biotransformation of the antiestrogen clomiphene to chemically … · animals. the female...

6
[CANCER RESEARCH 47, 4015-4019, August 1, 1987) Biotransformation of the Antiestrogen Clomiphene to Chemically Reactive Metabolites in the Immature Female Rat1 Peter C. Ruenitz,2 Richard F. Arrendale, Garry D. George, Carolyn B. Thompson, Corwin M. Mokier, and Nitin T. Nanavati College of Pharmacy, University of Georgia, Athens 30602 [P. C. R., G. D. G., C. B. T., C. M. M., N. T. N.], and Tobacco Safety Unit, Agricultural Research Service, United States Department of Agriculture, Athens 30613 [R. F. A.], Georgia ABSTRACT Novel metabolites of clomiphene (CLO), an antiestrogen effective in experimental breast cancer models, were characterized in studies using immature female rats. After i.p. administration, CLO was eliminated unchanged in feces and as two components which were chromatographi- cally identical to synthetic CLO analogues bearing a m-methoxy-p- hydroxyphenyl (guaiacol) moiety in place of one or the other of its phenyl rings. These components were also found in liver tissue. In the presence of liver microsomal homogenate, CLO underwent /»-hydroxylation of either of its phenyl rings, affording 4-hydroxy-CLO and 4'-hydroxy- CLO. These in turn underwent liver microsomal mediated conversion to the respective guaiacol metabolites. 4'-Hydroxy-CLO and its 3'-methoxy analogue, but not positional isomers of these, had arylating ability as shown by chemical and spectral studies, apparently due to spontaneous conversion to electrophilic allene-quinones. Reproductive tract abnor malities produced in neonatal rats by CLO were suggested not to be mediated via such metabolites, since similar such effects were caused by 4'-fluoro-CLO. However, the latent arylating potential of the 4'-hydrox- ylated metabolites of CLO suggests that these compounds may be useful in biochemical studies of breast cancer cell growth inhibition. INTRODUCTION The ability of CLO3 (Fig. 1) and related triarylethylene an- tiestrogens to antagonize the growth-promoting effect of estro gens in target tissues (1) has focused attention on the potential application of these compounds in estrogen receptor positive breast cancer treatment and prevention (2, 3). The constituent isomers of CLO have been shown to suppress the proliferation of cultured human breast cancer cells (4) and to inhibit the growth of chemically induced breast cancer in the rat (5). The immature female rat has served as the most widely used model for whole animal studies of antiestrogen action. In this species, the effects of triarylethylenes other than CLO have been suggested to be mediated in part via metabolites structur ally analogous to 4-HC (6-9). Although CLO was efficiently converted to 4-HC in vitro, relatively low levels of this were found in liver and uterine tissue of immature rats given CLO (10). Instead, one or more metabolites of unknown composition accompanied CLO in these tissues. We wanted to identify and characterize these metabolites chemically and spectrally and to establish a stepwise basis for their formation based on known routes of biotransformation of structurally related drugs. Received 2/13/87; revised 5/5/87; accepted 5/11/87. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1This research was supported by National Cancer Institute Grant CA 28928. 2To whom requests for reprints should be addressed. 3 The abbreviations used are: CLO, clomiphene, (E,Z)-2-[4-( 1,2-diphenyl-2- chlorovinyl)phenoxy]ethyl-/V,/V-diethylamine; 4'-FC, 4'-fluoroclomiphene; 4'- MC, 4'-methoxyclomiphene; 4'-HC, 4'-hydroxyclomiphene; CG', 3'-methoxy- 4'-hydroxyclomiphene; iso-CG', 3'-hydroxy-4'-methoxyclomiphene, 4-HC, 4- hydroxyclomiphene; CG, 3-methoxy-4-hydroxyclomiphene; TLC, thin layer chro- matography; TV retention time; FAB. fast atom bombardment; TMS, trimethyl- silyl. MATERIALS AND METHODS Chemicals and Biochemicals. CLO citrate was purchased from Sigma Chemical Co., St. Louis, MO. Clomiphene /V-oxide, 4-HC, deethyl- clomiphene, and 4'-MC were available from previous studies (11). CG, 4'-FC, and 4-hydroxy-4'-FC were prepared by adaptation of methods reported in detail by us (12). The preparation of 4'-HC, CG', and iso- CG' will be reported later. The structures of these compounds are summarized in Fig. 1. Catechol O-methyl transferase, NADP, and all other biochemicals were obtained from Sigma. Radiochemicals. The preparation of [I4C]CLO (0.69 mCi/mmol, labeled at the carbon bearing the chloro substituent) and \\-cthyl~l -*\\\ CLO (16.3 mCi/mmol) has been described (13, 14). \S-methyl-3H]-S- adenosyl-L-methionine (12.2 Ci/mmol) was obtained from New Eng land Nuclear Research Products, Boston, MA. Chromatography. TLC was carried out using 20- x 20-cm plastic backed sheets coated with Silica Gel 60 F2S4(EM Reagents Catalog 5775). Plates were developed using chloroform:methanol:28% aqueous ammonia [90:10:0.5, by volume (Solvent 1); 95:5:0.5, by volume (Sol vent 2)]. Rf values of standard compounds determined under these conditions are listed in Table 1. Reversed phase high performance liquid chromatography was carried out using an Alltech 25-cm x 4.6- mm inside diameter 10-jim ( ',» column which was eluted with 50 mM NaH2PO4 buffer, pH 4.8:acetonitrile (30:70, v/v) at a rate of 2.0 ml/min. Eluent was monitored at 280 nm. Tg values of standard compounds are listed in Table 1. Liver micro somal extract concentrates were reconstituted in 25-50 ><1 of methanol, centrifugea for 10 min at 200 x g, and filtered through 0.2-fim nylon syringe filters prior to chromatography Mass Spectrometry. Direct insertion probe electron ionization mass spectrometry and coupled gas-liquid chromatography-electron ioniza tion mass spectrometry were carried out using a Hewlett-Packard 5985B coupled gas-liquid chromatography-electron ionization mass spectrometry system. Capillary gas-liquid chromatography conditions were as follows: 30-m x 0.3-mm inside diameter fused Silica SE-54 capillary column; carrier gas, helium, 40 cm/s linear velocity; temper ature program, 100'C for 1 min, 100-300'C at 6°C/min; cold on column injection (15). Samples were dissolved in pyridinc:.V.W- bis(trimethylsilyl)trifluoroacetamide (1:1) and heated for 30 to 45 min at 76°Cprior to injection of \-2-n\ aliquots. The capillary gas-liquid chromatography column was connected to the mass spectrometer via an open split interface (16). The mass spectrometer was operated at an ion source temperature of 200°C, an electron multiplier voltage of 2400 V, and a scan range of 40-600 amu. Conditions for the direct insertion probe/mass spectrometry analysis were 2 min at ambient temperature and ambient temperature to 250*C at 20°C/min.Spectra were recorded at a rate of 24/min. Direct inlet probe FAB mass spectra were obtained under conditions described previously (17). Animals. The female Sprague-Dawley rats (neonatal and 20-25 days old) used in these studies were obtained from our breeding colony. After weaning, animals were maintained on a normal laboratory diet. In one experiment, hepatic drug metabolizing enzymes were induced by i.p. administration of 80 mg/kg (about 4 mg/animal) of sodium phénobarbitalin 0.25 ml of 1.15% KCI once a day for 3 days, culmi nating 24 h prior to sacrifice. Preparation of Liver Microsomal Fraction. Microsomal suspension was prepared and analyzed for protein concentration as described (10), using livers from 4-6 animals. Microsomal pellets were rehomogenized in 10 mM Tris buffer, pH 7.4, containing 1 mM EDTA and 20% (v/v) glycerol. 4015 on April 29, 2021. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Upload: others

Post on 16-Nov-2020

1 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Biotransformation of the Antiestrogen Clomiphene to Chemically … · Animals. The female Sprague-Dawley rats (neonatal and 20-25 days old) used in these studies were obtained from

[CANCER RESEARCH 47, 4015-4019, August 1, 1987)

Biotransformation of the Antiestrogen Clomiphene to Chemically ReactiveMetabolites in the Immature Female Rat1

Peter C. Ruenitz,2 Richard F. Arrendale, Garry D. George, Carolyn B. Thompson, Corwin M. Mokier, and

Nitin T. NanavatiCollege of Pharmacy, University of Georgia, Athens 30602 [P. C. R., G. D. G., C. B. T., C. M. M., N. T. N.], and Tobacco Safety Unit, Agricultural Research Service,United States Department of Agriculture, Athens 30613 [R. F. A.], Georgia

ABSTRACT

Novel metabolites of clomiphene (CLO), an antiestrogen effective inexperimental breast cancer models, were characterized in studies usingimmature female rats. After i.p. administration, CLO was eliminatedunchanged in feces and as two components which were chromatographi-cally identical to synthetic CLO analogues bearing a m-methoxy-p-

hydroxyphenyl (guaiacol) moiety in place of one or the other of its phenylrings. These components were also found in liver tissue. In the presenceof liver microsomal homogenate, CLO underwent /»-hydroxylation ofeither of its phenyl rings, affording 4-hydroxy-CLO and 4'-hydroxy-

CLO. These in turn underwent liver microsomal mediated conversion tothe respective guaiacol metabolites. 4'-Hydroxy-CLO and its 3'-methoxy

analogue, but not positional isomers of these, had arylating ability asshown by chemical and spectral studies, apparently due to spontaneousconversion to electrophilic allene-quinones. Reproductive tract abnor

malities produced in neonatal rats by CLO were suggested not to bemediated via such metabolites, since similar such effects were caused by4'-fluoro-CLO. However, the latent arylating potential of the 4'-hydrox-

ylated metabolites of CLO suggests that these compounds may be usefulin biochemical studies of breast cancer cell growth inhibition.

INTRODUCTION

The ability of CLO3 (Fig. 1) and related triarylethylene an-tiestrogens to antagonize the growth-promoting effect of estrogens in target tissues ( 1) has focused attention on the potentialapplication of these compounds in estrogen receptor positivebreast cancer treatment and prevention (2, 3). The constituentisomers of CLO have been shown to suppress the proliferationof cultured human breast cancer cells (4) and to inhibit thegrowth of chemically induced breast cancer in the rat (5).

The immature female rat has served as the most widely usedmodel for whole animal studies of antiestrogen action. In thisspecies, the effects of triarylethylenes other than CLO havebeen suggested to be mediated in part via metabolites structurally analogous to 4-HC (6-9). Although CLO was efficientlyconverted to 4-HC in vitro, relatively low levels of this werefound in liver and uterine tissue of immature rats given CLO(10). Instead, one or more metabolites of unknown compositionaccompanied CLO in these tissues. We wanted to identify andcharacterize these metabolites chemically and spectrally and toestablish a stepwise basis for their formation based on knownroutes of biotransformation of structurally related drugs.

Received 2/13/87; revised 5/5/87; accepted 5/11/87.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1This research was supported by National Cancer Institute Grant CA 28928.2To whom requests for reprints should be addressed.3The abbreviations used are: CLO, clomiphene, (E,Z)-2-[4-( 1,2-diphenyl-2-

chlorovinyl)phenoxy]ethyl-/V,/V-diethylamine; 4'-FC, 4'-fluoroclomiphene; 4'-MC, 4'-methoxyclomiphene; 4'-HC, 4'-hydroxyclomiphene; CG', 3'-methoxy-4'-hydroxyclomiphene; iso-CG', 3'-hydroxy-4'-methoxyclomiphene, 4-HC, 4-hydroxyclomiphene; CG, 3-methoxy-4-hydroxyclomiphene; TLC, thin layer chro-matography; TV retention time; FAB. fast atom bombardment; TMS, trimethyl-

silyl.

MATERIALS AND METHODS

Chemicals and Biochemicals. CLO citrate was purchased from SigmaChemical Co., St. Louis, MO. Clomiphene /V-oxide, 4-HC, deethyl-clomiphene, and 4'-MC were available from previous studies (11). CG,4'-FC, and 4-hydroxy-4'-FC were prepared by adaptation of methodsreported in detail by us (12). The preparation of 4'-HC, CG', and iso-CG' will be reported later. The structures of these compounds aresummarized in Fig. 1. Catechol O-methyl transferase, NADP, and allother biochemicals were obtained from Sigma.

Radiochemicals. The preparation of [I4C]CLO (0.69 mCi/mmol,labeled at the carbon bearing the chloro substituent) and \\-cthyl~l -*\\\CLO (16.3 mCi/mmol) has been described (13, 14). \S-methyl-3H]-S-adenosyl-L-methionine (12.2 Ci/mmol) was obtained from New England Nuclear Research Products, Boston, MA.

Chromatography. TLC was carried out using 20- x 20-cm plasticbacked sheets coated with Silica Gel 60 F2S4(EM Reagents Catalog5775). Plates were developed using chloroform:methanol:28% aqueousammonia [90:10:0.5, by volume (Solvent 1); 95:5:0.5, by volume (Solvent 2)]. Rf values of standard compounds determined under theseconditions are listed in Table 1.

Reversed phase high performance liquid chromatography was carriedout using an Alltech 25-cm x 4.6- mm inside diameter 10-jim ( ',»column

which was eluted with 50 mM NaH2PO4 buffer, pH 4.8:acetonitrile(30:70, v/v) at a rate of 2.0 ml/min. Eluent was monitored at 280 nm.Tg values of standard compounds are listed in Table 1. Liver microsomal extract concentrates were reconstituted in 25-50 ><1of methanol,centrifugea for 10 min at 200 x g, and filtered through 0.2-fim nylonsyringe filters prior to chromatography

Mass Spectrometry. Direct insertion probe electron ionization massspectrometry and coupled gas-liquid chromatography-electron ionization mass spectrometry were carried out using a Hewlett-Packard5985B coupled gas-liquid chromatography-electron ionization massspectrometry system. Capillary gas-liquid chromatography conditionswere as follows: 30-m x 0.3-mm inside diameter fused Silica SE-54capillary column; carrier gas, helium, 40 cm/s linear velocity; temperature program, 100'C for 1 min, 100-300'C at 6°C/min; cold on

column injection (15). Samples were dissolved in pyridinc:.V.W-bis(trimethylsilyl)trifluoroacetamide (1:1) and heated for 30 to 45 minat 76°Cprior to injection of \-2-n\ aliquots. The capillary gas-liquid

chromatography column was connected to the mass spectrometer viaan open split interface (16). The mass spectrometer was operated at anion source temperature of 200°C,an electron multiplier voltage of 2400

V, and a scan range of 40-600 amu. Conditions for the direct insertionprobe/mass spectrometry analysis were 2 min at ambient temperatureand ambient temperature to 250*C at 20°C/min.Spectra were recorded

at a rate of 24/min. Direct inlet probe FAB mass spectra were obtainedunder conditions described previously (17).

Animals. The female Sprague-Dawley rats (neonatal and 20-25 daysold) used in these studies were obtained from our breeding colony.After weaning, animals were maintained on a normal laboratory diet.In one experiment, hepatic drug metabolizing enzymes were inducedby i.p. administration of 80 mg/kg (about 4 mg/animal) of sodiumphénobarbitalin 0.25 ml of 1.15% KCI once a day for 3 days, culminating 24 h prior to sacrifice.

Preparation of Liver Microsomal Fraction. Microsomal suspensionwas prepared and analyzed for protein concentration as described (10),using livers from 4-6 animals. Microsomal pellets were rehomogenizedin 10 mM Tris buffer, pH 7.4, containing 1 mM EDTA and 20% (v/v)glycerol.

4015

on April 29, 2021. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 2: Biotransformation of the Antiestrogen Clomiphene to Chemically … · Animals. The female Sprague-Dawley rats (neonatal and 20-25 days old) used in these studies were obtained from

CLOMIPHENE BIOTRANSFORMATION

CLO

4'-FC

4'-MC

Õ'-HC

4-HC

4-H-4'-FC

4-H-4'-NC

CO'

iso-CG'

CC

CI

x

H

H

H

H

4-hydroxy

4-hydroxy

4-hydroxy

H

H

3-methoxy-4-hydroxy

Y

H

4'-fluoro

4'-methoxy

4'-hydroxy

H

4'-fluoro

4'-methoxy

3'-methoxy-4'-hydroxy

3'-hydroxy-4'-methoxy

H

Fig. l. Structures of CLO and ring substituted analogues. For convenience,only the (ram structure is shown, even though each compound is composed ofvarying amounts of both geometric isomers.

Table I TLC and HPLC°of CW and related compounds

Dual entries indicate separation of geometric (/: and Z) isomers. Compoundson developed TLC plates were viewed under 254 nm wavelength light.

CompoundCLO4'-FC4--MCDECCOiso-CG'CG'4-HC4-H-4'-FC4-H-4'-MC4'-HCCNOTLCR,Solvent1 Solvent2'0.620.620.620.510.520.36,0.510.41,0.490.360.370.330.30,0.410.180.810.690.660.660.460.24HPLC7>*(min)22.813.77.36.6,

9.0>30

" HPLC, high performance liquid chromatography; DEC, deethylclomiphene;CNO, clomiphene yv-oxide.

" /.' isomers were analyzed by this procedure, except for 4 IK which was a

mixture of £and Z isomers.' Plates were developed serially twice using this solvent.

Assay for Conversion of Phenolic Substrates to Guaiacol Metabolites.Incubation mixtures were prepared as described (18), using 1 mg ofmicrosomal protein (unwashed) per 0.5 ml of incubation. Each incubation was run for 15 min; then 2.5 ml of ether were added. The mixturewas shaken for 10 min and centrifuged for 10 min at 200 x g, and 2.0ml of the upper phase were aspirated and concentrated. The residuewas redissolved in 50 /il of ethyl acetate and 40 ^I of this solution weresubjected to TLC (Solvent 1) after coapplication of the appropriatereference compound. After development of the chromatogram, thereference compound was located under UV at a wavelength of 254 nmand cut out, and comigrating 'II was eluted and counted in 8 ml of

Scintiverse I.

RESULTS AND DISCUSSION

In female rats, diethylstilbestrol and tamoxifen (each giveni.p.) are eliminated as metabolites bearing methoxy groups nextto preexisting and metabolically introduced phenolic hydroxyls,respectively (19, 20). By analogy, CLO would be eliminated asCG or CG', metabolites in which a 3-methoxy-4-hydroxyphenyl

(guaiacol) moiety replaces one or the other of its benzene rings.Synthetic CG comigrated with the major zone of metabolite

radioactivity in chromatograms of fecal extracts from animals

Table 2 Fecal elimination of radioactivity after administration ofrn"C]CLO citrate

A solution of 0.15 mg of [3HUC)CLO (0.27 nG of 3H, 0.26 i<Ci of UC)containing 0.1 % citric acid in a total volume of 0.25 ml was given i.p. to threegroups of four animals. Feces were collected and processed as described (10).Results are averages per animal. TLC was carried out using Solvent 2.

Time period 0-24 h 24-48 h 48-120 hCLO equivalents found (mimi) 266° 33 22

% of dose 70.5 8.7 5.8

TLC radioactivity (nmol) comigrating with

CLO 189.1 8.2 3.7DEC* 6.0 2.5 2.0

CG 13.0 6.9 7.54-HC 5.9 2.6 2.2CNO 4A 1.9 1.1" Chromatography of this extract using Solvent 1 resulted in relative amounts

of radioactivity associated with the standard compounds similar to those indicatedabove, except that associated with CG. The amount associated with this standarddecreased by 80% to the equivalent of 2.6 nmol; the remaining 10.4 nmol had anRr between 0.40 and 0.50.

* DEC, deethylclomiphene; CNO, clomiphene /V-oxide.

IO-

»5-

16

hrs. after ip dosing

24

Fig. 2. Levels of radioactivity comigrating with CLO (•),CG (•),and CG'(O) in chromatograms of liver tissue extracts. [MC]CLO was administered to threegroups of four animals as indicated in Table 2. Radioactivity in livers was extractedand analyzed as described (10), using Solvent 1 for TLC. Each point representsaverage levels per animal. Amounts of deethylclomiphene, 4-HC, and clomiphene/V-oxide were each <0.9 nmol/liver at all time points. Ratios of I4C associatedwith CG' or CG to that associated with CLO were similar using Solvent 1,Solvent 2, or benzene-triethylamine (85:15, v/v) for TLC of the 16-h extract.

receiving [3H,I4C]CLO (Table 2). However, TLC of selected

extracts using Solvent 1 instead of Solvent 2 resulted in comi-gration with CG of only 20% of the 3H/14C originally associatedwith this; the remaining 80% was found between 4-HC andCG. The 3H/I4C ratio in this zone was the same as that

associated with CG and CLO, indicating that this componentretained an intact diethylaminoethoxy side chain.

Subsequent to these experiments, synthetic CG' was preparedand found to cochromatograph with the fecal 3H/I4C distinctfrom 4-HC and CG. These results suggested that CG andparticularly CG' were significant fecal metabolites of CLO,

especially at longer time intervals after dosing. Chromatography of liver extracts from animals receiving [14C]CLO underconditions suitable for separation of CG and CG' indicatedqualitatively similar relative percentages of 14Cassociated with

these compounds (Fig. 2) compared to those found in fecalextracts.

In order to provide a mechanistic basis for the formation of4016

on April 29, 2021. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 3: Biotransformation of the Antiestrogen Clomiphene to Chemically … · Animals. The female Sprague-Dawley rats (neonatal and 20-25 days old) used in these studies were obtained from

CLOMIPHENE BIOTRANSFORMATION

Table 3 Characterization of phenolic metabolites ofCLO and 4'-MC by mass spectrometry

Metabolites were purified by TLC using Solvent 1 and eluted from silica gel with methanol unless otherwise stated.FAB MS:" MFT, m/z

Metabolite4-Hydroxy-4'-MC

4-HC*4'-HC fromCLO*'C4'-HC

from CLO*4'-HC from 4'-MC%100

10032

68100Found452.1911

422.1907422.1919418.2385

ND418.2400Calculated452.1992

422.1887422.1887418.2382418.2382M=CnHjoClNOj

C26H2,C1NO2C26H2,C1NO2C27H3,N03C2,H3,N03EIMS:M*,

m/z451

421417431

ND" FAB MS, fast atom bombardment mass spectrometry; EIMS, electron ionization mass spectrometry; ND, not determined.* Authentic 4-HC and 4'-HC each had molecular ions (M *) at m/z 421.'Coupled gas-liquid chromatography-electron ionization mass spectrometry (relative intensity, retention time) M * (TMS ether) m/z 493 (40%, 25.8 min), m/z

489 (60%, 25.3 and 26.0 min).d Purification was carried out using chloroform:ethanol:28% aqueous ammonia (90:10:0.5), and the metabolite was eluted with ethanol. Substitution of ethanol for

methanol in the TLC solvent did not affect R/ values of relevant external standards.

CG and CG' and to establish the intermediacy of 4- and 4'-

HC, respectively, in the formation of these metabolites, studieswith subcellular fractions of hepatic drug metabolizing enzymeswere carried out.

Incubation of CLO with liver microsomes resulted in theNADPH dependent formation of components with TLC Rrvalues and high performance liquid chromatography TKvaluesidentical to those of deethylclomiphene, 4-HC, 4'-HC, andclomiphene A'-oxide. Similar experiments with 4'-MC and 4'-FC afforded components identified mass spectrally and chro-matographically as the respective 4-hydroxy metabolites, butonly 4'-MC gave a metabolite chromatographically identical to4'-HC.

Several lines of evidence from mass spectral experimentswith liver microsomal phenolic metabolites (Table 3) showedthat 4'-HC, unlike 4-hydroxylated metabolites, underwent 68-

100% alcoholysis during Chromatographie purification. Themolecular ion in the FAB spectrum of metabolic 4'-HC derived

from CLO was accompanied by an ion with comparative nominal mass decreased by 4 units. The exact mass determinationshowed this to be due formally to replacement of the chlorosubstituent with a methoxy group. Electron ionization spectraexhibited only the ion corresponding to this product to theexclusion of the expected molecular ion, in contrast to authenticunchromatographed 4'-HC, in which only the molecular ion

(m/e 421) was seen. Use of ethanol rather than methanol forTLC purification resulted in an increase in the nominal massof the alcoholysis product from m/e 417 to m/e 431, sinceethoxy rather than methoxy addition had occurred. Mass chro-matograms of TLC purified metabolic 4'-HC TMS ether indi

cated that the chlorine bearing molecular ion, m/e 493, washomogeneous, but the ion due to methanolysis of this, m/e 489,was composed of two components. This suggested that methanolysis had taken place at two or more points in the structureof4'-HC.

While the above spectral data indicated the Chromatographieinstability of 4'-HC, no evidence for this was found by TLC.Synthetic 4'-HC was cleanly separated into two components,

thought to be its geometric isomers, each showing similar massspectral properties. Other 3'- and 4'-hydroxylated triarylethy-

lenes were similarly found to be readily separable into theirconstituent geometric isomers by TLC (21 ).

As shown in Table 4, the rate of liver microsomal formationof 4'-HC was about one-half that of 4-HC and substantiallygreater than the rate of covalent binding of MC to microsomal

protein. These results suggested that the chemical reactivity of4'-HC did not result in excessive covalent interactions with

protein, assuming that such interactions take place via thismetabolite at all. Phénobarbitalinduced the yV-deethylation ofCLO but did not induce its A/-oxidation, ring hydroxylation, orcovalent binding to protein. This last result differed from that

Table 4 Metabolism off'4C]CLO in the presence of liver microsomes from

uninduced and phénobarbitalinduced ratsIncubation mixtures were prepared and run as described (10), as were ether

extractions and TLC analyses of extract concentrates. Reactions were started byaddition of 50 nmol (34.5 nCi) of |14C]CLO citrate. Results are averages ±SD of

at least three experiments, corrected for NADPH independent metabolism bysubtraction of corresponding results obtained from NADPH generating systemabsent incubations.

pmol/mg of protein/15 min

I4C recovered as

CNO4'HC (both isomers)

4HCDECCovalently

bound"No

pretreatment9220

±370940 ±120

1990 ±80610 ±12079

±8Pretreatment

withphénobarbital136

±35787 ±19

1785 ±1895618±30484

±7* After ether extraction, protein was precipitated from each incubation mixture

by addition of 2.0 ml (4 volumes) of ethanol. The mixture was vortexed and thencentrifuged at 500 x g for 10 min. The supernatant was decanted and the pelletwas homogenized in 1.0 ml of ethanol. The mixture was centrifuged as before,and the supernatant was decanted. The pellet was reextracted in this way 6 moretimes. Then, it was solubilized in 0.5 ml of 2 N NaOH and "C was counted in 8

ml of Aquasol.

Table 5 Comparative rates of conversion ofestradiol, 4-HC, and 4'-HC torespective m-methoxy-p-hydroxy (guaiacol) metabolites in the presence

of liver microsomesIncubation mixtures contained 50 units of catechol O-methyl transferase and

8.2 pmol (100 nCi) of (1S'-mi7/i>'/-'H]-.$"-adenosyl-L-methioninein addition to othercomponents referred to in "Materials and Methods." Substrates were added in10 til of /V,jV-dimethylacetamide to give final concentrations of 50 U.M.Reconstituted incubation mixture extracts were subjected to reverse isotope dilution TLCanalysis using appropriate reference compounds. Data are averages ±SE of sixexperiments with each compound.

Rate of formation (fmol/mgof protein/15 min) in the

presence of

Product2-Methoxyestradiol

CGCG'0

minNADPH44+11

14±411 ±30.4

minNADPH991

±82285 ±66

84 ±6

observed for a related triarylethylene, chlorotrianisene, withwhich this was significantly induced by phénobarbitalpretreatment of male rats (22).

The results in Table 4 together with those in Table 5 confirmthat 4- and 4'-HC can serve as intermediates in the formationof CG and CG', respectively, from CLO. These in vitro resultssuggest CG formation to be favored over that of CG', but in

vivo the reverse was found. Other processes may differentiallyaffect the accumulation and further metabolism and dispositionof the hydroxylated intermediates in the whole animal in a waywhich favors hepatic accumulation of CG'.

The chemically reactive nature of 4'-HC, initially suggested

4017

on April 29, 2021. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 4: Biotransformation of the Antiestrogen Clomiphene to Chemically … · Animals. The female Sprague-Dawley rats (neonatal and 20-25 days old) used in these studies were obtained from

CLOMIPHENE BIOTRANSFORMATION

in mass spectral studies, was further confirmed by its ability toreact with 7-(p-nitrobenzyl)pyridine, a reagent for electrophiles(Fig. 3). CG' also reacted with 7-(p-nitrobenzyl)pyridine butpositional isomers of CG' and 4'-HC did not. These data, plusthose in Table 3, indicate that chemical reactivity requires a 4'-

hydroxyl group. A mechanism consistent with these findings isproposed in Fig. 4. Reaction with methanol could also occur atthe allenii- carbon to give an acid labile enol ether. We have notyet determined to what extent mild acid treatment of metha-nolysis products of CG' or 4'-HC results in replacement of

OCH, with OH.Substances containing quinone or quinone-type moieties have

been shown to have toxic effects (25-27). Our data suggestedthat a metabolite endowed with a latent quinone-type moiety,CG', accumulated in rats given CLO. CLO was previously

shown to cause reproductive tract abnormalities in neonatalrats (28) and in offspring of pregnant rats to which CLO wasgiven (29). In order to find out if this could be mediated via 4'-

0.5-

0.4-

O O3-

in

0.2-1

01

o.o80

nmolft of compound

I6O

Fig. 3. Comparative arylating efficiencies of CLO analogues and a nitrogenmustard. Solutions of 4-HC (A), 4'-HC (A), CG (•),CG' (O), iso-CG' (C), andbis(/j-chloroethyl)amine i •I were each heated at 100'C for 20 min in the presenceof excess -y-(p-nitrobenzyl)pyridine in 0.2 M acetate buffer, pH 4.6, and the extentof reaction was determined as reported previously (23, 24).

hydroxylation of CLO, we compared the effects of CLO withthose of 4'-FC, an analogue shown to be incapable of undergo

ing such a conversion. Results in Table 6 suggest similar inhibitory effects of these compounds on uterine growth. Also,animals in both drug treated groups showed poorly developedovaries and an increased incidence of ovarian and uterine abnormalities compared to these tissues obtained from drug freeanimals. Thus, the reproductive tract toxicity of CLO wouldappear on the basis of these results to be due to its inherentestrogenic effects (30) rather than to conversion to electrophilicmetabolites of the type postulated in Fig. 4. In humans, CLOproduces a variety of side effects (30) which have limited its useto short term therapeutic applications. These effects also havebeen attributed to unchanged CLO on the basis of its ability tointeract not only with estrogen receptors but also with dopa-minergic and muscarinic receptors (31, 32).

Biotransformation of antiestrogens often affords metabolitesthe effects of which equal or exceed the potencies of the antiestrogens from which they are derived (33). Thus, 4-HC wasshown to inhibit estrogen dependent uterine growth and MCF7 breast cancer cell proliferation more effectively than doesCLO, due to greatly increased affinity for estrogen receptors(12, 34). The present results show that 4'-HC and CG' are

additional metabolites of CLO that could contribute to theeffects of this antiestrogen. This is especially true with regardto CG', at least in the immature rat, since CG' appeared to be

the major metabolite of CLO in liver and, by inference, inuterine tissue (10) in this animal species.

Complete attenuation of human breast cancer cell growth byantiestrogens occurs at relatively high concentrations and appears to be mediated via estrogen receptors (35). Thus, therehas been an interest in estrogen receptor ligands endowed withchemically reactive moieties (36, 37). While 4'-HC and CG'

each had novel arylating ability, studies of the dependence ofthis on temperature and pH, and on reactivity with biologicalnucleophiles, in conjunction with characterization of the natureof interaction of 4'-HC and CG' with estrogen receptors, will

be required to assess the significance of this in vivo.

ACKNOWLEDGMENTS

FAB mass spectrometry was carried out at the Massachusetts Institute of Technology Mass Spectrometry Laboratory (supported by NIHDivision of Research Resources Grant RR 00317).

REFERENCES

Fig. 4. Plausible mechanism for formation of adducts of 4'-HC and CG'.

Dechlorination, which proceeded only when a hydroxyl group was positioned asshown, affords an electrophilic allene-quinone intermediate which reacts as indicated. Alternatively, reaction with nucleophiles. especially methanol, could occurat the allenic carbon to afford enol ethers.

Table 6 Comparative effects of CLO and 4'-FC on reproductive

trocídevelopmentNewborn female rats were divided into 3 groups of 14. Each animal in the first

group received O.I ml of corn oil s.c. Animals in the second and third groupswere each similarly given, in turn, 0.2 mg of CLO and 4'-FC in corn oil. After

100 days, animals were killed, and uterine and whole body weights were determined.

Uterine wt (mg)Body wt (g)Ratio''°P<

0.001."P<0.01.

,. Uterine wt(mg)Body

wt (g)Corn

oil561.6

±19.1238.6 ±4.3238.4 ±7.9100CLO337.4

±21.3°265.5 ±6.8*128.4 ±9.0"4'-FC356.9

±8.6°260.4 ±4.6*137.6 ±4.2°

1. Sutherland, R. I... and Jordan, V. C. (eds.). Nonsteroidal Antiestrogens.Molecular Pharmacology and Antitumor Activity. New York: AcademicPress, 1981.

2. Lippman, M. E. Endocrine responsive cancers in man. In: ¡.D. Wilson andD. W. Foster (eds.), Williams Textbook of Endocrinology, Ed. 7, pp. 1309-1319. Philadelphia: W. B. Saunders, Inc., 1985.

3. Cuzick, J., Wang, D. Y., and Bulbrook, R. D. Prevention of breast cancer.Lancet, A-83-86, 1986.

4. Murphy, L. C., and Sutherland, R. L. Antitumor activity of clomipheneanalogs in vitro: relationship to affinity for the estrogen receptor and anotherhigh affinity antiestrogen-binding site. J. Clin. Endocrino!. Metab., 57: 373-379, 1983.

5. Hubinont, P. O., Leroy, F., and Galand, P. (eds.). Basic Action of SexSteroids on Target Organs, pp. 274-279. Basel: S. Karger AG, 1971.

6. Borgna, J-L., and Rochefort, H Hydroxylated metabolites of tamoxifen areformed in vivo and bound to estrogen receptor in target tissues. J. Biol.Chem., 256: 859-868, 1981.

7. Robertson, D. W., Katzenellenbogen, J. A., Long, D. J., Rorke, E. A., andKatzenellenbogen, B. S. Tamoxifen antiestrogens. A comparison of theactivity, pharmacokinetics, and metabolic activation of the as and transisomers of tamoxifen. J. Steroid Biochem.. 16: 1-13, 1982.

8. Katzenellenbogen, B. S., Katzenellenbogen, J. A., Ferguson, E. R., andKrauthammer, N. Antiestrogen interaction with uterine estrogen receptors.Studies with a radiolabelled antiestrogen (CI 628). J. Biol. Chem., 253:697-707, 1978.

9. Hayes, J. R., Rorke, E. A., Robertson, D. W., Katzenellenbogen, B. S., and

4018

on April 29, 2021. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 5: Biotransformation of the Antiestrogen Clomiphene to Chemically … · Animals. The female Sprague-Dawley rats (neonatal and 20-25 days old) used in these studies were obtained from

CLOMIPHENE BIOTRANSFORMATION

Katzenellenbogen, J. A. Biological potency and uterine estrogen receptorinteractions of the metabolites of the antiestrogens CI 628 and U 23,469.Endocrinology, 108: 164-172, 1981.

10. Ruenitz, P. C, and Bagley. J. R. Comparative fates of clomiphene andtamoxifen in the immature female rat. Drug Metab. Dispos., 13: 582-586,

1985.11. Ruenitz, P. C. Rabbit liver microscmi.il metabolism of enclomiphene. Drug

Metab. Dispos., 9:456-460, 1981.12. Ruenitz, P. C., Bagley, J. R., Watts, C. K. W., Hall, R. E., and Sutherland,

R. L. Substituted-vinyl hydroxytriarylethylenes, l-[4-[2-(diethylamino)-ethoxy)phenyl]-l-{4-hydroxyphenyl)-2-phenylethylenes: synthesis and effectson MCF 7 breast cancer cell proliferation. J. Med. Chem., 29: 2511-2519,

1986.13. Ruenitz, P. C., and Bagley, J. R. The preparation of carbon-14 labelled

clomiphene and nitromiphene. J. Labelled Compd. Radiopharm., 20:1113-1120,1983.

14. Ruenitz, P. C., and Bagley, J. R. Reduction of ¡miniumethers with sodiumboro['H]hydride. Preparation of tritium labelled clomiphene and nitromiphene (CI 628). J. Labelled Compds. Radiopharm., 21:857-863, 1984.

15. Arrendale, R. F., and Chortyk, O. T. Construction and application of a coldon-column injection system for capillary gas chromatography. J. High Resolution Chromatog. Chromatog. Commun., 8:62-68, 1985.

16. Arrendale, R. F., Severson, R. F., and Chortyk, O. T. Open split interfacefor capillary gas chromatography/mass spectrometry. Anal. Chem., 56:1533-1537, 1984.

17. Martin, S. A., Costello, C. E., and Biemann, K. Optimization of experimentalprocedures for fast atom bombardment mass spectrometry. Anal. Chem., 54:2362-2368,1982.

18. Li, S. A., Klicka, J. K., and Li, J. J. Estrogen 2- and 4-hydroxylase activity,caicchi il estrogen formation, and implications for estrogen carcinogenesis inthe hamster kidney. Cancer Res., 45: 181-185, 1985.

19. Metzler, M. Metabolic activation of diethylstilbestrol: indirect evidence forthe formation of a stilbene oxide intermediate in hamster and rat. Biochem.Pharmacol., 24:1449-1453, 1975.

20. Fromson, J. M , Pearson, S., and Bramah, S. The metabolism of tamoxifen(I.C.I. 46,474). Part I. In laboratory animals. Xenobiotica, 3:693-709,1973.

21. Ruenitz, P. C., Bagley, J. R., and Mokier, C. M. Estrogenic and antiestrogenicactivity of monophenolic analogues of tamoxifen. (Z)-2-[p-l,2-diphenyl-l-butenyl)phenoxy)-JV,yV-dimethylethylamine. J. Med. Chem., 25: 1056-1060,1982.

22. Bulger, W. H., Juedes, M. J., and Kupfer, D. Covalent binding of chlorotrian-

isene (TACE) metabolite(s) to rat hepatic microsomal components. Fed.Prpc., 45: 572,1986.

23. Friedman, O. M., and Boger, E. Colorimetrie estimation of nitrogen mustardsin aqueous media. Anal. Chem., 33:906-910, 1961.

24. Epstein, J., Rosenthal, R. W., and Ess, R. J. Use of v(4-nitrobenzyl)pyridineas analytical reagent for ethylenimines and alkylating agents. Anal. Chem.,27:1435-1439, 1955.

25. Hinson, J. A., Pohl, L. R., Monks, T. J., and Gillette, J. R. Acetaminophen-induced hepatotoxicity. Life Sci., 29:107-116, 1981.

26. Smith, M. T. Quiñonesas mutagens, carcinogens, and anticancer agents:introduction and overview. J. Toxicol. Environ. Health, 16:665-672, 1985.

27. Lau, S. S., Monks, T. J., and Gillette, J. R. Identification of 2-bromphydro-quinone as a metabolite of bromobenzene and o-bromophenol: implicationsfor bromobenzene-induced nephrotoxicity. J. Pharmacol. Exp. 'Iher.. 230:

360-366, 1984.28. Clark, J. H., and McCormack, S. Clomid or nafoxidine administered to

neonatal rats causes reproductive tract abnormalities. Science (Wash. DC),797; 164-165, 1977.

29. McCormack, S., and Clark, J. H. Clomid administration to pregnant ratscauses abnormalities of the reproductive tract in offspring and mothers.Science (Wash. DC), 204:629-631, 1979.

30. Clark, J. H., and Markaverich, B. M. The agonistic-antagonistic propertiesof clomiphene: a review. Pharmacol. Ther., 15:467-519, 1982.

31. Hiemke, C., and Ghraf, R. Interaction of non-steroidal antiestrogens withdopamine receptor binding. J. Steroid Biochem., 21:663-667, 1984.

32. Ben-Baruch, G., Schreiber, G., and Sokolovsky, M. Cooperativity pattern inthe interaction of the antiestrogen drug clomiphene with the muscarinicreceptors. Mol. Pharmacol., 21: 287-293, 1982.

33. Jordan, V. C. Biochemical pharmacology of antiestrogen action. Pharmacol.Rev., 36: 245-276, 1984.

34. Ruenitz, P. C., Bagley, J. R., and Mokier, C. M. Metabolism of clomiphenein the rat:estrogen receptor affinity and antiestrogenic activity of clomiphenemetabolites. Biochem. Pharmacol., 32:2941-2947, 1983.

35. Sutherland, R. L., Watts, C. K. W., and Ruenitz, P. C. Definition of twodistinct mechanisms of action of antiestrogens on human breast cancer cellproliferation using hydroxytriphenylethylenes with high affinity for the estrogen receptor. Biochem. Biophys. Res. Commun., 140: 523-529,1986.

36. Leclercq, G., Devleeschouwer, V. and Heuson, J. C. Guidelines in the designof new antiestrogens and cytotoxic-Iinked estrogens for the treatment ofbreast cancer. J. Steroid Biochem., 19: 75-85, 1983.

37. Katzenellenbogen, J. A., Carlson, K. E., Heiman, D. F., Robertson, D. W.,' Wei, L. L., and Katzenellenbogen, B. S. Efficient and highly selective covalentlabeling of the estrogen receptor with ('Hjtamoxifen aziridine. J. Biol. Chem.,258: 3487-3495,1983.

4019

on April 29, 2021. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 6: Biotransformation of the Antiestrogen Clomiphene to Chemically … · Animals. The female Sprague-Dawley rats (neonatal and 20-25 days old) used in these studies were obtained from

1987;47:4015-4019. Cancer Res   Peter C. Ruenitz, Richard F. Arrendale, Garry D. George, et al.   Chemically Reactive Metabolites in the Immature Female RatBiotransformation of the Antiestrogen Clomiphene to

  Updated version

  http://cancerres.aacrjournals.org/content/47/15/4015

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  [email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

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

.http://cancerres.aacrjournals.org/content/47/15/4015To request permission to re-use all or part of this article, use this link

on April 29, 2021. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from