminize the E-SCREEN assay; that is, the estro- weight [water imbibition, hypertrophy sentinel fish (38). Alkylphenols accu- genic properties of compounds character- (which is also produced by estrogen antag- mulate in river sediment and in the fat ized using animal bioassays was also onists), and hyperplasia] (5), while on the of exposed fish (39). Some phenolic ascertained using the E-SCREEN test. other hand, the human E-SCREEN bio- antioxidants such as butylated hydroxy- From a pharmacokinetic perspective, the assay measures cell proliferation only. This toluene (BHT) and t-butylhydroxy- latter measures estrogenicity at the target is a necessary and sufficient parameter to anisole (BHA) are used to prolong the cell level under conditions where estrogen define estrogen action (2). shelf life of foodstuffs and to reduce

Volume 103, Supplement 7, October 1995 1 19 SOTO ETAL.

nutritional losses by retarding oxidation. than that achieved with E2 or DDT PR as well; this confirms their estrogen- Interesting observations pertaining to (45,46). Therefore, the estrogenicity of mimicking properties and the specificity of structure-function relationships were PCB mixtures would be best ascer- the E-SCREEN assay as a tool to identify made: a) the alkyl chain must at least tained by first determining which con- estrogens. These xenoestrogens compete have 3 carbons; b) the p-isomers are geners are estrogenic. Korach et al. (47) with estradiol for binding to its receptor; more potent estrogens than the m-iso- demonstrated the ability of certain their RBA's correlated well with their mers; c) polyalkylated, hindered phenols hydroxy-PCBs to bind to estrogen potency to induce cell proliferation, pS2, like BHT and Irganox 1640 (Ciba- receptors and to produce an uterotropic and PR. Recent data suggest that alkylphe- Geigy, Basel, Switzerland) are not estro- effect that correlated with their relative nols bind to the estrogen-binding domain genic while being effective antioxidants; binding affinity to the estrogen recep- of the estrogen receptor (22). Binding to and d) fused rings like naphthols are tor. It is generally assumed that hydrox- the receptor is a necessary but insufficient not estrogenic in spite of being an inte- ylated PCBs are estrogenic while property to define estrogenicity. Tetra- gral part of the A and B ring of natural nonhydroxylated PCBs are not. In this hydronaphthol, which is not estrogenic, is steroids. Instead, substituted naphthols paper we show that 5 of the 18 con- effective in preventing the forward binding such as 6-Br naphthol and allenolic acid geners studied were estrogenic in the of estradiol to its receptor while alkylphe- are estrogenic; more studies are needed E-SCREEN assay (Table 4). It is nols, which are estrogenic, are effective in to assess whether these substituted unknown whether they were estrogenic displacing prebound estradiol from its naphthols are active due to a bulk effect, per se or they were hydroxylated by receptor (54). These data were interpreted electronegativity, or because flat mole- MCF-7 cells; hydroxylated PCBs are as indicative of interactions with more than cules sucli as naphthols and coplanar more potent than their nonhydroxy- one site on the estradiol receptor. PCBs are unable to bind tightly to the lated counterparts. The phenyl rings in estrogen receptor. the active compounds were not copla- Cumulative Effect ofXenoestrogens In addition to the estrogenic alkylphenol nar. Among the hydroxylated PCBs, Xenoestrogens may act cumulatively (Figure antioxidants described above, we found that those p-hydroxylated were more potent 3) and with endogenous estrogens thus dis- BHA was estrogenic. BHA is a widely used than m-hydroxylated; o-hydroxylated rupting the endocrine system of exposed antioxidant; because it controls oxidation of compounds were even less active. wildlife and humans. Hence, measuring short-chain fatty acids such as coconut oil Hindered phenols, such as in 4-OH- the total estrogenic burden due to envi- (40). Maximal usage levels of BHA permit- 3,5,-DCB were less estrogenic than ronmental contaminants present in a ted by U.S. Food and Drug Administration unobstructed ones. plasma/tissue sample may be more mean- (U.S. FDA) varies according to the food The pesticides dieldrin and toxaphene ingful than to assess exposure by measur- type, from 50 ppm in dry breakfast cereals are estrogenic. Their use has been ing the levels of each of the known to 1000 ppm in active yeast (41). restricted in the United States since xenoestrogens. Recently, an epidemiologi- Plasticizers are used to decrease the 1974 and 1982, respectively (23). cal study showed a positive correlation rigidity of certain polymers. For the These compounds are highly lipophilic between breast cancer and serum levels of most part, they are di- and triesters of and bioaccumulate in ecosystems; they DDE, a DDT metabolite (55-57), leaving organic acids. Phthalate esters are widely are still found in wildlife, coincidentally open the possibility that xenoestrogen expo- used plasticizers (42). These compounds with signs of reproductive impairment. sure increases the incidence of breast can- leach from plastics, and they have been Toxaphene is a main airborne pollutant cer. Since xenoestrogens are postulated to found to be ubiquitously distributed in in North America, and its residues be a risk factor for breast cancer, measuring the environment, including marine appeared in regions where it has never a single xenoestrogen may not be a reliable ecosystems (43,44). Among phthalate been used, like the Arctic and indicator of exposure because different per- esters butyl benzyl phthalate was estro- Scandinavia (48). It is present in Arctic sons eating different diets may be exposed genic whereas those derived from alkyl and Baltic salmon muscle fat at concen- to different xenoestrogens. Therefore, mea- alcohols such as dibutyl phthalate and trations of 700 to 7000 ppb (49); this suring total xenoestrogen burden represents diamyl phthalate were not. concentration is well within those a more reliable approach to assess the link * Aroclor 1221 was not estrogenic by the producing estrogenic effects in the between xenoestrogens and breast cancer. E-SCREEN assay (data not shown). E-SCREEN assay (10 pM = 4800 ppb). The E-SCREEN test may be used to this Many congeners are present in Aroclor Endosulfan was introduced in 1954; it end once a protocol is developed to separate mixtures; therefore, it is likely that the is presently used for agricultural purposes environmental estrogens from endogenous maximal concentration used, 10 pM, in the United States and other countries ones. In addition, in a preventive approach, resulted in levels of individual con- (50). Proliferative, estrogenlike effects in the E-SCREEN test may be used to screen geners lower than those needed to MCF-7 cells were found at doses of 10 pM chemicals for their estrogenicity before they induce cell proliferation. PCB mixtures (4060 ppb). Endosulfan was shown to pro- are released into the environment. such as Aroclor 1221 were reported to duce testicular atrophy in male rats fed a be estrogenic using uterotropic assays as diet containing 10 ppm (51,52); it also end points (increased uterine wet lowered gonadotrophin and REFERENCES weight, increased uterine glycogen con- plasma levels (53). These results are consis- 1. Colborn T, vom Saal FS, Soto AM. tent) (45). However, the magnitude of tent with its estrogenicity revealed by the Developmental effects of endocrine- the uterine wet weight increase was E-SCREEN test. disrupting chemicals in wildlife and only marginally significant, analyzed by These newly identified estrogens not humans. Environ Health Perspect inappropriate statistical tests and lower only induced cell proliferation but pS2 and 101:378-384 (1993).

120 Environmental Health Perspectives THE E-SCREEN TESTAND NOVEL XENOESTROGENS

2. Hertz R. The estrogen problem - retrospect and prospect. In: 23. Soto AM, Chung KL, Sonnenschein C. The pesticides endosul-. Estrogens in the Environment II-Influences on Development fan, toxaphene, and dieldrin have estrogenic effects on human (McLachlan JA, ed). New York:Elsevier, 1985;1-1 1. estrogen sensitive cells. Environ Health Perspect 102:380-383 3. Hammond B, Katzenellenbogen BS, Kranthammer N, (1994). McConnell J. Estrogenic activity of the insecticide chlordecone 24. Gyling M, Leclercq G. Estrogenic and antiestrogenic down-reg- (kepone) and interaction with uterine estrogen receptors. Proc ulation of estrogen receptor levels: evidence for two different Natl Acad Sci USA 76:6641-6645 (1979). mechanisms. J Recept Res 10:217-234 (1990). 4. Meyers CY, Matthews WS, Ho LL, Kolb VM, Parady TE. 25. Reiner GCA, Katzenellenbogen BS. Characterization of estro- Carboxylic acid formation from kepone. In: Catalysis in gen and progesterone receptors and the dissociated regulation Organic Synthesis (Smith GW, ed). New York:Academic Press, of growth and progesterone receptor stimulation by estrogen in 1977;213-255. MDA-MB-134 human breast cancer cells. Cancer Res 5. Clark JH, Watson C, Upchurch S, Padykula H, Markaverich 46:1124-1131 (1986). B, Hardin JW. Estrogen action in normal and abnormal cell 26. Thomas KB, Colborn T. Organochlorine endocrine disprup- growth. In: Estrogens in the Environment (McLachlan JA, ed). tors in human tissue. In: Chemically-induced Alterations in New York:Elsevier, 1980;53-67. Sexual Development: The Wildlife/Human Connection 6. Lieberman ME, Maurer RA, Gorski J. Estrogen control of pro- (Colborn T, Clement C, eds). Princeton,NJ:Princeton lactin synthesis in vitro. Proc Natd Acad Sci USA 75:5946-5949 Scientific Publishing, 1992;365-394. (1978). 27. Sharpe RM, Skakkebaek NE. Are oestrogens involved in falling 7. Ramsdell JS, Tashjian AH. Thyro-tropin releasing hormone sperm count and disorders of the male reproductive tract? and epidermal growth factor stimulate prolactin synthesis by a Lancet 341:1392-1395 (1993). pathway(s) that differs from that used by phorbol esters: disso- 28. Giwercman A, Carlsen E, Keiding N, Skakkebaek NE. ciation of actions by calcium dependency additivity. Evidence for increasing incidence of abnormalities of the Endocrinology 117:2050-2060 (1985). human testis: a review. Environ Health Perspect 101(Suppl 8. Palmiter RD. Quantitation of parameters that determine the 2):65-71 (1993). rate ofovalbumin synthesis. Cell 4:189-197 (1975). 29. Guzelian PS. Comparative toxicology of chlorodecone (kepone) 9. Soto AM, Lin T-M, Justicia H, Silvia RM, Sonnenschein C. in humans and experimental animals. Annu Rev Pharmocol An "in cu!ture" bioassay to assess the estrogenicity of xenobi- Toxicol 22:89-113 (1982). otics. In: Chemically Induced Alterations in Sexual 30. Bergink EW, Wallace RA, VandeBerg JA, Bos ES, Ab G. Development: The Wildlife/Human Connection (Colborn T, Estrogen-induced synthesis of yolk proteins in roosters. Am Clement C, eds). Princeton, NJ:Princeton Scientific Zool 14:1177 (1974). Publishing, 1992;295-309. 31. Redshaw MR, Follett BK. Physiology of egg yolk production 10. Soto AM, Justicia H, Wray JW, Sonnenschein C. p-Nonyl- by the fowl: the measurement of circulating levels of vitel- phenol: an estrogenic xenobiotic released from "modified" logenin employing a specific radioimmunoassay. Comp polystyrene. Environ Health Perspect 92:167-173 (1991). Biochem Physiol A 55:399-405 (1976). 11. Soule HD, Vasquez J, Long A, Albert S, Brennan MJ. A 32. Allen E, Doisy EA. An ovarian hormone: preliminary report on human cell line from a pleural effusion derived from a breast its localization, extraction and partial purification, and action in carcinoma. J Natl Cancer Inst 51:1409-1413 (1973). test animals. JAMA 81:819-821 (1923). 12. Soto AM, Sonnenschein C. The role of estrogens on the prolif- 33. Dodds EC, Lawson W. Molecular structure in relation to eration of human breast tumor cells (MCF-7). J Steroid oestrogenic activity. Compounds without a phenathrene Biochem 23:87-94 (1985). nucleus. Proc Roy Soc Lond B Biol Sci 125:222-232 (1938). 13. Soto AM, Sonnenschein C. Mechanism of estrogen action on 34. Clark LB, Rosen RB, Hartman TG, Louis JB, Suffet IH, cellular proliferation: evidence for indirect and negative control Lippincott RL, Rosen JD. Determination of alkylphenol on cloned breast tumor cells. Biochem Biophys Res Commun ethoxylates and their acetic acid derivatives in drinking water 122:1097-1103 (1984). by particle beam liquid chromatography/mass spectrometry. Int 14. Lykkesfeldt AE, Briand P. Indirect mechanism of oestradiol J Environ Anal Chem 47:167-180 (1992). stimulation of cell proliferation of human breast cancer cell 35. Junk GA, Svec HJ, Vick RD, Avery MJ. Contamination of lines. BrJ Cancer 53:29-35 (1986). water by synthetic polymer tubes. Environ Sci Technol 15. Briand P. Lykkesfeldt AE. Long-term cultivation of a human 8:1100-1106 (1974). breast cancer cell line, MCF7, in chemically defined medium. 36. Gilbert MA, Shepherd MK, Startin JR, Wallwork MA. Effect ofestradiol. Anticancer Res 6:85-90 (1986). Identification by gas chromatography-mass spectrometry of 16. Soto AM, Silvia RM, Sonnenschein C. A plasma-borne specific vinylchloride oligomers and other low molecular weight com- inhibitor of the proliferation of human estrogen-sensitive breast ponents of poly(vinylchloride) resins for food package applica- tumor cells (estrocolyone-I). J Steroid Biochem Mol Biol tions. J Chromatogr 237:249-261 (1992). 43:703-712 (1992). 37. Knaak JB, Elridge JM, Sullivan LJ. Excretion of certain poly- 17. Maddedu L, Legros N, Devleeschouwer N, Bosman C, Piccart ethylene glycol ether adducts of nonylphenol by the rat. M, Leclercq G. Estrogen receptor status and estradiol sensitivity Toxicol Appl Pharmacol 9:331-340 (1966). of MCF-7 cells in exponential growth phase. Eur J Cancer Clin 38. Purdom CE, Hardiman PA, Bye VJ, Eno NC, Tyler CR, Oncol 24:385-390 (1988). Sumpter JP. Estrogenic effects from sewage treatment works. 18. Sonnenschein C, Olea N, Pasanen ME, Soto AM. Negative Chem Ecol 8:275-285 (1994). controls of cell proliferation: human prostate cancer cel s and 39. Shiraishi H, Carter DS, Hites RA. Identification and determi- androgens. Cancer Res 49:3474-3481 (1989). nation of tert-amylphenols in carp from the Trenton channel of 19. Winer BJ. Design and analysis of single-factor experiments. In: the Detroit river, Michigan, U.S.A. Biomed Environ Mass Statistical Principles in Experimental Design. 2nd ed. New Spectrom 18:478-483 (1989). York:McGraw-Hill, 1962;149-260. 40. Shahidi F, Wanasundara PK. Phenolic antioxidants. Crit Rev 20. Korach KS, Metzler M, McLachlan JA. Diethylstilbestrol Food Sci Nutr 32:67-103 (1992). metabolites and analogs. J Biol Chem 254:8963-8968 (1979). 41. Singer PL. Occupational oligospermia. JAMA 140:1249 21. Martin L, Pollard JW, Fagg B. Oestriol, oestradiol-17,B and the (1949). proliferation and death of uterine cells. J Endocrinol 69: 42. Tepper L. Phthalic acid esters - an overview. Environ Health 103-115 (1976). Perspect 3:179-182 (1973). 22. White R, Jobling S, Hoare SA, Sumpter JP, Parker MG. 43. Giam CS, Neff GS, Atlas EL, Chan HS. Phthalate ester plasti- Environmentally persistent alkylphenolic compounds. cizers: a new class of marine pollutant. Science 199:419-421 Endocrinology 135: 175-182 (1994). (1978).

Volume 103, Supplement 7, October 1995 121 SOTO ETAL.

44. Randall R, Ozretich R, Boese B. Acute toxicity of butyl benzyl 51. NCI. Bioassay of endosulfan for possible carcinogenicity. Tech phthalate to the saltwater fish English sole, Parophrys vetulus. Rpt Ser NCI-CG-TR-62. National Cancer Institute 62:18-54 Environ Sci Technol 17:670-672 (1983). (1978). 45. Ecobichon DJ, MacKenzie DO. The uterotropic activity of 52. Gupta PK, Gupta RC. Pharmacology, toxicology and degrada- commercial and isomerically-pure chlorobiphenyls in the rat. tion of endosulfan. A review. Toxicology 13:115-130 (1979). Res Commun Chem Path Pharmacol 9:85-94 (1974). 53. Singh SK, Pandey RS. Effect of sub-chronic endosulfan expo- 46. Gellert RJ. Uterotrophic activity of polychlorinated biphenyls sures on plasma gonadotrophins, testosterone, testicular testos- (PCB) and induction of precocious reproductive aging in terone and enzymes of androgen biosynthesis in rat. Indian J neonatally treated female rats. Environ Res 16:123-130 (1978). Exp Biol 28:953-956 (1990). 47. Korach KS, Sarver P, Chae K, McLachlan JA, McKinney JD. 54. Mueller GC, Kim U-H. Displacement of estradiol from estro- Estrogen receptor-binding activity of polychlorinated hydroxy- gen receptors by simple alkylphenols. Endocrinology biphenyls: conformationally restricted structural probes. Mol 102:1429-1435 (1978). Pharmacol 33:120-126 (1987). 55. Wolff MS, Toniolo PG, Lee EW, Rivera M, Dubin N. Blood 48. ATSDR. Toxicological Profile for Toxaphene. Rpt No TP-90- levels of organochlorine residues and risk of breast cancer. J 26. Atlanta:Agency for Toxic Substances and Disease Registry, Natl Cancer Inst 85:648-652 (1993). 1990. 56. Krieger N, Wolff M. Breast cancer and serum organochlorines: 49. Paasivirta J, Rantio T. Chloroterpenes and other organochlo- a prospective study among white, black, and Asian women. J rines in Baltic, Finnish and Arctic wildlife. Chemosphere Natl Cancer Inst 86:589-599 (1994). 22(1):47-55 (1991). 57. Savitz D. Re: Breast cancer and serum organochlorines: a 50. ATSDR. Toxicological Profile for Endosulfan. Rpt No TP- prospective study among white, black and Asian women. J Natd 91/16. Atlanta:Agency for Toxic Substances and Disease Cancer Inst 86: 1255 (1994). Registry, 1991.

122 Environmental Health Perspectives