comparison of the mutagenicity and teratogenicity of ... · teratogenicity testing. the...

[CANCER RESEARCH 42, 3016-3021. August 1982]0008-5472/82/0042-OOOOS02.00

Comparison of the Mutagenicity and Teratogenicity of Cyclophosphamideand Its Active Metabolites, 4-Hydroxycyclophosphamide, PhosphoramideMustard, and Acrolein1

Barbara F. Hales2

Department of Pharmacology and Therapeutics, McGi/1 University, Montreal, Quebec, Canada H3G 1Y6

ABSTRACT

Cyclophosphamide must be metabolically activated to havemaximal mutagenic or teratogenic activity. The first step in thisactivation is hydroxylation to 4-hydroxycyclophosphamide; this

metabolite breaks down to form two cytotoxic metabolites,phosphoramide mustard and acrolein. In this report, the mu-tagenicity and teratogenicity of Cyclophosphamide, 4-hydro-peroxycyclophosphamide (which forms 4-hydroxycyclophos

phamide spontaneously in solution), phosphoramide mustard,and acrolein were compared. Mutagenicity was assessed usinga Salmonella typhimurium TA 1535 test system; teratogenicitywas studied in rats on Day 20 of gestation after intraamnioticinjection of drug on Day 13. The activation of Cyclophosphamide to mutagenic metabolites was dependent on the presenceof liver microsomes and reduced nicotinamide adenine dinu-cleotide phosphate while both phosphoramide mustard and 4-

hydroperoxycyclophosphamide were mutagenic without activation, with the latter being the most potent. The third metabolite, acrolein, was bacteriotoxic at low concentrations; it wasnot mutagenic in the absence and only very weakly mutagenicin the presence of liver microsomes. All four compounds testedwere teratogenic. The malformations produced by Cyclophosphamide, 4-hydroperoxycyclophosphamide, and acrolein included edema, hydrocephaly, open eyes, cleft palate, micro-

gnathia, omphalocele, bent tail, and forelimb and hindlimbdefects, whereas phosphoramide mustard produced only hydrocephaly and tail, forelimb, and hindlimb defects. 4-Hydro-

peroxycyclophosphamide and its breakdown product, acrolein,were more potent as teratogens than either Cyclophosphamideor phosphoramide mustard. Cyclophosphamide produced malformations in both the injected and contralateral uninjectedfetuses while the other three compounds all produced malformations only in the injected fetuses. Thus, the site of activationof Cyclophosphamide to a teratogen is probably maternal.Because acrolein plays a major role in the teratogenicity ofCyclophosphamide but is only weakly, if at all, mutagenic, theteratogenicity and mutagenicity of metabolites of this drug aredissociable.

INTRODUCTION

Cyclophosphamide is widely used as a cancer chemothera-

peutic agent and as an immunosuppressant. Its use is associated, however, with certain undesirable or toxic effects includ-

1This work was supported by the Medical Research Council of Canada (Grant

MA-7078) and the Conseil de la Recherche en SantÃ©du QuÃ©bec.Part of thismaterial has been published previously in abstract form (20).

2 Scholar of the Medical Research Council of Canada.

Received December 28, 1981; accepted May 3, 1982.

ing fetal malformations, secondary cancer, and cystitis (32,34, 35). Metabolic activation of Cyclophosphamide is requiredto produce both the therapeutic and the toxic effects (3, 5, 13,14, 21, 24, 26, 31, 36, 37, 39). Several attempts have beenmade to modify this drug or its metabolism to improve therelative ratio of desirable to undesirable effects (11, 15, 18,19, 22,33,36).

Activation of Cyclophosphamide is thought to occur predominantly in the liver and is catalyzed by the microsomal mixed-function oxidase system. The first step in the activation ofCyclophosphamide is C-4 hydroxylation to form the unstableintermediate, 4-hydroxycyclophosphamide (Chart 1). This me

tabolite, or its aldehyde tautorner, aldophosphamide, decomposes spontaneously to yield phosphoramide mustard andacrolein. Several investigators (6, 7, 23, 25, 37) have presented evidence that the selective therapeutic specificity ofCyclophosphamide is due to the intermediate, 4-hydroxycyclo-phosphamide-aldophosphamide. With respect to one of theundesirable toxic effects, the bladder toxicity or sterile hem-

orrhagic cystitis, it has been possible to demonstrate that theresponsible metabolite is acrolein (8, 10). Administration of athiol compound provides protection from the bladder toxicityof a high dose of Cyclophosphamide without affecting thecytotoxic specificity (11, 33). With respect to mutagenicity,Ellenberger and Mohn (12) found that, with the exception ofacrolein, all the metabolites of Cyclophosphamide tested weremutagenic. The identity of the metabolite(s) of Cyclophosphamide that is responsible for the teratogenic effects of this drugremains to be resolved. Evidence has, however, been obtainedin different laboratories that either phosphoramide mustard oracrolein is the "ultimate" teratogenic metabolite of Cyclophos

phamide (9, 27, 39).The objective of this study was to compare the mutagenicity

and teratogenicity of Cyclophosphamide with that of some ofits active metabolites: 4-hydroxycyclophosphamide; phosphor-amide mustard; and acrolein. Since 4-hydrcxycyclophospha-mide is unstable, we have used 4-hydroperoxycyclophosphamide; this compound is rapidly hydrolyzed to 4-hydroxycyclo

phosphamide in aqueous solution (Chart 1). In this report, it isdemonstrated that metabolically activated Cyclophosphamide,4-hydroperoxycyclophosphamide, and phosphoramide mustard are all mutagenic to Salmonella typhimurium TA 1535.The most potent mutagen is 4-hydroperoxycyclophosphamide.Cyclophosphamide, 4-hydroperoxycyclophosphamide, phosphoramide mustard, and acrolein are all teratogenic after Â¡n-traamniotic injection. Only 4-hydroperoxycyclophosphamide

and acrolein produce the same spectrum of malformations asCyclophosphamide; these compounds are at least 100 timesmore potent as teratogens than Cyclophosphamide. Therefore,it is possible to dissociate the mutagenic and teratogenic

3016 CANCER RESEARCH VOL. 42

on March 6, 2021. © 1982 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

Cyclophosphamide: Mutagenicity and Teratogenicity

OOH

CICH2CH2

>CICH2CH2

,NH CH4Â°xNO CH,

CICH2CHj /CICHjCHj

NH CH-X.

-CH,

LIVER

MICROSOMES

O2, NADPH

CYCLOPHOSPHAMIDE

Chart 1. Activation of Cyclophosphamide.

4-HYDROPEROXYCYCLOPHOSPHAMIDE

OH

It CHCICHjCH, M

CICH2CH2v

CH

4-HYDROXYCYCLOPHOSPHAMIDE

CICH2CH2 '

CICH2CH2

NH2

P= 0

O â€” CH -CH2â€”CHO

ALDOPHOSPHAMIDE

CH2=CH CHO

ACROLEIN

CICH2CH2 NH-

>-<='CICH2CH2 OH

PHOSPHORAMIDE MUSTARD

effects of metabolites of cyclophosphamide using its 2 breakdown products, phosphoramide mustard and acrolein.

MATERIALS AND METHODS

Chemicals. Cyclophosphamide was purchased from Koch-LightLaboratories Ltd., Colnbrook, England. 4-Hydroperoxycyclophos-phamide and phosphoramide mustard (ASTA-5317) were gifts from Dr.N. Brock, Asta-Werke, Bielefeld, Germany. Acrolein was purchased

from Eastman Chemical Co., Rochester, N. Y. Solutions of thesechemicals were prepared immediately prior to use. PhÃ©nobarbital waspurchased from Allen & Hanburys, Toronto, Ontario, Canada. Glucose6-phosphate, glucose-6-phosphate dehydrogenase, and NADP were

purchased from Sigma Chemical Co., St. Louis, Mo.Mutagenicity Testing. S. typhimurium strain TA 1535 (kindly pro

vided by Dr. Bruce Ames) was used to assay for mutagenic activity. TA1535 is a histidine-requiring auxotroph that is reverted to prototrophyby mutagens that cause DMA base-pair substitutions. The plate incor

poration assay was done according to the procedure of Ames ef al. (2)as described previously (21, 22). This mutagenicity assay was linearwith both the concentration of cyclophosphamide and the microsomalfraction (21). Maximal activation of cyclophosphamide to mutagenicmetabolites was obtained with 50 Â¿ilof the hepatic microsomal fractionfrom phenobarbital-pretreated rats; this is the enzyme preparation used

in these experiments.To evaluate bacteriotoxicity, tubes containing the designated drug

concentration (0.1 ml), microsomal activation system (0.5 ml), andbacteria (0.1 ml) were incubated for 60 min at 37Â°.Aliquots (0.1 ml) ofthe 10"3 and 10~4 dilutions were plated on complete agar for detection

of survival; colonies were counted after incubation for 24 hr at 37Â°.

Teratogenicity Testing. The fetotoxicity and teratogenicity of cyclophosphamide, 4-hydroperoxycyclophosphamide, phosphoramide mus

tard, and acrolein were tested by injection into the amniotic fluid ofembryos on Day 13 of gestation. Timed-gestation pregnant Sprague-

Dawley rats (225 to 250 g) were obtained from Charles River Canada,Inc. (St. Constant, QuÃ©bec,Canada). The day on which spermatozoawere found in the vaginal smear was considered Day 0 of pregnancy.Pregnant rats were housed on Beta chips in the Mclntyre Animal Centre(McGill University, Montreal, QuÃ©bec, Canada) and given Purina ratchow and water ad libitum. On Day 13 of gestation, the pregnant ratswere laparotomized under ether anesthesia and the uterus was exposed. Embryos in one uterine horn received an intraamniotic injection,using a 28-gauge hypodermic needle, of 0.9% 10 jtl of 0.9% NaCI

solution (controls) or drug dissolved in 10 /il of 0.9% NaCI solution.The uterus was repositioned in the abdominal cavity and the laparotomywas closed with nylon sutures in 2 layers. Rats were killed by decapitation on Day 20 of gestation. Fetuses were removed, examined forexternal malformations, blotted dry, and weighed.

Statistical Analysis. Mutagenicity data were evaluated by linearregression analysis and 2-way analysis of variance; teratogenicity datawere analyzed by the Mann-Whitney U test and the fetal weight data

by paired and unpaired i tests as appropriate. These procedures aredescribed by Snedecor and Cochran (38).

RESULTS

Mutagenicity. The relative mutagenicity of cyclophosphamide, 4-hydroperoxycyclophosphamide, phosphoramide mus

tard, and acrolein was assessed by their ability to revert S.typhimurium TA 1535 to histidine independence (Chart 2). Inthe absence of a microsomal enzyme-activating system, cyclo

phosphamide was only weakly mutagenic. After activation mediated by liver microsomes and a NADPH-generating system.

AUGUST 1982 3017



ÃŸ.F. Hales

DRUG:

S

y,

^ io. <i) *

O

i

2000160012008004000400800120016002OOO~â€”

- 4-HrDrÂ¿rÂ«rÂ°-"-SET----*

s0i---*-:filuP*_,-1*--Â¥*1_**-_--_-â€¢* * . Â¿Â¿.----*"200016001200800400040080012001600onnn

O

niai

o

lmw\-or

m

100 200 300 40O 500

CONCENTRATION: 25 50 100 250 500 5 10 25 50 100 25 50 100 250 500 0.001 0.01 0.1 1.0 10 50(^g/PLATE)

Chart 2. Mutagenicity of cyclophosphamide, 4-hydroperoxycyclophosphamide, phosphoramide mustard, and acrolein. Mutagenicity was measured using S.typhimurium TA 1535 in the presence and absence of an activating system, as described in the text. Columns, means for 3 experiments (n = 3); bars, S.E.;â€¢,bacteriotoxicity.

cyclophosphamide was mutagenic to S. typhimurium TA 1535.There was a linear relationship (correlation coefficient, 0.943)between the concentration of cyclophosphamide per plate andthe number of revertant colonies per plate.

Of the 3 metabolites of cyclophosphamide tested, 2 [4-hy-droxycyclophosphamide (4-hydroperoxycyclophosphamide)

and phosphoramide mustard] were mutagenic themselves; mutagenicity was not enhanced in the presence of liver micro-

somes. There was a linear relationship between the concentration of the compound per plate and the number of revertantcolonies for both 4-hydroperoxycyclophosphamide (range, 5to 50 ^g/plate; correlation coefficient, 0.935) and phosphor-

amide mustard (range, 25 to 500 fig/plate; correlation coefficient, 0.957). At a concentration of 100 Â¡ug/plate, 4-hydro

peroxycyclophosphamide was highly bacteriotoxic (Chart 3,<20% survival); phosphoramide mustard was less bacteriotoxic than 4-hydroperoxycyclophosphamide while activated

cyclophosphamide was not toxic to the bacteria.The mutagenic potencies of bioactivated cyclophosphamide,

4-hydroperoxycyclophosphamide, and phosphoramide mus

tard are compared in Chart 4. The relative mutagenicity ofthese 3 compounds, expressed in number of revertant coloniesper nmol, is calculated from linear regression analysis of thedata in Chart 2. 4-Hydroperoxycyclophosphamide is 6.2 timesmore mutagenic than cyclophosphamide following activation.Phosphoramide mustard is 3.2 times more mutagenic thanactivated cyclophosphamide. Thus, 4-hydroperoxycyclophos

phamide is a more potent mutagen than its breakdown product,phosphoramide mustard.

The third metabolite of cyclophosphamide tested, acrolein,was bacteriotoxic even at concentrations as low as 10 fig/plate(Chart 3); at concentrations of 0.01, 0.1, and 1.0 /Â¿g,survivalrates were adequate (97, 91, and 77%, respectively) to permitmutagenicity testing. Acrolein was not mutagenic itself and was

CONCENTRATION(^g plate)

Chart 3. Bacteriotoxicity of cyclophosphamide, 4-hydroperoxycyclophosphamide, phosphoramide mustard, and acrolein. Bacteriotoxicity was measuredwith S. typhimurium TA 1535 in the presence of an activating system, asdescribed in the text. Survival is expressed as the log of the surviving cell fraction

Ã‘o). Point, mean of 2 experiments.

CYCLOPHOSPHAMIDE

4-HYDROPEROXYCYCLOPHOSPHAMIDE

PHOSPHORAMIDEMUSTARD

0 0.5 1.0 1.5 2.0

REVERTANT COLONIES/NMOL

Chart 4. Relative mutagenicity of bioactivated cyclophosphamide, 4-hydroperoxycyclophosphamide, and phosphoramide mustard. The relative mutagenicity (revertant colonies/nmol) is calculated from linear regression analysis of thedata Â¡nChart 2, as described in the text.

only weakly mutagenic in the presence of liver microsomes(Chart 2). There was no linear relationship between acroleinconcentration and number of revertant colonies per plate byregression analysis.




Teratogenicity. The injection of 0.9% NaCI solution intoembryos in one uterine horn significantly (p < 0.05) increasedthe number of dead or resorbed fetuses in that horn (Chart 5).However, all the injected fetuses alive on Day 20 of gestationwere normal. Intraamniotic injection of 100 /Â¿g(0.36 /tmol) ofcyclophosphamide per fetus had no effect on pregnancy outcome measured by the number of fetuses dead, resorbed, ormalformed. Injection of 1000 jug (3.6 ;iimol) of cyclophosphamide per fetus did not significantly increase the number ofinjected fetuses that were dead or resorbed but did increasethe number of fetuses that were malformed. In addition tofinding 91.7% of the injected fetuses malformed, 100% of thelive uninjected fetuses on the contralateral side were alsomalformed. The malformations observed included edema(29%), hydrocephaly (97%), open eyes (55%), micrognathia(18%), cleft palate (16%), omphalocele (18%), bent tail (5%),and forelimb (5%) and hindlimb (74%) defects (adactyly, syn-

dactyly, polydactyly).A dose of 1000 ftg (3.4 jumol/fetus) of 4-hydroperoxycyclo-

phosphamide, in the same range as the dose of cyclophosphamide required to produce malformations, was toxic to all theinjected fetuses. In contrast to the observation with cyclophosphamide, 4-hydroperoxycyclophosphamide had no effect onthe contralateral uninjected fetuses. Lower doses of 4-hydro

peroxycyclophosphamide (10 ftg, 0.034 /Â¿mol/fÃ©tus;100 Â¿Â¿g,0.34 /Â¿mol/fÃ©tus)were still fetotoxic but now, to the survivingfetuses, were also teratogenic. The 2 lower doses of thiscompound (1 Â¡Â¿g,0.003 /unol/fetus; 10 jig, 0.03 /Â¿mol/fetus)


caused malformations in 41.2 and 66.7%, respectively, of theinjected fetuses. The spectrum of malformations induced by4hydroperoxycyclophosphamide was identical to that inducedby cyclophosphamide. These malformations were: edema(21%), hydrocephaly (36%), open eyes (29%), micrognathia(14%), cleft palate (14%), omphalocele (7%), bent tail (14%),and forelimb (18%) and hindlimb (39%) defects.

A high dose (1000 jug, 4.5 jumol/fetus) of phosphoramidemustard killed both injected and uninjected fetuses. A lowerdose (100 jig, 0.45 /tmol/fetus) of phosphoramide mustardwas fetotoxic only to injected fetuses; this treatment inducedopen eyes in one fetus of the 5 surviving injected fetuses.Treatment with a 10-fold lower dose (10 /Â¿g,0.05 jamol/fetus)

did not significantly increase fetal deaths compared to the0.9% NaCI solution controls but did induce malformations onthe injected side in 16% of the live fetuses; these includedhydrocephaly (4%) and forelimb (8%), hindlimb (16%), and tail(4%) defects. No malformed fetuses were observed at lowerdoses of phosphoramide mustard (1 or 0.1 /Â¿g/fetus).

High doses (100 jig, 1.8/xmol/fetus; 10ftg, 0.18/Â¿mol/fetus)of acrolein killed 98 and 100% of the injected fetuses, respectively. Injection of a lower dose (1 /ig, 0.018 /unol/fetus) ofacrolein induced malformations in 85.7% of the injected livefetuses. The types of malformations induced by acrolein werevery similar to those induced by cyclophosphamide and 4-

hydroperoxycyclophosphamide. These included: edema (7%),hydrocephaly (43%), open eyes (7%), cleft palate (7%), omphalocele (29%), and forelimb (50%), hindlimb (50%), and tail

DRUG:

v>LUV)

QLU

KOU.

mLU(0

LUIL

OLUmceOinLUaceOO

luQ

1.0

0.8

0.6

0.4

0.2

0.2

0.4

0.6

0.8

SALINE

1.0

CYCLO

PHOSPHAMIDE

4-HYDROPEROXYCYCLO-

PHOSPHAMIDE

PHOSPHORAMIDE

MUSTARDACROLEIN

1.0

0.8 >â€¢nO

0.6 Â§m

Â°-4 3

0.2 Ico

0.2

0.4

0.6

0.8

1.0

om

o313)mv>O310)moâ€¢nm

m(A

1 10 100 1000 10 100 1000 0.1 1 10 100DOSE/FETUS(>jg) - 100 1000

Chart 5. Teratogenicity and embryotoxicity of cyclophosphamide, 4-hydroperoxycyclophosphamide, phosphoramide mustard, and acrolein. Teratogenicity wasassessed in Sprague-Dawley rats on Day 20 of gestation after intraamniotic injection on Day 13 (HH , injected fetuses; [*Â£Â£),uninjected contralateral fetuses).Malformed fetuses are expressed as the mean number of malformed/live injected or uninjected fetuses per litter in each treatment group. Dead or resorbed fetusesare the mean number of dead or resorbed/injected or uninjected fetuses per litter in each treatment group. Bars, S.E.; â€¢,no live fetuses; saline. 0.9% NaCI solution.

AUGUST 1982 3019



B. F. Hales

(21%) defects. A 10-fold lower dose (0.1 jug, 0.002 Â¿Â¿mol/fetus)of acrolein had no effect on the rÃ©sorptionor malformation raterelative to 0.9% NaCI solution-treated controls. Thus, the doserange of acrolein capable of producing malformations is narrower than that for either 4-hydroperoxycyclophosphamide or

phosphoramide mustard.4-Hydroperoxycyclophosphamide, phosphoramide mustard,

and acrolein were all embryolethal at the doses equivalent tothe dose of cyclophosphamide required to produce teratogeniceffects. Both 4-hydroperoxycyclophosphamide and acrolein

induced the same spectrum of malformations as cyclophosphamide while phosphoramide mustard induced only hydro-cephaly and limb and tail defects. The doses of 4-hydroperox

ycyclophosphamide and acrolein required to produce the teratogenic effects were at least 100 times less than the dose ofcyclophosphamide necessary to produce the same defects.Moreover, cyclophosphamide was the only compound testedthat induced malformations in the contralateral uninjected fetuses as well as in the injected fetuses.

Administration of a low dose (1 00 jug/fetus) of cyclophosphamide had no significant effect on fetal weight [3.55 Â±0.61(S.D.) g, as compared to 3.67 Â±0.21 g for 0.9% NaCI solution-

injected controls]. However, treatment with a teratogenic doseof cyclophosphamide (1000 /Â¿g/fetus) caused a marked decrease (60%) not only in the weight of injected fetuses (1 .59Â±0.68 g) but also in the weight of the contralateral uninjectedfetuses (1.82 Â±0.59 g). Administration of 4-hydroperoxycy

clophosphamide (doses from 1 to 100 /ig/fetus) caused decreases varying from 19 to 48% in the weight of injectedfetuses relative to the 0.9% NaCI solution-injected controls. In

contrast to the results with cyclophosphamide, weights of thecontralateral uninjected fetuses were less affected (decreasesfrom 3 to 12% relative to the 0.9% NaCI solution-injected

controls). Treatment with doses (1 0 to 100 /Â¿g/fÃ©tus)of phosphoramide mustard that did induce malformations caused smalldecreases (9 to 18%) in the weight of injected fetuses. Theweights of the contralateral uninjected fetuses were not significantly affected. Treatment with acrolein (0.1 or 1 /Â¿g/fÃ©tus)also decreased (24%) the weight of injected fetuses withoutsignificantly affecting the weight of the uninjected contralateralcontrols.

DISCUSSION

The objective of this study was to investigate the relativemutagenicity and teratogenicity of cyclophosphamide, its synthetic preactivated analog, 4-hydroperoxycyclophosphamide,

and its cytotoxic metabolites, phosphoramide mustard andacrolein. With the exception of acrolein which was extremelybacteriotoxic, all the compounds tested were mutagenic to S.typhimurium TA 1535. Unlike cyclophosphamide, neither 4-

hydroperoxycyclophosphamide nor phosphoramide mustardrequired metabolic activation to be mutagenic.

Ellenberger and Mohn (1 2) previously investigated the mutagenicity of metabolites of cyclophosphamide in the Esche-

richia coli strain 343/1 13 test system. They found that acrolein(doses not reported) was not mutagenic whereas both 4-hydro

peroxycyclophosphamide and phosphoramide mustard werebase substitution mutagens. Our results confirm this observation in a different test system. In addition, in this report, it isdemonstrated that 4-hydroperoxycyclophosphamide is a more

potent mutagen than either of its breakdown products, phosphoramide mustard or acrolein. Other investigators have alsoused a S. typhimurium test system to measure mutagenicity ofbioactivated cyclophosphamide found in urine or blood (4, 41).In some test systems, acrolein, as well as phosphoramidemustard, causes chromosomal damage (3, 5). However, themechanism involved in chromosomal breaks or tangling maybe different from that in base substitution mutagenicity.

The spectrum of malformations produced by cyclophosphamide can be duplicated by treatment with 4-hydroperoxycyclo

phosphamide or acrolein at much lower doses. Both of thesecompounds have a selective local effect on injected fetuses. Inthese studies, phosphoramide mustard was less potent as ateratogen than either acrolein or 4-hydroperoxycyclophos

phamide; this metabolite produced both a lower incidence andfewer types of malformations. Using a variety of model systems,different laboratories have come to different conclusions withrespect to the "ultimate" teratogenic metabolite of cyclophos

phamide. Studying preimplantation mouse embryos in vitro,Spielmann and Jacob-MÃ¼ller (39) concluded that, since phosphoramide mustard and 4-hydroperoxycyclophosphamide ex

hibited identical embryotoxicity, phosphoramide mustard wasthe ultimate embryotoxic metabolite in vivo. Mirkes ef al. (27),using rat whole-embryo cultures undergoing organogÃ©nesis,

also demonstrated that phosphoramide mustard was the teratogenic metabolite of cyclophosphamide. Acrolein, at dosesequimolar to those of bioactivated cyclophosphamide, had noeffect on any of the parameters measured. However, followingin vivo i.p. injection in mice, Gibson and Becker (16) reportedthat phosphoramide mustard was less potent as a teratogenthan cyclophosphamide. Claussen ef al. (9) tested embryotoxicity in vivo with injections into the yolksac in rabbit embryos.These investigators did not study phosphoramide mustard, butthey did demonstrate that acrolein was both embryotoxic andteratogenic. They concluded that this metabolite is importantin the embryotoxicity of cyclophosphamide. Previously, MÃ¼ller(29) demonstrated that nitrogen mustard was teratogenic afterin utero injection in rats.

Our results suggest that acrolein and phosphoramide mustard both play a role in the embryotoxicity and teratogenicity ofcyclophosphamide. The failure to detect the teratogenicity ofacrolein in the in vitro embryo culture system is interesting.Acrolein itself may require metabolic activation to be embryotoxic; in vivo, this compound is metabolized to glycidaldehydeand glyceraldehyde or to acrylic acid (30). This conversionmay be possible in vivo near the fetus following an intraamnioticinjection but not in vitro in the culture system. Alternatively,because acrolein is extremely volatile, it may not persist in theculture flasks. The narrow dose-response relationship observed in the present study for the teratogenicity of acroleinsuggests that this effect could be missed. It is likely that thereis a dose threshold for the teratogenicity of this compound,similar to that reported for acetaminophen-induced hepatic

necrosis (28). Indeed, acrolein (like acetaminophen) can deplete glutathione (17), and the formation of the glutathioneconjugate of acrolein (3-hydroxypropylmercapturic acid) hasbeen suggested as an index of the activation of cyclophosphamide (1).

In this study, the most teratogenic metabolite of cyclophosphamide is acrolein, whereas the most mutagenic metabolite is4-hydroxycyclophosphamide. Acrolein, although extremely




bacteriotoxic, does not appear to be mutagenic. Thus, themutagenicity and teratogenicity of metabolites of cyclophos-

phamide are dissociable. This comparison of the relative mutagenicity and teratogenicity of "proximal" metabolites of cy-

clophosphamide should help delineate the mechanism of actionof this drug as a teratogen.

The metabolites of cyclophosphamide responsible for itstherapeutic effects are probably 4-hydroxycyclophosphamide-

aldophosphamide or phosphoramide mustard (6, 7, 23, 25,37, 40). If this is so, it may be possible to modify eithercyclophosphamide itself or its disposition to prevent the tera-

togenic effects mediated by acrolein, as has been suggestedfor the bladder toxicity produced by this metabolite.

ACKNOWLEDGMENTS

I thank Professor Norbert Brock for providing the 4-hydroperoxycyclophos-phamide and phosphoramide mustard and Dr. Bruce Ames for providing the S.typhimurium. The excellent technical assistance of Ranjana Jain is acknowledged.

REFERENCES

1. Alarcon, R. A. Studies on the in vivo formation of acrolein: 3-hydroxypro-pylmercapturic acid as an index of cyclophosphamide (NSC-26271) activation. Cancer Treat. Rep.. 60. 327-335. 1976.

2. Ames, B. N., McCann, J., and Yamasaki, E. Methods for detecting carcinogens and mutagens with the Sa/mone//a/mammalian-microsome mutagenicity test. MutÃ¢t.Res., 31: 347-364, 1975.

3. Au. W., Sokova, O. I., Kopnin, B., and Arrighi, F. E. Cytogenetic toxicity ofcyclophosphamide and its metabolites in vitro. Cytogenet. Cell Genet., 26.108-116, 1980.

4. Baibinder, E., Reich, C. I., Shugarts, D.. Keogh, J., Fibiger, R., Jones, T.,and Banks, A. Relative mutagenicity of some urinary metabolites of theantitumor drug cyclophosphamide. Cancer Res., 41: 2967-2972, 1981.

5. Benedict, W. F., Banerjee. A., and Venkatesan, N. Cyclophosphamide-induced oncogenic transformation, chromosomal breakage and sister chro-matid exchange following microsomal activation. Cancer Res., 38: 2922-

2924, 1978.6. Brock, N. Comparative pharmacologie study in vitro and in vivo with cyclo

phosphamide (NSC-26271). cyclophosphamide metabolites, and plain nitrogen mustard compounds. Cancer Treat. Rep., 60. 301-307, 1976.

7. Brock, N., and Hohorst, H. J. The problem of specificity and selectivity ofalkylating cytostatics: studies on /V-2-chloroethylamido-oxazaphosphorines.Z. Krebsforsch., 88. 185-215, 1977.

8. Brock. N.. Stekar, J., Pohl, J., Niemeyer, U., and Scheffler, G. Acrolein, thecausative factor of urotoxic side-effects of cyclophosphamide, ifosfamide,trofosfamide and sufosfamide. Arzneim. Forsch., 29: 659-661, 1979.

9. Claussen, U., Hellnann, W., and Pache, G. The embryotoxicity of thecyclophosphamide metabolite acrolein in rabbits, tested in vivo by i.v.injection and by the yolk-sac method. Arzneim. Forsch.. 30: 2080-2083,

1980.10. Cox, P. J. Cyclophosphamide cystitisâ€”identification of acrolein as the

causative agent. Biochem. Pharmacol.. 28: 2045-2049, 1979.11. Cox, P. J., and Abel, G. Cyclophosphamide cystitisâ€”studies aimed at its'

minimization. Biochem. Pharmacol., 28. 3499-3502, 1979.12. Ellenberger, J., and Mohn, G. R. Mutagenic activity of major mammalian

metabolites of cyclophosphamide toward several genes of Escherichia coli.J. Toxicol. Environ. Health, 3: 637-650, 1977.

13. Fantel, A. G., Greenway, J. C., Juchau, M. R., and Shepard. T. H. Terato-genic bioactivation of cyclophosphamide in vitro. Life Sci., 25:67-72,1979.

14. Foley, G. E., Friedman, O. M., and Drolet, B. P. Studies on the mechanismof action of cytoxanâ€”evidence of activation in vivo and in vitro. CancerRes., 21: 57-63, 1961.

15. Gibson, J. E., and Becker, B. A. Effect of phÃ©nobarbital and SKF 525-A onthe teratogenicity of cyclophosphamide in mice. Teratology, ): 393-398,

1968.16. Gibson, J. E., and Becker, B. A. Teratogenicity of structural truncates of

cyclophosphamide in mice. Teratology, 4: 141-150, 1971.


17. Gurtoo, H. L., Hipkens, J. H., and Sharma, S. D. Role of glutathione in themetabolism-dependent toxicity and chemotherapy of cyclophosphamide.Cancer Res., 41: 3584-3591, 1981.

18. Hales, B. Modification of the teratogenicity and mutagenicity of cyclophosphamide with thiol compounds. Teratology, 23: 373-381. 1981.

19. Hales, B. Modification of the mutagenicity and teratogenicity of cyclophosphamide in rats with Â¡nducersof the cytochromes P-450. Teratology, 24:1 -11, 1981.

20. Hales. B. F. Comparison of the teratogenicity and mutagenicity of somemetabolites and structural analogs of cyclophosphamide (CPA). Teratology,23: 40A, 1981.

21. Hales, B. F.. and Jain, R. Characteristics of the activation of cyclophosphamide to a mutagen by rat liver. Biochem. Pharmacol., 29: 256-259, 1980.

22. Hales. B. F., and Jain, R. Effects of phÃ©nobarbital and 0-naphthoflavone onthe activation of cyclophosphamide to mutagenic metabolites in vitro by liverand kidney from male and female rats. Biochem. Pharmacol., 29: 2031-2037, 1980.

23. Hohorst, H. J., Draeger, U., Peter, G., and Voelcker, G. The problem ofoncostatic specificity of cyclophosphamide (NSC-26271): studies on reactions that control the alkylating and cytotoxic activity. Cancer Treat. Rep.,60:309-315, 1976.

24. Klein, N. W., Vogler, M. A., Chatot, C. L., and Pierre, L. J. The use ofcultured rat embryos to evaluate the teratogenic activity of serum: cadmiumand cyclophosphamide. Teratology, 27: 199-208, 1980.

25. Lelieveld, P., and van Putten, L. M. Biologic activity of two derivatives andsix possible metabolites of cyclophosphamide (NSC-26271). Cancer Treat.Rep.. 60:373-379, 1976.

26. Manson, J. M., and Simons, R. In vitro metabolism of cyclophosphamide inlimb bud culture. Teratology, Ã•9:149-158, 1979.

27. Mirkes, P. E., Fantel. A. G.. Greenaway. J. C., and Shepard. T. H. Teratogenicity of cyclophosphamide metabolites: phosphoramide mustard, acrolein and 4-ketocyclophosphamide in rat embryos cultured in vitro. Toxicol.Appi. Pharmacol., 58: 322-330, 1981.

28. Mitchell, J. R.. Jollow, D. J., Potter. W. Z., Gillette, J. R., and Brodie, B. B.Acetaminophen-induced necrosis IV. Protective role of glutathione. J. Pharmacol. Exp. Ther., 187: 211-217, 1973.

29. MÃ¼ller,M. Does nitrogen mustard affect the foetus directly or secondarily byits effect on the mother? Experentia (Basel). 22: 247, 1966.

30. Patel, J. M., Wood, J. C., and Leibman, K. C. The biotransformation of allylalcohol and acrolein in rat liver and lung preparations. Drug Metab. Dispos.,8:305-308, 1980.

31. Sanyal. M. K., Kitchin. K. T., and Dixon. R. L. Rat conceptus developmentin vitro: comparative effects of alkylating agents. Toxicol. Appi. Pharmacol.,57: 14-19, 1981.

32. Schardein, J. L. Drugs as Teratogens, pp. 169-172. Cleveland: CRC Press,Inc., 1976.

33. Scheef. W.. Klein, H. 0., Brock. N., Burkert, H., GÃ¼nther,U., Hoefer-Jankey.H., Mitrenga, D., Schnitker, J., and Voightmann. R. Controlled clinicalstudies with an antidote against the urotoxicity of oxazaphosphorines: preliminary results. Cancer Treat. Rep., 63. 501-505. 1979.

34. Schmahl, D.. and Habs, M. Carcinogenic action of low-dose cyclophosphamide given orally to Sprague-Dawley rats in a lifetime experiment. Int. J.Cancer, 23: 706-712, 1979.

35. Sieber, S. M., and Adamson, R. H. Toxicity of antineoplastic agents in man:chromosomal aberrations, antifertility effects, congenital malformations, andcarcinogenic potential. Adv. Cancer Res., 22: 57-155. 1975.

36. Sladek, N. E. Therapeutic efficacy of cyclophosphamide as a function of itsmetabolism. Cancer Res., 32: 535-542, 1972.

37. Sladek, N. E. Bioassay and relative cytotoxic potency of cyclophosphamidemetabolites generated in vitro and in vivo. Cancer Res., 33: 1150-1158,1973.

38. Snedecor. G. W., and Cochran, W. G. Statistical Methods. Ed. 6. Ames,Iowa: Iowa State University Press, 1967.

39. Spielmann, H., and Jacob-MÃ¼ller, U. Investigations on cyclophosphamidetreatment during the preimplantation period. II. In vitro studies on the effectsof cyclophosphamide and its metabolites 4-OH-cyclophosphamide, phosphoramide mustard, and acrolein on blastulation of four-cell and eight-cellmouse embryos and on their subsequent development during implantation.Teratology. 23: 7-13, 1981.

40. Struck, R. F., Kirk, M. C., Witt, M. H., and Laster, W. R., Jr. Isolation andmass spectral identification of blood metabolites of cyclophosphamide:evidence for phosphoramide mustard as the biologically active metabolite.Biomed. Mass Spectrom., 2: 46-52, 1975.

41. Suling, W. J., Struck, R. F.. Woolley. C. W.. and Shannon, W. M. Comparativedisposition of phosphoramide mustard and other cyclophosphamide metabolites in the mouse using the Sa/moneHa/mutagenesis assay. Cancer Treat.Rep., 62: 1321-1328, 1978.

AUGUST 1982 3021



1982;42:3016-3021. Cancer Res Barbara F. Hales Acrolein4-Hydroxycyclophosphamide, Phosphoramide Mustard, andCyclophosphamide and Its Active Metabolites, Comparison of the Mutagenicity and Teratogenicity of

Updated version

http://cancerres.aacrjournals.org/content/42/8/3016

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/42/8/3016To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



comparison of the mutagenicity and teratogenicity of ... · teratogenicity testing. the...

Documents