Also included in the test panel were17,Bestradiol as a positive control and ethanol as solvent control

Helle RaunAndersen,1 Anna-Maria Andersson,2 Steven F. Arnold,3 Herman Autrup,4 Marianne Barfoed,5 Nicola A.

Beresford,6

Poul

Bjerregaard,7

Lisette B.Christiansen,7 Birgitte Gissel,5 Rene Hummel,2 Eva Bonefeld J0rgensen,4 Bodil

Korsgaard,7

Remy LeGuevel,8 Henrik Leffers,2 John

McLachlan,3

AnetteM0ller,1 JesperBo Nielsen,1 Nicolas

Olea,9

Anita

Oles-Karasko,10

Farzad Pakdel,8 Knud L. Pedersen,7 Pilar Perez,9 Niels Erik Skakkebcek,2 Carlos

Sonnenschein,10

Ana M.Soto,10 John P. Sumpter,6 SusanM.Thorpe,5and PhilippeGrandjean1

The aim of this study was tocompare results obtained by eight different short-term assays of estrogenlike actions of chemicals conducted in 10 different laboratories in five countries. Twenty chemicals were selected to represent direct-acting estrogens, compounds with estrogenic metabolites, estrogenic antagonists, and a known cytotoxic agent. Also included in the test panel were17,Bestradiol as a positive control and ethanol as solvent control. The test compounds were codedbefore distribution. Test methods included direct binding to the estrogen receptor (ER), proliferation of MCF-7 cells, transient reporter gene expression in MCF-7 cells, reporter gene expression in yeast strains stably transfected with the human ER and an estrogen-responsive reporter gene, andvitellogenin production in juvenile rainbow trout.

17p-Estradiol,

177a-ethynyl estradiol, and diethylstilbestrol induced a strong estrogenic response in all test systems.

Colchicine caused cytotoxicity only. Bisphenol A inducedan estrogenic response in all assays.

The results obtainedfor the remaining test compounds-tamoxifen, ICI 182.780, testosterone, bisphenol A dimethacrylate, 4-n-octylphenol, 4-n-nonylphenol, nonylphenol dodecylethoxylate, butylbenzylphthalate, dibutylphthalate, methoxychlor, o,p'-DDT, p,p'-DDE, endosulfan, chlomequat chloride, andethanol-varied among the assays. The results demonstrate that careful standardization isnecessary toobtainareasonable degree of reproducibility. Also, similar methods varyintheir sensitivity toestrogenic compounds. Thus, short-term tests are useful for screening purposes, but the methodsmustbefurther validated by additional interlaboratory and interassaycomparisons to document the reliabilityof the methods.- Environ Health Perspect 107(Suppl 1):89-108 (1999). http://ehpnetl.niehs.nih.gov/docs/1999/Suppl-1/89-108andersen/

abstract.html

Key words:estrogenic chemicals, estrogens, antiestrogens, estrogenicity tests, binding assay, yeast, MCF-7, vitellogenin

Several man-made chemicals that are andthe toxicologyof the substances are widely distributed inthe environment may poorly understood.

have the potential to mimicestrogens or Estrogenic chemicals identifiedtodate otherwise disrupt the endocrine system includesome organochlorine pesticides, (1-4). Reliableshort-term methods are suchas o,p'-DDT and methoxychlor, and needed toidentifysuchchemicals,tochar- industrial chemicals and byproducts, acterize and control the environmental including some

polychlorinated

biphenyl load, and to evaluatehuman exposures. congeners, alkyl phenols, phthalates, and Attempts to develop these methods must bisphenolA (5-8).Thechemicalstructure confronttheproblemthat thebiologicfate of these chemicals varies

substantially,

which makes it difficult to predict their estrogenicity solely on astructural basis.

Estrogenicity was first defined as a physiologic response to a compound that inducedestrus in vivo. An in vivo rodent uterinebioassay wasdeveloped in which an estrogen-induced uterotropic response was estimated asan increase inuterine tissue mass in ovariectomized or immature rodents (9). This assay is oftenregarded as agold standard. However, although it has beenwidelyused formany years, the assay has still not been fullystandardized (10), and even when the same protocol is used, the results sometimes vary between differ- entlaboratories (11,12). Finally, different tissues mayrespond differently to estro- genic chemicals. A well-known example is tamoxifen, anantiestrogen in breast tissue but a uterotropic estrogenic agonist in uterine tissue (13). Morebroadlybased in vivo assays should aim atdetecting the response in different tissues, thereby becoming more sensitive and informative.

Theadvantages ofthe rodentutero- tropic assay and other in vivo assays are that theytake into consideration the effects of metabolism, plasma-protein binding, and pharmacokinetics. These methods can also detectestrogenicresponses due toaltered metabolism ofexogenous orendogenous estrogensas causedby, for example,some hepatotoxic chemicals such as carbon tetra- chloride(14).However,in vivomethodsare in general expensive andtime-consuming,

Manuscript receivedatEHP 26 June1998;accepted21 September1998.

1Departmentof Environmental Medicine,OdenseUniversity, Odense, Denmark; 2Departmentof Growth andReproduction,The NationalUniversityHospital, Copenhagen, Denmark; 3EnvironmentalEndocrinology Laboratory,Center for Bioenvironmental Research,TulaneUniversity,NewOrleans, Louisiana;4Departmentof Environmental Medicine,UniversityofArhus,Arhus,Denmark; 5NovoNordiskANS,Malov, Denmark;6DepartmentofBiologyandBiochemistry, BrunelUniversity, Uxbridge,Middlesex,UnitedKingdom;7DepartmentofBiology,OdenseUniversity, Odense,Denmark;8Laboratorie DEndocrinologieMoleculaire delaReproduction, CampusdeBeaulieu,Rennes,Cedex, France;9Laboratoryof MedicalInvestigation, Departmentof Radiology,UniversityofGranada, Granada,Spain; 10Departmentof AnatomyandCellularBiology, TuftsUniversity School ofMedicine,Boston,Massachusetts.

AddresscorrespondencetoH.R.Andersen,Environmental Medicine,Winsloewparken 17,OdenseUniversity, DK-5000 OdenseC, Denmark.Telephone:45 6557 3765.Fax:456591 1458.E-mail:[email protected]

Abbreviations used:APE, alkylphenolpolyethoxylate; BBP,benzylbutylphthalate;BSA,bovineserumalbumin;cat,chloramphenicol acetyltransferase;CD,charcoal dextran;DBP,dibutylphthalate;DCC,dextran-coatedcharcoal;DES,diethylstilbestrol; DME, Dulbecco's modification ofEagle'smedium; ELISA,enzyme-linked immunosorbent assay;ER, estrogen receptor;ERE,estrogen-responsiveelement;FBS,fetal bovineserum;GST,glutathioneStransferase; hER,humanestrogenrecep- tor;1C50,concentration that inhibits50%;LBA,ligand-bindingassay;LBD,ligand-bindingdomain; NP,nonylphenol;NPE,nonylphenolpolyethoxylate;NP12E0,nonylphe- noldodecylethoxylate; ONPG, o-nitrophenylo-galactopyranoside; OP, octylphenol; PAGE,polyacrylamide gelelectrophoresis;PBS,phosphate-bufferedsaline; RBA, relativebindingaffinity; RP,relativepotency;rtER,rainbowtroutestrogenreceptor;RT-PCR,reversetranscriptasepolymerasechainreaction; SDS, sodiumdodecylsuk fate;SHBG,sex-hormonebindingglobulin;SRB, sulforhodamine-B; TCA,trichloraceticacid; tk,thymidinekinase.

ANDERSENETAL

and avariety of short-term assaysare there- forebeingapplied toidentify estrogenic chemicals and todetermine the relative potencies for hormonal responses. Some of these assays may besuitable forscreening large numbers of chemicals andcontami- natedmedia,such as waterandfood,and may therefore be useful tools forpriori- tizingchemicalsfor more extensive studies in vivo.

However,most in vitroassays estimate primarily the intrinsic estrogenic activity of thechemicalsasreflectedby theirbind- ing toandactivationof theestrogen recep- tor (ER) without taking into account factorsthat mayaffecttheir activityinthe intactorganism. These factors include:

bindingafflnityofthe chemical topro- teins (especially sex-hormone binding globulin [SHBG] andalbumin), abilityto enter targetcells,degradation timeinthe organism, and the concentration of endogenousestrogen.

Mostcirculating endogenousestrogen (>99%) is boundtoplasmaproteins and onlyaminorfractionisable topenetrate into the cells and activate the ER. Estro- genicchemicals such as

o,p'-DDT,

octyl- phenol (15), bisphenol A (16), and the potentsyntheticestrogendiethylstilbestrol (DES)

(15,1/)

haveamuch loweraffinity for these proteins than does

17p-estradiol.

Hence, the major part ofthesechemicals in theblood wouldbe availablefor activation ofthe cellularER. Therelative potencyof thechemicalscompared toestradiol may thereforebeunderestimated byin vitro tests conducted withoutaddition ofplasmapro- teins, as has

recendy

beendemonstratedfor bisphenolA(16).

The biodegradation of estrogenic chemicals may influence theestrogenic response, as alongdegradation time could enhance theresponse. Someenvironmental chemicalsareverylipophilic andresistant tometabolism andmaythereforeaccumu- late inorganismsand reachconcentrations sufficienttoproduce estrogenic responses in vivo. Hence,although theenvironmen- talestrogenscharacterizedto date are con- siderablyless potent thanestradiol(1/50th to

1/10000th)

based on ERbindingaffini- ties oreffects seen in cell cultures (5,18), lowconcentrationsmay still bedetrimental to the reproductive success ofexposed organisms. Forexample, feminizationand decreased reproduction ingulls wasappar-

endy

elicited by

o,p'-DDT

in eggs at con- centrations as low as 5

mg/kg,

which is similar to concentrations reported in the environment(19).

Because

estrogenic

chemicals compete with

endogenous

estrogens for

binding

to

ERs,the concentration of

endogenous

^estro- gen mayinfluence the

estrogenic

effects of the

chemicals,

andachemicalthaton its

own shows aweak

estrogenicity might

therefore

potentially

act asan

antiestrogen

in vivo.In

addition,

humans and wildlifeare

exposed

toseveral

potential

estrogenic chemicals

simultaneously.

Addition oftwo ormoreweakestrogensmayinduceanaddi- tive

effect,

asdemonstratedinMCF-7 cells

(5)

and

by

inductionof

vitellogenin

in male fish

(20).

Toobtainareliable

screening

system for

estrogenicity,

itis necessarytovalidate the short-term assays and compare the results

tothose obtainedindifferenttypes of in vivostudies. Further

development

ofthe short-term assay systemsmaybe

needed,

andseveral assays may needtobecombined in a

screening panel.

Asanimportantstep in this process, itisnecessarytocompare andvalidate thoseshort-termassays

already developed

and used

by

differentlaborato- ries.

Therefore,

acomparison study of short-term tests forestrogenicactivityof chemicals wasinitiated. Previous studies have

compared

the results obtained in different assay systems

using

the same

chemicals within thesame

laboratory (7,21-23).

In this

study,

10laboratories

agreed

toparticipateandtest apanelof20 chemicalswithout

knowing

theidentityof thechemicalspriorto

testing.

Theshort-term methodswere chosen to representdifferent types of response.

Binding

assaysmeasurethespecific

binding

to the ER but

provide

^no informationon

activationof thereceptor. The

prolifera-

tion assay inMCF-7 cells

(E-SCREEN)

measures acellular responseknown tobe induced

by

estrogens but reveals^nodirect information ofthe mechanism involved.

The reportergene

expression

^assaysdetect activation of the ER

leading

to

binding

of

an

estrogen-responsive

element

(ERE)

and

expression

of^areporter gene. Several reportergene assayswere

included,

i.e.,a

mammalian cell type

(MCF-7) transiently

transfectedwithERE and ^areportergene anddifferentyeast assays

stably

transfected with ERandERE-linkedreporter genes.

Finally,

^an in vivo assay

using

inductionof

vitellogenin

ⁱⁿ

juvenile

rainbowtroutwas

includedbecause

vitellogenin

inductionin

juvenile

ormalefish is

reportedly

^avery sensitive

biologic

marker for

estrogenicity (20).

Severalof the assays

(e.g.,

the E- SCREEN and the yeast

assays)

have been

widely

used in an attempt to

identify estrogenic

chemicals

(5,8,24-28).

Thetestchemicals

(Table

1;

Figure 1)

were selected ^to represent documented

direct-acting

^{potentestrogens}

(1 7p-estra- diol, 17a-ethynyl estradiol, DES),

acom-

plete antiestrogen (ICI 182.780),

a

partial antiestrogen (tamoxifen),

^apotentandro- gen

(testosterone),

and environmental

pol-

lutants

reported

^to be

estrogenic

in differentinvitroassaysaswell^as in vivo

by inducing

^a

uterotropic

responseinrodents.

The responseswere inducedeither

directly

[i.e., octylphenol (OP) (15,27,29-31), nonylphenol (NP) (12,23,27,31-35),

and

o,p'-DDT (5,21,27,36-39)]

or after

Table 1. Testchemicals.

Compound(CAS no.) 173-Estradiol(50-28-2) 17j3-Ethynylestradiol(57-63-6) Diethylstilbestrol(56-53-1) Tamoxifen(10540-29-1) ICI 182.780(129453-61-8)a Testosterone(58-22-0) BisphenolA(80-05-7)

Bisphenol A-dimethacrylate(3253-39-2) 4-rn-Octylphenol(1806-26-4)

4-n-Nonylphenol (25154-52-3) Nonylphenoldodecylethoxylate Benzylbutylphthalate(85-68-7) Dibutylphthalate(84-74-2) Methoxychlor(72-43-5) o,p'-DDT(789-02-6) p,p'-DDE(72-55-9)

Endosulfan(mixed isomers)b(115-29-7) Chlormequatchloride(999-81-5) Colchicine(64-86-8)

Ethanol

Supplier(Codeno.) Sigma(E1132) Sigma(E0882) Sigma(D4628)

Aldrich(28,161-1) Zeneca Sigma(T1268) Aldrich(23,965-8) Aldrich(15,632-9) Ehrenstorfer(C157120) Ehrenstorfer(C156300) Promochem

Riedel-deHaen(36927) Riedel-de Haen(36736) Ehrenstorfer(C150600) Ehrenstorfer(C120810)

Ehrenstorfer(C120410) Ehrenstorfer(C131200) Ehrenstorfer(Cl13400) Sigma(C3915) BDH(15338SE)

Purity,% 99.4 99.1 99 96 99.3 100

99 99.7 99.4 99.9 7 97 98 98.4 99.8 99.7 99 97 95 99.8

?,unknown.aGiftfromZenecaPharmaceuticals.h75%a-isomer and 24%

13-isomer.

Testno.

A B C D E F G H J K L MN 0 p 0 R S T

Environmental HealthPerspectives *Vol 107,

Supplement

1 * February 1999 90

17ax-Ethynylestradiol

CH3

HO

\ /t \OH

CH3

Diethyistilbestrol

/CH3 OCH2CH2N\

Tamoxifen

ICI 182.780

HO / CgH19

4-n-Nonylphenol

H19C9 (OCH2CH2),2-OH 4-n-Nonylphenol dodecylethoxylate, 12 ether groups

Testosterone

CH3

H OH

CH3 BisphenolA

CH2 CH3 CH2

co/o c ooo-~

C- O O C- C

OH3 CH3 OH3

BisphenolAdimethacrylate

HO C8H17

4-n-Octylphenol

Benzylbutylphthalate

0

11

C,O-O-C4H9

C -O-C4H9

0

Dibutylphthalate

CC13

CH30 CH 0H3

Methoxychlor

ClCC13

o,p'-DDT 0012

p,p'-DDE

Cl

cl

t°'s=sO

0I~~

cl

Endosulfan, mixed

r ~~~CH3 +

Cl-CH2-CH2-N-CH31 Clo

CH3

Chlormequat chloride

CH30 0

N,,

CH3O0 OH H OCH3

OCH3

Colchicine Figure 1. Chemicalstructureof thetestcompounds.

metabolism [i.e., methoxychlor (5,12,40)].

Other testcompounds have been reported

estrogenic in vitrobutnot uterotropic in

vivo: benzylbutylphthalate (BBP), dibutyl- phthalate (DPB) (6,23), and endosulfan (5,41,42), and twoof the testcompounds have been reported estrogenic in vitro, but

no informationonuterotropicresponses is

available: bisphenol A (8,27,28) and bisphenol A-dimethacrylate (8). Inaddi- tion, ^anantiandrogen [p,p'-DDE (43)], ^a nonylphenolpolyethoxylate (nonylphenol dodecylethoxylate [NP12EO]) that degradestoNP(31),^achlorinated^nonaro- maticcompound (chlormequat chloride), ^a chemical with known cytotoxic effect (colchicine), and ethanolassolvent control

wereincluded.

Materials and Methods

Chemicals

The chemicalswereobtained from Sigma Chemical Company (St. Louis, MO), Ehrenstorfer (Augsburg, Germany), Riedel-de Haen(Seelze, Germany),Aldrich

(Steinheim, Germany), Promochem, (Wesel, Germany), or Zeneca Pharmaceuticals (Cheshire, UK), aslisted in Table 1. The purity of the chemicals stated in the table

wasreported by the supplier.

Identical test panels consistingof18 vials eachcontainingapproximately 2 ml stocksolutions of10mmole/liter ofthetest

substanceswereset up.Alltestcompounds

were dissolvedin pure ethanol from BDH (Poole, UK). Inaddition, the testpanel included one vial containing 2-ml 10

imol/liter 17p-estradiol

used ^as

positive

control andonevialcontaining 2ml pure

ethanolassolvent control. Thesetestpanels

were used for allassaysexceptinduction of vitellogenin production inrainbowtrout

because thisassayrequiredmuch moreof the testcompounds. Thesame batches of chemicals were used for this assay. The chemicalswereweighed, coded, andtrans-

ferred tobrownglassvials (10.0 ml, La- Pha-Pack, Langerwerde, Germany) before delivery ^to laboratory 9. The chemicals

weredissolved in laboratory 9. Details of thisassayaregivenbelow and in "Results."

Alltestcompounds ^wereweighed and dissolved in ethanol and transferred ^to brown glass vials (4.0 ml, La-Pha-Pack) with screw caps ofbutyl^gummy with Teflon foliation(Brown Chromatography, Wiirzburg, Germany). DES, 17ax-ethynyl estradiol, and 1

7p-estradiol

^were handled separately after all the othertest com-

poundswereweighedand dissolved. As the laststep, ethanol for the solvent controls

wastransferred directly to the testvials.

Because the solvent controlswereprepared after thestronglyactive estrogens,the risk of cross-contamination was maximized, although rigorous efforts were made to

limit this risk. All testvialswereplaced at

-20°C until their distribution ^to thepar-

ticipating laboratories. Duringtransport

the testvials ^werekepton dry ice. After arrival thetest panelwas kept at-200C until theanalyses^wereperformed.All par-

ticipating institutions wereinformed that the testpanelcontained hazardous toxic chemicals and should be handled accord- ingly. The laboratories and methods are

summarizedinTable2.

OH

1713-Estradiol

H

ANDERSEN ETAL.

Table 2. Laboratories and methods includedinthestudy Laboratory name (laboratoryno.)

Department ofEnvironmentalMedicine, OdenseUniversity,Odense,Denmark(1) Department ofAnatomyandCellular Biology,

TuftsUniversity,Boston,Massachusetts(2) Laboratory of MedicalInvestigation,

UniversityofGranada,Granada, Spain(3) DepartmentofBiologyandBiochemistry,

BrunelUniversity, Uxbridge,Middlesex,UK(41

LaboratorieD'Endocrinologie Moleculairede laReproduction,CampusdeBeaulieu, Rennes, Cedex, France (5)

Department of Growth andReproduction, TheNational University Hospital, Copenhagen,Denmark(6)

DepartmentofEnvironmental Medicine, University ofArhus,Arhus,Denmark(7) NovoNordiskA/S,Malav, Denmark(8) DepartmentofBiology,OdenseUniversity,

Odense,Denmark(9)

EnvironmentalEndocrinology Laboratory, TulaneUniversity, New Orleans, Louisiana(10)

Direct Competitive Estrogenin Vitro BindingAssay (Laboratory 6) Basedon RecombinantHumanEstrogenReceptor Isolated by Reverse Transcriptase- Polymerase ChainReaction

from

MCF-7 Cells. The complete codingregion of the ER (amino acids 2 to 595) was prepared byreverse transcriptase-polymerase chain reaction (RT-PCR) fromcDNA prepared fromtotalRNAfrom MCF-7 cells. cDNA was synthesized from 1-pg total RNA using0.5-pg T14Vprimer (V corresponds toA, C, or G) as described inAusubel et al. (44). One microliter of the cDNA sample wasused forPCR using 5 p native Pfu-enzyme (Stratagene, La Jolla, CA) including30pmol of eachprimer:

5'-primer: 5'- CGGGGA TCCACCAT GACCCTCCACACCAAAG-3';

3'-primer: 5'-GAGGAATTCCGACT GTGGCAGGGAAACCCTC-3'.

Nucleotides in italics were addedto facilitate cloning. Thecycleconditionsona GeneAmp PCR System 9600 (Perkin- ElmerCorp.,Norwalk, CT) were960Cfor 3min,followed by40cyclesof96°Cfor 30 sec, 680C for 1 min, 740C for 3min, and finally, 740Cfor 8 min. Theligand-binding domain (LBD) (amino acids 282 to 595) was prepared from the complete coding regionDNAfragmentbyPCRasdescribed previously, except that the 5'-primer was exchanged with 5'-GAAGGATCCTCT

GCTGGAGACATGAG-3'

and only 15

92

Test method E-SCREEN,proliferationofMCF-7 cells E-SCREEN,proliferationofMCF-7 cells E-SCREEN, proliferationofMCF-7 cells

Recombinantyeast(Saccharomyces)estrogenscreen

expressing hER

Recombinant yeast(Saccharomyces)estrogenscreen

expressingrtER

Direct competitive estrogeninvitrobindingassay, recombinant hER

Recombinant yeast(Saccharomyces)estrogen

screenexpressinghER

Transientgene expression assayinMCF-7 cells In vitroERbindingassay,rabbituterine tissue

Vitellogenin production injuvenile rainbowtrouts 1n V/vo

Recombinantyeast(DY159)estrogenscreen expressinghER

cycles were performedusing an

annealing

temperature of520C. The resulting DNA fragments were digestedwith restriction enzymes BamHI and EcoRI and cloned directionallyinto pGEXGTH (45),allow- ing 32P-end labeling ofthe Est-LBD. The insert in the purified plasmid DNA wvas sequenced on an ALFexpress sequenator (Amersham-Pharmacia-Bioteck, Uppsala, Sweden) using the ThermoSequenase

(Amersham-Pharmacia-Bioteck)

and CY5 fluorescence labeled primers. The recombi- nant proteins were purifiedas

glutathione

S-transferase (GST) fusion proteins from Escherichia colilysates, using thebulkGST purification module as described by the manufacturer (Amersham-Pharmacia- Bioteck). To normalize the amount of fusionproteinused in each assay,analiquot ofthe Sepharose-GST-LBD (about 5% of total amount used) was 32p labeledwith

[y32P]adenosine

triphosphate (Amersham- Pharmacia-Bioteck) and 2.5-pbovine heart muscle kinase (Sigma Chemical) as

describedbythemanufacturer (Amersham- Pharmacia-Bioteck) and then mixed 1:20 (vol:vol) with the unlabeled Sepharose- GST-LBD, resulting in approximately 200cpm/10pl.

In the binding assay, 10 plaliquots of Sepharose-GST-LBD mixed with [32p]_

labeledSepharose-GST-LBD were trans- ferred to Eppendorftubes (Eppendorf- Netheler-HinzGmbH, Hamburg,Germany)

and 1

xphosphate-buffered

saline

(PBS) (140

mM

NaCI,

2.7 mM

KCI,

10.1 mM

Na2HPO4,

1.8 mM

KH2PO4; pH 7.3)

^was added to afinal

volume

of48

pl.

All the chemicals ^weredilutedⁱⁿ 96% ethanol to

the desiredconcentrations and mixed 1:1 with

3H-estradiol (10 pCi/ml).

Two micro-

liters of this mixture was added to the

Sepharose-GST-

1BD beadsand incubated

at room temperaturefor45 minwith slow agitation. The beads^were collected by^cen-

trifugation

andwashed fourtimeswith 500-

pl

^1xPBS before theywere

resuspended

ⁱⁿ

50-pl

1 xPBS and transferred to 5 ml

Ultima

Goldscintillation solution

(Packard

Instrument

Co.,

Meriden,

CT)

and mea-

sured in a scintillation ^counter measuring both 3H-estradiol and

32P

activity. The

[3P]-labeled Sepharose-GST-LBD

^was

used to normalize for the amount of

Sepharose-GST-LBD

used in eachassay.

The

binding

of

3H-estradiol

^to the

Sepharose-GST-LBD

fusion protein was

plotted

^{as a}

function

of the actual chemical

concentration usedin the assay. Thecon- centration thatinhibits50%

(ICJ0)

^wascal- culated as the chemical concentration

reducing

the 3H-estradiol

binding

to 50%

ofmaximal

binding.

Data representmeanof twvo

independent

experiments.

InVitro

Estrogen Receptor Binding Assay (Laboratory

8)Basedon Rabbit Uterine Tissue.Aclassical

ligand-binding

assay

(LBA) employing

dextran-coated charcoal

(DCC)

toseparatebound

and

free

ligands

^was used asdescribed in detail in EORTC

(46)

and

Thorpe (47).

Cytosol

prepared

from rabbit uterine tissue was the

source of the ER-rich

cytosol;

rabbit muscle wasthe sourceof

ER-poor cytosol.

Fresh

aliquots

of

cytosols

^were thawed ^on the

day

^of

analysis.

Both

cytosols

were

diluted with assay buffer

[PB:

10 mM

K2HPO4/KH2PO4,

1.5 mM

K2EDTA,

10 mM

monothioglycerol,

10mM

Na2MoO4 2H20,

10%

glycerol (v/v); pH

7.51 to

approximately

3 mg

cytosol protein/ml.

ER-rich

cytosol

^wasdilutedwith

ER-poor cytosol

^toachieve

approximately

^20to25%

maximal

binding

of 0.5 nM3H-

17J-estra-

diol

(Amersham-Pharmacia-Bioteck).

Radioinert 1

7f3-estradiol

^wasobtained from

Sigma

Chemical.

Thestocksolutionsoftest

compounds

were further diluted

by

ethanol for

analysis

with PB.

Aliquots

of10

pl

^test

compounds

were incubatedwith 20

pl

3H-estradiol

(assay

concentration 0.5 nM) and 50

pl cytosol

^inmicrotiter

plates

^for 18 to20 hr

at4°C. Forcontrol

samples

aswellasmaxi-

mal

binding samples,

10

pl

PB wasadded

EnvironmentalHealthPerspectives *Vol 07, Supplement - February 999

in lieu oftest compound. To assess DCC background counts, 50 pl=0.3% bovine serum albumin (BSA) in PB was added in lieu of cytosol. To terminate the binding reaction, 100 pl DCC slurry (0.5% acti- vated charcoal [Sigma Chemical] and 0.005% dextran T70 [Amersham-Pharma- cia-Bioteck] in PB) were added to each sample and incubated with continuous shaking for 15 min at4°C. To separate bound andfreetitrated estradiol, micro- titerplates (Microwell, Nunc, Roskilde, Denmark) were centrifuged for 10 min (800xg) at 40C. Aliquots of 100 plwere removedfrom each sample for scintillation counting usingOptiflour scintillation liq- uid(Packard BioScience B.V, Groningen, The Netherlands). Standards and control samples were incubatedinquadruplicate;

testcompoundswere incubated indupli- cate.Themean countsper minute ineach sample wascalculated, background (DCC) was subtracted, and percentof maximal

3H-17f-estradiol

bindingwas calculated.

This value wasplottedagainst theconcen- tration of test compound incubated (expressed logarithmically). IC50values wereusedtocomparebindingaffinities.

E-SCREENBasedonProliferation of Human BreastCancer Cells (MCF-7) (Laboratories 1, 2, and 3). Thisassay introduced bySoto etal. (48)isbased on theestrogen-sensitivehumanbreastcancer cellline MCF-7. These cells require the presenceofestrogento growas tumors in a host.WhenMCF-7 cellsaregrownin acul- ture mediumsupplemented with non- estrogenic charcoal-dextran (CD)-stripped humanserum, proliferation is prevented.

When estrogen isadded, the cells prolifer- ate. This assaywasperformed in threedif- ferent laboratoriesasdescribedbySotoetal.

(48) (laboratory 2)orslightly modifiedas describedbyVillalobosetal. (49)

(laborato-

ries 1 and3). Briefly, stock cultures of MCF-7 BUS cells (passage 143 to 148), weregrown inDulbecco's modificationof Eagle'smedium(DME) supplementedwith 5%fetal bovineserum (FBS)in anatmos- phere of5%C02/95% airundersaturating humidity at 37°C. MCF-7 cells were trypsinized andplated in 12-well

plates

(Costar, Cambridge, MA)

(laboratory

2) or 24-wellplates (Limbro,McLean,VA)(labo- ratories 1 and3) atinitial concentrationsof

104

cells per well. Cells wereallowed to attach for 24hr, thenthe

seeding

medium (5% FBSin DME [laboratories2 and3] or 10% FBS in DME [laboratory 1]) was

replaced with the

experimental

medium (5%CD-treatedFBS [laboratory 2]^or 10%

CD-treated human serum [laboratories 1 and 3] supplemented to phenol red-free DME). CD-treated FBS and CD-treated human serum were prepared as described by Soto etal. (48) based on plasma supplied from local blood banks. A range of concen- trationsof the test compounds was added to thismedium.Allchemicals were diluted to desired concentrations with DME immedi- ately prior to use. The bioassay was termi- nated on day 6 (late exponential phase) by removingthe media from the wells. In labo- ratory 2 acell-lysing solution (10% ethyl- hexadecyl-dimethylammonium bromide) (Eastman Kodak, Rochester, NY) in 0.5%

Triton X-100, 2 mM MgCl2, 12 mM NaCl, 5 mM phosphate buffer, pH 7.4) was added and thenuclei counted in a Model ZMCoulter Counter Apparatus (Coulter Electronics, Hialeah, FL). In laboratories 1 and 3 the cells werefixedand stained with sulforhodamine-B (SRB) asdescribed by Brotons etal. (24) andVillalobos et al.

(49). Briefly, cellsweretreated with cold 10% trichloraceticacid (TCA) and incu- bated at 40C for 30 min, then washed five timeswith tap water and left todry. TCA- fixed cells were stained for 10 min with 0.4% (w/v) SRB dissolved in 1% acetic acid.Wellswererinsed with 1% acetic acid and airdried. Bound dye was solubilized with 10 mM Tris base (pH 10.5) in a shakerfor 20 min. Finally, aliquotswere transferred to a96-well plate and read ina Titertek Multiscan plate reader (Titertek Instruments, Inc., Huntsville, AL) at492 nm (laboratory 3). Inall threelaboratories the mean cell numbers from eachexperi- ment were normalized to the steroid-free control cultures to correctfor differences in the initialseeding density. Datarepre- sent themeanandarepools of eitherone (laboratory 3), at least two(laboratory 1),or at least three (laboratory 2) independent experimentsrun induplicate.

Transient GeneExpressionAssayin MCF-7Cells

(Laboratory

7). MCF-7cells obtained from the Breast Cancer Task ForceCell Culture Bank (Mason Research Institute, Worcester, MA) (passage298 to 310) were propagated in DMEwithout phenol redsupplementedwith 1% CD- treatedFBS, 64 pg/mlGentamicin(Gara- mycin, Schering-Plough, Madison, NJ), 2.5 mMglutamine, and 6 pg/liter, insulin andtransfectedas describedbyJorgensen and Autrup (50,51) using a chimeric reporterconstructcontainingone ERE in front of the thymidine kinase

(tk)

pro- moterand the chloramphenicol acetyl- transferase (CAT) gene

pERE-tk-cat

(52).

Upontransfection the cells were treated for 48 hr with solvent (0.1% ethanol [96%, Merck, Darmstadt, Germany]), 10 nM 17,-estradiol (E 8875, Sigma Chemical), orthe test chemicals. The test chemicals weretested at the highest nontoxic concen- trationasdeduced by a nonradioactive cell proliferation/cytotoxicity assay (Promega, Madison, WI). CAT activities were normal- ized to transfection efficiency and protein content as described by Jorgensen and Autrup (50,51). Preliminary results indi- catedthat the use of CD-treated human serum and CD-treated FBS give similar results. Data are expressed as the mean±SD and are a pool of at least nine samples from three independent experiments.

Recombinant Yeast Estrogen Screen Using HumanEstrogen Receptor(Labora- tories4and6).Yeast (Saccharomyces) sta- bly transfected with the human estrogen receptor(hER) gene and expression plas- mids carrying an ERE and the reporter gene lacZencoding the enzyme ,-galactosidase was used and the assay performed as describedbyRoutledge and Sumpter (27).

The testchemicals were serially diluted in absolute ethanol and 10plof each concen- tration wastransferredto a96-welloptically flat-bottom microtiter plate (Titertek) and allowedtoevaporateto dryness on the assay plate.Aliquots (200

pl)

of medium contain- ingrecombinant yeast andthe chromogenic substratechlorophenol

red-P-D-galactopyra-

noside(Boehringer Mannheim, East Sussex, UK) were thendispensed to eachsample well. Details of preparation of medium componentsarediscussedin Routledgeand Sumpter(27). Eachplatecontainedatleast one rowof blanks (assay medium only) as wellas astandardcurvefor

17p-estradiol.

In laboratory 4, absorbance at 540 nm was measured after 72-hr incubation usinga Titertek MultiscanMCC/340 platereader (Titertek). Inlaboratory6, absorbance at 550nmwas measuredafter 72- and93-hr incubation using an Anthos 2010 plate reader (Anthos Labtec Instruments, Salzburg,Austria). Inlaboratory 6, testing ofthechemicalswasrepeated by addingthe chemicalsdirectlytothe medium insteadof evaporating them todrynesson theplate.

This repetition was to investigate if this modification of theprocedurehadany effect on the response obtained. Data represent mean values from a single experiment carriedout induplicate.

RecombinantYeastEstrogen Screen UsingHuman EstrogenReceptor and Yeast Strain DY150 (Laboratory 10).

The yeaststrainDY150(MATa ura3-1 leu

ANDERSEN ET AL.

2-3 112 trp 1-1 his 3-11 15 ade 2-1 can 1-100) contains the yeast expression plas- mid containing hER (YEPKB1) and the estrogen-sensitive LacZ reporter plasmid (2ERE-LacZ). This strain was grown overnight at 30°C in synthetic medium, supplementedwith uracil and tryptophan, in 2 ml cultures. The next day, 25

Pl

of the overnight culture was diluted into 975 plfresh medium andgrown overnight (18 hr) with 1

pl

of the various stock solutions oftestchemicals corresponding to a final concentration of10pM forthe testchem- icals and 10 nM for 17J3-estradiol. For 3- galactosidase assays, theyeast cells were collectedbycentrifugation, resuspended in 700 pl Z-buffer (60 mM Na2HPO4, 40 mM NaH2PO4, 10 mM KCl, 1 rmM MgSO4, 35 mM 3-mercaptoethanol), and permeabilized by the addition of 6

pl

CHCI3 and 4

pl

0.1% sodium dodecyl sulfate (SDS) followedby vortexing for25 sec. The reactions were equilibrated at 30°C for 10 min, then 160

pl

o-nitro- phenyl-D-galactopyranoside (ONPG) (4 mg/ml in Z-buffer) was added and the reactions returned to 30°C forbetween 5 and 60 min. The reactions were termi- nated by the addition of 400

pl

1 M NaCO3, the cell debris was removed by centrifugation, and the absorbance at 420 nM measured (A420). The growth ofthe yeast strains was monitored by measuring the absorbanceat600 nM (A600). Miller units weredetermined using the following formula: [A420/(A600 of1/10 dilution of cellsxvolume of culturexlength ofincu- bation)]x1000. The data represent mean from three independent experiments with determinations intriplicates.

Recombinant YeastEstrogen Screen UsingRainbow TroutEstrogen Receptor (Laboratory5). Saccharomyces cerevisiae (strain BJ-ECZ)werestably transfectedwith the rainbow troutestrogen receptor (rtER) gene andan estrogen-responsive reporter genecontainingtwoERElinkedtotheyeast CYC1 promoterlocated upstream oftheE.

coli gene for

f-galactosidase

(lacZ) as described byPetitetal. (53). The cells were grown inliquid culture intheabsence (neg- ative control) orpresence (positivecontrol) of 10 nM

17p3-estradiol

ortestchemicals at 10 PM (1:1000 dilution of thestocksolu- tionsdirectly in the yeastculture media) for 4 hr at28°C. Cells wereharvested, lysed, thecell densitywas determinedat 600 nm, and the

f3-galactosidase

activitywas mea- sured at420 nm usingONPGas substrate.

The data represent the mean from three replicatesinasingleexperiment.

Vitellogenin Production inJuvenile Rainbow Trout in Vivo (Laboratory9).

Juvenile rainbowtrout (46 to 174g),

kept

in steel tanks ata photoperiod of 12 hr light:12hr dark, were randomly assorted into experimental groups ofsixfish, anes-

thetized with 2-phenoxyethanol, and injected ip with asingle dose (adjusted to 1 ml/kg)of thecompoundstobe tested. The compounds weredissolvedin 50%ethanol, 99%ethanol,orpeanutoil andinjected in the concentrations shown inFigure2.Apre- exposure blood samplewastaken on day 0 and afinal blood sampleon day9, and the vitellogenin level in theplasma was mea- sured byadirect sandwich enzyme-linked immunosorbent assay(ELISA).

For the ELISA, purifiedvitellogenin for antibody production and standards was obtained from 1713-estradiol-treated rainbow trout. Protein for immunization of rabbits to raise polyclonal antibodies against vitellogenin was obtained by precipitation oftroutserum with EDTA and MgCl2 [modified from Wiley et al.

(54)]. Vitellogenin usedfor standards and affinity columns was purified by gel filtration (SephacrylS300HR. Amersham- Pharmacia-Bioteck) followed by anion exchange chromatography (DEAE Seph- acel, Amersham-Pharmacia-Bioteck).

Antibodies against vitellogenin were obtained from ammonium sulfate-pre- cipitated rabbit antiserum afteraffinity chromatography on a CNBr-activated Sepharose 4B column (Amersham- Pharmacia-Bioteck) coupled with rainbow trout vitellogenin. The specificity of the antibodies was verified byWestern blot- ting of trout plasma after native- and SDS-polyacrylamide gel electrophoresis (PAGE). The antibodies were used for constructing a direct sandwich ELISA.

Microtiter plates were coated withspecific antibodies and blockedwith BSA (3%).

Vitellogenin standards or samples were added followed by incubation with horse- radish peroxidase-coupled antibodies. The color development after adding enzyme substrate (OPD) was monitored at 490 and 650 nm. The resulting assay had a lin- earrange of 5 to 50 ng/mland a detection limit of 500 ng/ml serum. The concentra- tion ofvitellogenin in serum at day0 and day 9 wasmeasured insamples from each fish andthe difference invitellogenin cal- culated. In cases where the vitellogenin levels werebelow thedetection limitof the assay, the concentration ofthe samplewas set to the detection limit of 0.5 pg/ml.

The data express the mean increase in

serum

vitellogenin

for the six fish in the

experimental

group.

Results

All test chemicals were tested in a blind fashion,and theidentitiesofthechemicals

werenotrevealed until all results had

been reported

^to the project coordinator. All laboratories also tested theirown positive control (1

7j-estradiol)

anda

hormonle-

free control. After reporting, all datawere

recalculated in reference to the positive controlfrom thetest

panel

^{to ensure}com-

parability

ofthe results. Some laboratories examineda range ofconcentrations; other laboratorieschoseto use

only

a

single

coIn- centration of the test chemicals in accor-

dance with their rouitine procedure for short-termtests.

For each in vitroassay, the response to achemical was

judged

as

fully

estrogenicif the responsewas >75%,

partially

estrogenic if the responsewas25to75%,

weakly

estro-

genic ifthe responsewas 10 to25%, and negativeifthe responsewasbelow 1O0%of the response induced

by 17f3-estradiol.

In the in vivoassay in rainbow trout, the

increaseinvitellogenin productionwassteep upon stimulation with the three potent estrogens, 1

7p-estradiol,

17oc-ethynylestra-

diol, and DES. Because the maximal response oftheindividual ^test

compounds

inthis assayisunknownanddifferent

single

doses of thechemicalswereused for testing, itis notpossible tocompare the responses

directly

or to

judge

the potencies of the chemicalsas intheinvitroassays.

1

7p3-Estradiol

and 17(x-ethynylestradiol exhibited strong estrogenic activity in all assays (Figures2-7 and Table 3). DES induced apartialestrogenic responsein the reporter geneexpression assay in MCF-7 cells and in the DY150 yeast assay but induced a full estrogenic responsein all other assays. The relativebinding affinities (RBA) of17cx-ethynyl estradiol and DES were

higher

than that for 17f-estradiol in bothbindingassays. Inonelaboratoryusing the MCF-7 cell proliferation assay, the responsecurve for 17f-estradiolshowed a

decrease atconcentrations above 0.0001

pM,

indicatingatoxicresponse.Theknown cytotoxiccompound colchicineinducedno estrogenic response inanyofthe assays but was

clearly

toxic inmostassays.

The antiestrogens tamoxifen and ICI 182.780bind strongly tothe recombinant hER from MCF-7 cells with affinities similar to 17f-estradiol. ICI 182.780 also binds strongly to ER from rabbit uterus, whereas thebinding affinity of tamoxifen

Environmental HealthPerspectives *Vol 07 Supplement * Februmy 99' 94

Table 3. Relative binding affinities of the test chemicals in the two binding assays.

IC50,M RBAa IC50,M RBAa

recombinant recombinant rabbit rabbit

Test compound hER hER uterus ER uterus ER

17(1-Estradiol 1.45x 10- 1 2.0x10-11 1

17[-Ethynyl

estradiol 6.7x10-10 2.2 2.0 x10-12 10

DES 2.5x10-'0 5.8 7.0x10-15 2857

Tamoxifen 2.6x10-9 0.6 1.2x107 1.7x104

ICI 182.780 3.6x109 0.4 4.0x10-12 5

Testosterone >2x10 - >1.0x 10 -

BisphenolA 1.1x104 1.3x104 1.6x104 1.3x105

BisphenolAdimethacrylate >2x104 - 4.3x10 4.7x106

4-r-OP 4x104 3.6x104 >1.0x 10- -

4-n-NP 4.3x104 3.4x104 1.8x104 1.1x10-5

NP12E0 5.7x10- 2.5x10- >1.0x10-5 ^-

BBP 1.2x104 1.2x104 >1.0x105 -

DBP >2x104 - >1.0x105 -

Methoxychlor >2x104 - 6.5x104 3.1 x 107

o,p'-DDT 5x10-7 2.9x103 3.4x10 5.9x 10

pJp-DDE

1.6x10C 9.1 x10 >1.0x10-5 -

Endosulfan 1.3x10- 1.2x104 >1.0x105 -

Chlormequat chloride 5.6x10-5 2.6x10 >1.0x10-5 ^-

Colchicine >2x10-4 - >1.0x10 ^-

Ethanol NC - NC -

Abbreviations:NC,nocompetitionfor the binding of173-estradiol.RBA, relative binding affinities. 'RBA wascal- culatedastheratio between thebinding affinity(IC50) of 17fi-estradiol and the binding affinity of the test com- poundtothe ER inthe twobinding assays.

Ef E 1E2l alE

CD

CD1t

A B C 0 E F 6 H I J K L M N O P Q R S T

Tetcompounds

Figure2 Increase invitellogeninproductioninjuvenilerainbowtrout9daysafterasingle intraperitoneal injectionof thetestcompoundsatthe doses indicated. The increase invitellogenin productionwascalculatedastheserumcon- centration atday 9 minustheserumconcentrationatday0.A, 17-estradiol (0.5 mg/kgin50%ethanol);B, 17a- Ethynylestradiol(5mg/kg in peanutoil); C,DES(5 mg/kgin50%ethanol);

0,

tamoxifen(50 mg/kgin99%ethanol); E, ICI 182.780(50 mg/kgin99%ethanol);F,testosterone(not investigated);G,bisphenol A(50mg/kg in 50%ethanol);

H,bisphenol A dimethacrylate(50 mg/kgin peanutoil);I,4-n-octylphenol(50 mg/kgin peanutoil);J.4-nnonylphenol (50 mg/kgin peanutoil); K, nonylphenol dodecylethoxylate(50 mg/kgin50%ethanol);L,benzylbutylphthalate (500 mg/kg in peanutoil); M, dibutylphthalate(500 mg/kgin peanutoil); N, methoxychlor (100 mg/kgin peanutoil); 0,

oJp-

0DT(50mg/kg in peanutoil);P.p4p-DDE(100 mg/kgin peanutoil); aendosulfan(5 mg/kginpeanutoil); R.chlorme- quatchloride(250 mg/kginpeanutoil);S,colchicine(1 mg/kginpeanutoil); T,ethanol(1mI/kg).

for this receptorwasonly 1/6000th the bindingaffinityof

17f-estradiol

(Table 3).

In theproliferation assayinMCF-7cells, tamoxifen inducedanestrogenic response inonelaboratory (77%of the maximum responseinducedby

17[1-estradiol),

aweak responseinanotherlaboratory (21% ofthe 17,-estradiol response),andnoresponsein

thethirdlaboratory using this assay.ICI 182.780 inducednoestrogenic activityin anyofthe three laboratories using theE- SCREEN assay(Figure 3).Reporter gene expressioninMCF-7 cellswasbelow the hormone-freecontrol for both antiestrogens (Figure 5).Inyeast,ICI 182.780 exhibited filllestrogenic activityinall assays (Figures

4,7)

except

one(laboratory 10) using the yeast strain DY150 (Figure 5). Tamoxifen responded as a partial estrogenic agonist in all the yeast assays, though only weakly in yeast DY150. In Saccharomyces expressing thehER, a maximal response was observed attamoxifen concentrations just below 1

pM,

whereas the response at higher con- centrations wasdecreased because of toxic- ity (Figure4). Tamoxifen, but not ICI 182.780, induced vitellogenin production in rainbow trout. However, the level of vitellogenin produced was only 0.04% of the level inducedby17restradiol at 1%of the dose of tamoxifen (Figure 2).

Althoughno ERbindingwas registered inthe binding assays (Table 3), testosterone exhibited full estrogenic activity in the proliferationas well asthe reporter gene expression assays in MCF-7cells (Figures 3,5). Maximalproliferationwasinduced at 1/1000th to 1/10000th the

17p-estradiol

concentrationinducing maximal response.

Testosteroneinduced a weak reporter gene activity inSaccharomyces expressingthe hER butonly at concentrations above 10

PM

(Figure 4),while no response was induced in anyofthe other yeast assays (Figures 6,7) inwhich onlyone con- centration of10 pM testosterone was tested.Testosteronewasnotinvestigatedin the vitellogenin assay in rainbow trout becausewe wereunabletoobtain sufficient quantities ofthechemical because of legal restrictionsfor importing steroidhormones.

Inthebindingassay usingrecombinant hER, o,p'-DDTwas the most effective of the environmental chemicals testedin reducing the binding of

[3H] 17p-estradiol;

binding affinitieswere 1/350th ofthe binding affinity of 17,B-estradiol (Table3).

The binding affinityofp,p'-DDEwas considerably lower, i.e., approximately 1:10,000 comparedtothebindingaffinity of17-aestradiol. Also,4-n-OPand 4-n-NP had similaraffinities fortheERofapproxi- mately 1:3000 that of

17p-estradiol,

whereas theratio forbisphenolA, BBP, and endosulfanwas about 1:8000. Both NP12EO and chlormequat chlorideonly barely reducedthe bindingof

[3H] 17P-

estradioltohER withbinding affinities of approximately 1/40,000th of thebinding affinityof

17p-estradiol.

Bisphenol A

dimethacrylate, DBP, methoxychlor, and colchicine all had IC50values above the limitof200pMfornonspecific binding.

The bindingaffinities of theenviron- mental chemicals for the ERfrom rabbit uterus were ingeneral lower than for the hERfrom MCF-7 cells, and only five

17,B-Estradiol

29 _ _ f f

1E-13 1E-12 1E-11 1E-10 1E-09 1E-08 1E-07 2E-08 1E-00 1E-04 Mobrcon08n&~on

17a-Ethynylestradiol

8-O- ---i--

1E-13 1E-12 1E-11 1E-10 1E-09 1E-08 1E-07 1EGS 1E-05 1E-04 Modac02on0Un

DES

1E-13 1E-12 1E-11 1E-10 1E-08 1E-08 1E-07 12E08 1E-05 1E-04 Mo~arconcenrafOos

Tamoxifen

9

8- _ _ _

7- _ -7

-6

!L-- 7i

^-

'i

i/

o$- - 101

-1

BisphenolA

Bisphenol A dimethacrylate

1E4-9 1E-08 1E-07 1E-08 1E

Mod0cncenU

\

E-0

ICI 182.780

9-

8

9⁷

~Eg5-

n3-

Y

02

1E13 1E-12 1E-11 1E-10 12E08 1E-8 1E-07 12E-0 1E-05 1E-04 IE-09 1E-08 1E-07 1E2- 1E-05 0.0001 0.001

Mwolncen Mdoar hb

Figure3.Effects of thetestchemicalsonproliferationofMCF-7 cellsexpressedasfold increaseincellnumber above hormone-freecontrolas a

function of thelogconcentrationof thetestcompound.Thetestcompoundswereaddedtocellsgrowinginmedium supplemented with10%

CD-treated humanserum(laboratory1and3)or5%CD-treated FBS(laboratory 2) and incubated for 6 days. Laboratory1(m), laboratory 2 (0), and laboratory 3(A). (Continuedonnextpage)

Environmental HealthPerspectives*Vol 107, Supplement 1 ^-February 1999 ANDERSEN ETAL.

Testosterone

0,0001 ,0081 -.

i M i\

1 X X

i -1~

\

a# ,,,& a,., l...;.

0-4

-82

g

8

1₀

22

8

c9

A

3211-0

4

96 W'01

-0

4-NP-12E0

°1

i2-,

^{- 1}

1--

,c .- ...g... ...

1-0-- 1E00 1E007 1E.0 1E0 0,O01 0.001

0mebAi

BBP

1E.0

M ~

p,p'-DDE

1E.00 1E-0 10E- 1E.0 10E.0 0.0001 0.

Endosulfan

1E-0 1 IE-07 1E-E 10-0 0.0001 0.01

Mcnor*iv

Chlormequat

chloride

00a

1E.0 1E.0 1E.07 IE

Mo no⁰⁸n05I000^{1E0.0 O,ODDt 0,001}

Methoxychlor

0

S 32

i0

~. .. ..-1 ''''--.. -9ece.

10E- 1E0.0 10E07 10E. 10E.0 O.O01 0,001

MMa 0obn

v ....... i._Ii

1E409 10.00 1E7 10.0 10-0 0.0001 0.001

M mw*o

11 10 9

~E7- i4

2

0-

1E4- 0.0001 0.001 O0. 0.1

Conooh .,%

Is5

4 4aS

0

DBP

4

2

10

Is

1-

Colchicine

o,p'-DDT Ethanol

Figure3.Continued.

10 9

a

il 1A

If -Q0v-1

9

27

Z

4

f

T F

^{t _r}

1 1

91

a

-1

t-j

^Ra

-a3