ASPECTS or cvcmpuosmmms mxncmr m PERINATAL MICE Thesis for the Degree of M. S. MICHIGAN “STATE UNIVERSITY ROBERT-DOWNS SHORT JR. 1971 LIBRARY L W8” . naive-iv ” alkaline-av ? HUM} & SflNS' 300K BINDERY INC. I mannv SINGERS RT. IICHIGA] ABSTRACT ASPECTS OF CYCLOPHOSPHAMIDE TOXICITY IN PERINATAL MICE By Robert Downs Short Jr. Cyclophosphamide is a potent teratogen in mice. Metabolites of the parent compound are active alkylating agents. There is evidence, how— ever, that cyclophosphamide may produce toxic effects on developing tissues by a mechanism other than alkylation. These investigations were undertaken to further characterize the form of cyclophosphamide responsible for developmental toxicity in mice and to define biochemical lesions responsible for its teratogenic action. Cyclophosphamide administration to day-old mice produced a reduced growth rate, increased lethality, and morphologic abnormalities. The administration of an equimolar dose of nor-nitrogen mustard to day—old mice did not affect development. The ability of perinatal mice to form alkylating metabolites of cyclophosphamide was examined in an in vitro system. These studies indicated that neither the fetus, placenta, neo- natal liver, nor neonatal body was able to produce alkylating metabolites to the same degree as the adult liver. Cyclophosphamide developmental toxicity, therefore, occurred at a time when the affected organism had not fully developed an ability to activate the parent compound. These Robert Downs Short Jr. observations suggest that a non-alkylating form of cyclophosphamide was responsible for perinatal toxicity in mice. The in vitro study, in addition, showed that cyclophosphamide acti- vation proceeded at a greater rate in nursing females than in virgin females. Furthermore, there were no differences between males and females in their ability to activate cyclophosphamide. Phenobarbital pretreat- ment was shown to increase activation while SKF 525-A pretreatment inhibited the formation of alkylating metabolites. The incorporation of l4C-leucine into embryonic protein was inhibited by 72 and 96 hours after a teratogenic dose of cyclophosphamide. There was, in addition, a decreased incorporation of the precursor into the maternal liver by 96 hours and the kidney by 72 hours after cyclophospha- mide. The placenta, on the other hand, did not show an inhibition of incorporation at any of the times studied. The incorporation of this precursor into embryonic protein, when pregnant mice were treated 72 hours earlier with the drug, was inhibited at 30 and 60 but not at 15 and 120 minutes after precursor. Since the cyclOphosphamide induced inhibition of protein synthesis occurred at a time when the embryo was visibly abnormal it was not possible to conclude that this was the mechanism of teratogenic action. The synthesis of nucleic acids in embryos was studied by measuring the incorporation of 14C—orotic acid, l4C-uridine, and 14C-thymidine into nucleic acids. There was no significant effect of cyclophosphamide on the rate of synthesis of nucleic acids in embryos at 3, 24, or 72 hours after drug treatment, although DNA content of the embryos was signifi- cantly reduced. The data suggested that cyclophosphamide induced effects in the embryo, in contrast to the mother, were temporarily dissociated or occurred by different mechanisms. ASPECTS OF CYCLOPHOSPHAMIDE TOXICITY IN PERINATAL MICE BY Robert Downs Short Jr. A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Pharmacology 1971 ACKNOWLEDGMENTS The author wishes to thank the graduate committee members: Drs. Theodore M. Brody, Hyram Kitchen, David A. Reinke and especially James E. Gibson for their assistance and guidance throughout the various phases of this project. The author, in addition, would like to thank Dr. K. S. Rao for valuable comments and suggestions. ii TABLE OF CONTENTS CHAPTER I. INTRODUCTION. . . . . . . . . . . . . . . . . . CHAPTER II. METHODS . . . . . . . . . . . . . . . . . . . . A. Animals. . . . . . . . . . . . . . . B. Postnatal Toxicity . . . . . . . . . C. In vitro Cyclophosphamide Activation Animals . . . . . . . . . . Tissue Preparation. . . . . . Incubation Protocol . . . . . . Alkylating Analysis . . . . . bWNH D. Biosynthesis of Macromolecules . . . 1. Amino Acid Incorporation. . . 2 Nucleic Acid Precursor Incorporation. . . . . . . 3. Determination of RNA and DNA. 4. Estimation of Pool Size . . . E. Statistical Analysis . . . . . . . . CWTER III. RESULTS 0 O O O O O O I O O O O O O O O O O O O A. Postnatal Toxicity . . . . . . . . . B. In vitro Cyclophosphamide Activation C. Biosynthesis of Macromolecules . . . 1. Protein Synthesis . . . . . . 2. RNA Synthesis . . . . . . . 3 DNA Synthesis . . . . . . . CHAPTER IV. DISCUSSION. . . . . . . . . . . . . . . . . . . BIBLIOGRAPHY. . . . . . . . . . . . . . . . . . . . . . . . . iii Page 12 13 13 14 15 16 16 22 3O 30 36 46 50 56 Table LIST OF TABLES Cumulative litter mortality produced by nor— nitrogen mustard or cyclophosphamide administration to day—old mice . . . . . . . . . . . . . . . Optimization of pH for the in vitro activation of cyclophosphamide. . . . . . . . . . . . . . . . . . Optimization of cofactors for the in vitro activa- tion of cyclophosphamide. . . . . . . . . . . . . Effect of phenobarbital and SKF 525-A on cyclophos— phamide activation and liver/body weight ratio. . . Activation of cyclophosphamide by the 9000 g super- natant of livers from mature and Adweek—old mice. Activation of cyclophosphamide by the 9000 g super- natant of tissues obtained from pregnant mice . . . Activation of cyclophosphamide by the 9000 g super— natant of perinatal tissue obtained from mice . . DNA content after cyclophosphamide treatment. . . . DNA synthesis after cyclophosphamide treatment. . . iv Page 21 23 23 27 28 29 33 47 48 Figure LIST OF FIGURES Page Growth curves of litters treated with either cyclophosphamide or normal saline 24 to 48 hours after birth 0 O O O O O O O O O O O O O O O O O O 0 O O I 17 Growth curves of cyclophosphamide and nor-nitrogen mustard treated litters . . . . . . . . . . . . . . . . . 19 The kinetics of cyclophosphamide activation . . . . . . . 24 The ontogeny of cyclophosphamide activation . . . . . . . 31 14 . . . . . C—LeuCine incorporation into embryonic protein as a function of time after cyclophosphamide. . . . . . . 34 4 . . . C—Leucine incorporation into embryonic protein as a function of time after precursor . . . . . . . . . . 37 The incorporation of 14C-orotic acid into RNA . . . . . . 40 The incorporation of either l4C-uridine or 14C- orotic acid into RNA of embryos . . . . . . . . . . . . . 42 Milligrams RNA/g wet weight of embryo as a function of gestational age and treatment. . . . . . . . . . . . . 44 I. INTRODUCTION Cyclophosphamide was synthesized by Arnold and Bourseaux (1958) in an effort to provide an alkylating agent that required specific bioactivation by neoplastic tissues for the production of its cytotoxic action. The molecule consists of nitrogen mustard to which a cyclic phosphoramide ring structure has been attached. Higher concentrations of phosphatases and phosphamidases in tumor tissues (Gomori, 1948) were postulated as sites of activation of the inactive transport form. The activation of cyclOphosphamide was proposed to occur by the hydro- lytic cleavage of the P-N bond between nitrogen mustard and the phospha— mide ring to yield nitrogen mustard, an active alkylating agent. Cyclophosphamide has been successfully employed in the treatment of a broad spectrum of malignant and non—malignant diseases, including the lymphoproliferative diseases, such as Hodgkin's disease, the malig— nant lymphomas, chronic lymphatic leukemia and mycosis fungoides (Abele and Dodson, 1960). Cyclophosphamide has been shown to be unusually effective in treating leukemia in children, some of whom had developed a resistance to antimetabolite therapy (Brubaker et aZ., 1962). Alkylat- ing agents have been employed as potential inhibitors of the immune response and cyclophosphamide has been successfully used in the treatment of the nephrotic syndrome in children (Drummond et al., 1968). Cyclo- phosphamide therapy may produce the following signs of toxicity: alopecia, leucopenia, hemorrhagic cystitis, and minor degrees of anorexia and nausea. The usual clinic dose of-cyclophOSphamide in man is l 2 approximately 3 mg/kg/day p.o. or 15 mg/kg/week i.v. after a priming dose of 40—50 mg/kg i.v. The requirement of bioactivation for the conversion of the inactive cyclophosphamide parent compound to active cytotoxic metabolites has been demonstrated by various techniques. Foley et al. (1961) demonstrated that cyclophosphamide lacked an in vitro inhibitory effect on mammalian cell cultures. The sera of rats treated with cyclophosphamide, however, had an inhibitory effect on the cultures. Homogenates of neoplastic tissues, on the other hand, were not able to conduct the activation reaction to a sufficient degree to produce an inhibitory effect on cell growth. Brock and Hohorst (1963) used the formation of nitrobenzyl pyridine reactive material as an estimation of cyclophosphamide activa— tion. They demonstrated activation in mice, rats, guinea pigs, rabbits, cats, dogs, and humans. Activation was also demonstrated in liver slices and, to a lesser degree, in both adrenal cortical and lung tissue slices. Activation, however, was not demonstrated by acid prostrate phosphatase, alkaline renal phosphatase, or phosphamidases. Brock and Hohorst (1967) suggested that activation occurred by an NADPH and oxygen dependent microsomal system as a result of experiments involving fractionation of rat liver homogenates. Cohen and Jao (1970) have subsequently shown, in addition, that in vitro activation by a liver microsomal system can be inhibited by the simultaneous addition of either parahydroxymercuribenzoate, cytochrome c, hexobarbital, or testosterone. These observations suggested that the activation process was inhibited by chemicals that interfered with activity of the mixed function oxidase system or were metabolized by this system. Cyclophosphamide activation was shown (Sladek, 1971) to be inhibited by carbon monoxide. The parent 3 compound, in addition, was shown to have a type 1 spectral interaction with cytochrome P450. The previously described observations concerning cyclophosphamide activation led to the conclusion that the molecule existed as the inactive parent compound that required bioactivation for the production of cytotoxic effects. The bioactivation process, however, did not occur by the action of hydrolytic enzymes in tumor tissues but rather by the action of the mixed function oxidase system of the liver microsomal fraction. It is generally agreed that alkylating agents react with nucleo- philic centers in biological systems. Alkylating agents have the poten— tial for a variety of interactions and effects since there are a variety of nucle0philic centers within cells, e.g., phosphate, amino, and sulf- hydryl groups. The in vitro treatment of calf thymus DNA with nor- nitrogen mustard, for example, reduced its template activity in reactions catalyzed by E. coli DNA and RNA polymerase (Ruddon and Johnson, 1968). Nor—nitrogen mustard treatment, in addition, reduced the coding capacity of poly A, poly U, and poly C in an E. coli cell free protein synthesizing system (Johnson and Ruddon, 1967). The in viva treatment of hamster plasmacytomas with cyclOphosphamide produced a maximal decrease in DNA and RNA synthesis by 24-48 hours. The decrease in DNA synthesis was accompanied by a decrease in DNA nucleotidyl transferase activity in a crude cell free system (Wheeler and Alexander, 1969). The first report of teratogenic effects of alkylating agents in mammals appeared in 1948 (Haskin). A single administration of nitrogen mustard (0.5 and 1.0 mg/kg) to pregnant rats produced developmental abnormalities. The abnormalities observed included reduced size and weight, cleft palate, fusion of the digits, abnormal posture of the hind 4 limbs and short tail. Cyclophosphamide has been shown to be teratogenic in chicks (Gerlinger at aZ., 1963), rabbits (Gerlinger, 1964), rats (Kreybig, 1965), and mice (Gibson and Becker, 1968a). The intraperitoneal administration of 20 mg/kg cyclophosphamide to pregnant Swiss-Webster mice on gestational days 10 through 14 produced an increased number of resorptions and a variety of teratogenic effects (Gibson and Becker, 1968a). The surviving fetuses demonstrated cleft palates, exencephaly, digital defects, fusion of the long bones, hydro— cephalus and hydronephrosis. The widest spectrum of anomalies was pro- duced by the administration of cyclophosphamide on day 11. Fetal mortality curves were biphasic with peaks occurring when the drug was administered on days 10 to 11 and days 14 to 15. Doses of 5 and 10 mg/kg resulted in increased resorption and decreased growth rates in a dose-related manner but failed to produce detectable anomalies. The role of the parent compound and its metabolites in the produc- tion of cyclophosphamide teratogenicity was investigated by means of enzyme induction or inhibition (Gibson and Becker, 1968b). Pretreatment of pregnant mice with phenobarbital or SKF 525-A produced a corresponding increase or decrease in plasma alkylating products after cyclophosphamide administration. Since the teratogenic effects of cyclophosphamide were decreased by phenobarbital pretreatment and increased by SKF 525-A pretreatment it was proposed that teratogenicity was associated with the parent compound rather than metabolites. This suggestion was supported by the observation that the distribution of l4C-cyclophosphamide, under conditions of altered maternal metabolism, indicated that the increased teratogenic effects were associated with levels of the parent compound rather than metabolites. The levels of metabolites in the embryo were observed to remain relatively-constant regardless of the pretreatment. 5 This observation may indicate that there is a placental barrier to the transport of the more polar cyclophosphamide metabolites (Gibson and Becker, 1971a). In another study, Gibson and Becker (1971b) found that certain theoretical alkylating metabolites of cyclophosphamide did not possess the same potency or spectrum of teratogenic effects as cyclophosphamide. Cyclophosphamide was recently reported (Nordlinder, 1969) to produce toxic effects in developing mice treated 1 day after birth with a single dose of the drug. The treated mice showed increased lethality, lower adult body weights, delayed development of hair and abnormal morphology in comparison to controls. Animals that received cyclophosphamide 1 day after birth had microencephaly and short snout, ears, and tail at maturity (40 days). These observations suggest that cyclophosphamide was able to produce toxic effects in the neonates in the absence of acti- vation since the perinatal period is characterized by a reduced capacity of the mixed function oxidase system (Jondorf at aZ., 1959). The neo- natal mouse, therefore, may provide a system, in addition to the mouse embryo, that shows susceptibility to the non-alkylating form of cyclophosphamide. The present investigations were undertaken to further characterize the form of cyclophosphamide responsible for developmental toxicity and to define biochemical lesions responsible for its teratogenic action. The form of cyclophosphamide responsible for developmental toxicity was evaluated by 2 methods. The first method involved the confirmation and extension of data concerning the effects of cyclophosphamide in day-old mice. The effect of different drug doses was determined in order to establish dose related effects. The effect of alkylating activity on the development of mice was studied by treating day—old mice with 6 nor-nitrogen mustard. The validity of this comparison was tested by determining the pharmacokinetic properties of the 2 agents. The second method involved a determination of the ontogeny of cyclophosphamide activation by various perinatal tissues sensitive to its toxic effects. These observations permitted a correlation to be made between perinatal toxicity and the ability of the organism to metabolize the parent compound. The synthesis of DNA, RNA, and protein are important developmental processes. The disruption of these pathways may produce dramatic effects on development. The ability of cyclophosphamide to disrupt these path- ways in embryos was studied in an attempt to define drug induced bio- chemical lesions. II. METHODS A. Animals Virgin Swiss Webster mice were obtained from Spartan Research Animals (Haslett, Mich.) and housed at 70-75° F in stainless steel cages with wire mesh bottoms. The animals were maintained on a 12-hour light-dark cycle in order to synchronize estrus cycles. The cycle began at 8 a.m. when the lights were turned on. The animals were given free access to food and tap water. Timed pregnancies were obtained by the daily mating of 150-200 females. One male was placed in a cage containing 10 females at 8 a.m. and the females were examined 1 hour later for signs of copulation as indicated by the presence of vaginal plugs. Approximately 4% of the females are bred each day by this procedure. Mice were isolated and identified as being on day 1 of gestation if plugs were found. Ovula- tion occurs independently of copulation in mice and there is about a 5 hour delay between ovulation and fertilization. B. Postnatal Toxicity The effects of cyclophosphamide and nor-nitrogen mustard on post- natal development were studied in litters obtained from the previously described breeding program. The mice were allowed to deliver spon- taneously and drugs were administered 1 day later. The litters were housed on San—i-cell (Paxton Processing Co.) for 2 to 4 weeks. 8 Six litters, born on the same day, were assigned to each drug treat— ment group. The group was subdivided into 3 treated and 3 control litters on the basis of litter size and average body weight. The drugs were administered at doses of 45 and 80 mg/kg for cyclophosphamide treated litters and 54 mg/kg for nor—nitrogen mustard treated litters. The doses were calculated on the basis of the average mouse weight for each litter. The drugs were prepared immediately before use and injected subcutaneously in a volume of 50 Li of normal saline (Gibson and Becker, 1967). Mice in control litters received an equivalent volume of the vehicle. The average body weights were measured at the indicated times after birth by determining the total litter weight and size. Urine and plasma levels of nor-nitrogen mustard and cyclophosphamide were measured by the alkylating analysis (see below). After drug administration one—day-old mice were either isolated (0-4 hours) or returned to the mother (8 hours) prior to sacrifice. Blood was obtained by decapitation and exsanguination and urine was collected from excised bladders. The samples from cyclophosphamide treated animals were hydrolyzed in 0.5 ml of 1N HCl for 5 minutes on a boiling water bath to convert the parent compound to products that can be detected in the alkylating analysis. As a result of this treatment the analysis measures total drug in blood or urine. C. In vitro Cyclophosphamide Activation The ability of mice to activate cyclophosphamide was studied as a function of sex and age in an in vitro system. The effects of inducers and inhibitors of this system were studied, in addition, by pretreating virgin females with either phenobarbital (100 mg/kg for 3 days) or SKF 525-A (32 mg/kg one hour before sacrifice). 1. Animals Mature and 4-week-old mice were purchased from Spartan Research Animals prior to use. The remaining animals were obtained from the breeding program described above. 2. Tissue Preparation All tissue samples were obtained between 9 and 11 am. Mature and 4—week-old mice were sacrificed by cervical dislocation. Livers of these animals were perfused with 1.15% KCl in 0.5M tris (hydroxymethyl) aminomethane (tris, Sigma) buffer pH 7.4 (KCl—tris) prior to removal. The livers of a whole litter were pooled for single determinations. When determinations were made on the ability of the neonatal body to activate cyclophosphamide the liver, bladder, and viscera containing ingested milk were removed. The bladder and milk were removed in an effort to exclude variable factors that could affect the activation reaction. In prenatal studies the mother was killed by cervical dislo— cation and a laparotomy was performed. The uterine wall was exposed and cut to permit removal of the conceptus and its placenta. The tissues were maintained in an ice bath for up to 2 hours prior to being weighed and homogenized in 2 m1 KCl-tris per gram tissue. The tissues were homogenized in a loosely fitting, motor driven, teflon-glass Potter- Elvehjem homogenizer. The tissue fraction for incubation was obtained by centrifuging the homogenate at 9000 g for 30 minutes in an Inter- national centrifuge model B-20. 3. Incubation Protocol All incubations were conducted in a Dubnoff metabolic shaking apparatus at 100 oscillations/min and 37° C in room air. The incubation media consisted of glucose—6—phosphate (G6P, Sigma), nicotinamide adenine 10 dinucleotide phosphate (NADP, Sigma), magnesium sulfate (MgSO4), cyclo- phosphamide, 0.5 m1 of 0.5 M tris buffer pH 7.4, and 0.5 m1 of the 9000 g supernatant for a total volume of 2.5 ml. The components were added to a 50 m1 beaker and a marble was included to insure adequate agitation (Fouts, 1970). The amounts of the components added are indicated in the results section. Since the reaction did not proceed in the presence of 0.5 ml of 95% ethanol it was decided to use this relatively mild technique for terminating the reaction. The media was centrifuged briefly, after the addition of ethanol, to remove precipitable material. A sample of the supernatant was analyzed for alkylating metabolites (see below). The results were quantified with a standard curve prepared by hydrolyzing cyclophosphamide in 1N HCl for 30 minutes on a boiling water bath. An estimation of the non—enzymatic formation of alkylating metabolites was obtained from an incubation media that consisted of cyclophosphamide, tris buffer, and distilled water. The amount of alkylating material determined in this system was subtracted from the amount of alkylating material formed by the various tissue samples. Two incubations were conducted for each sample and the results were averaged to represent a single determination. Protein determinations were made on the 9000 g supernatant by the method of Lowry (Lowry et aZ., 1951) and crystalline bovine serum albumin (Armour) served as a standard. The product formed was expressed either as mumoles of alkylating activity/mg protein or umoles of alkylating activity/gram tissue. The kinetics of activation were determined by the double reciprocal plot of Lineweaver and Burk and a least squares regression analysis (Steel and Torrie, 1960). 11 4. Alkylating Analysis The alkylating analysis of Friedman and Boger (1961) as modi— fied by Gibson and Becker (1968b) was used in this study. This method depends on the ability of an alkylating agent to react with nitrobenzyl pyridine (NBP) to form a quaternary pyridinium ion that is highly colored in an alkaline solution. A 0.5 m1 sample of the supernatant was diluted to 3.0 ml with distilled water. The tubes were placed in an ice bath and 1.0 ml of cold 0.2 M acetate buffer pH 4.6 was added to each sample. The NBP reagent (0.4 m1 of 5% NBP in acetone) was added and the tubes were shaken. A11 tubes were heated for 20 minutes on a boiling water bath. The tubes were removed at the end of this period, cooled in ice, and shaken. The color was developed by adding 2.0 m1 acetone, 5.0 m1 ethyl acetate, and 1.5 ml 0.25 N sodium hydroxide to each tube. The color was extracted into the organic phase by shaking 10 times, adding approximately 200 mg sodium chloride, and then shaking 10 more times. The optical density of the organic layer was determined at 540 mu using a Bausch and Lomb Spectronic 20. D. Biosynthesis of Macromolecules L-Leucine (UL)-14C (197-240 mC/mM), orotic acid-6-14C (3.17 mC/mM), uridine-2-14C (38.8 mC/mM) and thymidine—2-14C (55.7 mC/mM) were purchased from ICN Tracer Lab (Irvine, Calif.) Cyclophosphamide was administered at a teratogenic dose of 20 mg/kg on day 11 of gestation (Gibson and Becker, 1968a). Control animals were injected at the same time with distilled water. The drug was injected in a volume of 0.1 m1/10 g body weight. The precursors were administered at various intervals after cyclophosphamide and the animals were sacri- ficed at the indicated times. Embryos and placentae were removed by previously described techniques, weighed, and homogenized in 10% tri- 12 chloroacetic acid (TCA). Maternal liver and kidney were also removed in some experiments and processed by the same techniques. The supernatant was saved, after centrifugation, for the determination of the acid soluble precursor pool. 1. Amino Acid Incorporation l4C—leucine was administered to pregnant mice at a dose of 5 uC/mouse to measure protein synthetic rates. After the initial centri- fugation the precipitate was washed with successive 5 ml portions of 10% TCA containing 20 mg/ml L-leucine, 10% TCA, and 25% potassium acetate in 95% ethanol. The precipitate was washed with TCA containing the non— radioactive precursor in order to minimize the binding of the radio— active precursor to acid insoluble material. The precipitate was then heated at 60° C for 5 minutes in an alcohol-ether solution (3:1) to extract lipids. After centrifugation the precipitate was heated in 10% TCA at 90° C for 15 minutes. The insoluble material was then treated with successive 5 ml washes of 25% potassium acetate in 95% ethanol, alcohol-ether (3:1) and finally 2 ether washes. A portion of the precipitate was transferred to a tared counting vial, dried, weighed, and solubilized in 1 ml of Soluene 100 (Packard) at 40° C for at least 8 hours. The radioactivity was determined in a Beckman LS-100 liquid scintillation system after the addition of 15 m1 of a toluene base counting solution (5 g 2,5-diphenyloxazole (PPO), 200 mg 1,4 bis [2—(4-methyl-5-phenyloxazolyl)]benzene (POPOP), and 1 liter toluene) to the vial containing the protein. The samples were counted for a suf- ficient time to give results with a maximum of 5% error. The counts/ min were converted to distintegrations/min with 14C—toluene and internal standardization techniques. The data were expressed as dpm/mg protein. 13 2. Nucleic Acid Precursor Incorporation 14C—labeled precursors of RNA (orotic acid and uridine) and DNA (thymidine) were administered to pregnant mice at a dose of 100 uC/kg to determine nucleic acid synthetic rates. After removal of the TCA soluble radioactivity the precipitate was washed with 10% TCA con— taining the appropriate non—radioactive precursor in an effort to remove unincorporated radioactivity. The precipitate from uridine and thymidine treated animals was washed with an additional portion of 10% TCA. All the precipitates were heated for 5 minutes at 60° C in an alcohol-ether solution (3:1) to extract lipids. The precipitate was heated for 15 minutes at 90° C in 5 m1 of 5% TCA to solubilize the nucleic acids. The supernatant was saved for the determination of radioactivity and nu- cleic acids. If the yield of radioactivity was expected to be low, then a wash of the precipitate was omitted. The radioactivity in 1.0 m1 of the TCA extract was determined in 15 ml of a dioxane base counting solution (60 g naphthalene, 4 g PPO, 200 mg POPOP, 100 ml absolute methanol, 20 ml ethylene glycol and dioxane for a final volume of 1 liter) using a 13703 external standard in a Beckman LS-100 system. The calibrated channel ratios from the external standard were used to com- pute dpm from the observed count rate. The samples were counted to at least 5% error and the results were expressed as dpm/m1 TCA extract. 3. Determination of RNA and DNA RNA.was estimated in the TCA extract by the orcinol procedure for determining ribose as recommended by Ceriotti (1955). Yeast RNA (Sigma) was used as the standard to obtain quantitative results. The orcinol reagent was prepared daily by mixing 200 mg orcinol, 10 m1 of 4mM CuCl2 in concentrated HCl, and a sufficient volume of concentrated 14 HCl to give 100 ml of the reagent. A sample of the TCA extract was brought to a volume of 5 ml by the addition of 5% TCA and 5 m1 of the orcinol reagent was added. The solutions were heated for 40 minutes on a boiling water bath and then cooled. The color was extracted with 5 m1 of isoamyl alcohol and the optical density of the organic layer was determined at 675 mu. The results were expressed as mg RNA/ml of the TCA extract. DNA was estimated by the reaction of deoxyribose with diphenylamine using Dischés' method as modified by Burton (1956). The diphenylamine reagent was prepared by dissolving 1.5 g diphenylamine in 100 m1 of glacial acetic acid and adding 1.5 ml of concentrated sulfuric acid. The solution was placed in small dark bottles and stored in the refrig— erator until needed. Acetaldehyde (0.1 m1 of a 16 mg/ml solution) was added to the diphenylamine reagent prior to use. A sample of the nucleic acid extract was brought to a volume of 1.0 ml by the addition of 5% TCA and 2.0 m1 of the diphenylamine reagent was added. The color was developed for 16-20 hours at room temperature and the optical density of the solution was determined at 600 mu. The DNA was quantified with a standard curve prepared from calf thymus DNA (Sigma). The data were expressed as mg DNA/m1 TCA extract. 4. Estimation of Pool Size A 1 ml portion of the acid soluble supernatant was added to 15 ml of the Dioxane base counting solution previously described. The radioactivity was determined on a Beckman LS-100 using either internal (amino acid pool) or external (nucleic acid pool) standardization tech- niques. The samples were counted to a maximum error of 5% and the data were expressed as dpm/g of tissue. 15 E. Statistical Analysis Statistical analysis was performed by Student's t test (Steel and Torrie, 1960). The level of significance was selected as P < 0.05. III. RESULTS A. Postnatal Toxicity The data in Figure 1 show that day-old mice treated with cyclo- phosphamide have a reduced growth rate and lower adult body weights than controls. The data in Figure 2 and Table 1 compare the effect of cyclo- phosphamide (45 and 80 mg/kg) and nor-nitrogen mustard on weight gains and mortality of litters treated 1 day after birth. The dose of nor- nitrogen mustard used in this study was calculated to be equimolar with the higher dose of cyclophosphamide. The data indicate that there was a dose related effect of cyclophosphamide in reducing the growth rates and increasing the mortality of treated litters. Mice treated with nor- nitrogen mustard, on the other hand, were similar to controls in regards to weight gain and mortality. These animals, in addition, did not exhibit any gross morphologic defects. The cyclophosphamide treated animals, however, demonstrated many of the morphologic abnormalities reported by Nordlinder (1969); for example, delayed development of hair, microcephaly, and short nose, ears, and tail. Mice that received the lower dose of cyclophosphamide did not exhibit morphologic defects, at maturity, to the same degree as mice receiving the higher dose. The pharmacokinetic properties of these 2 agents in day-old mice were determined. However, a combination of factors such as sample size, dose, and sensitivity of the assay did not permit quantitative determina- tions. The data did, however, permit qualitative conclusions concerning 16 17 Figure 1. Growth curves of litters treated with either cyclophos- phamide or normal saline at 24-48 hours after birth. The treated group received the subcutaneous administration of 45 mg/kg cyclophosphamide and the control group received an equal volume of the vehicle. The values plotted represent the mean i_S. E. of the ratio of the body weights, at a given age, to the average adult control body weight at the end of 6-7 weeks. The value used in calculating this ratio was 28.6 grams. The closed circles and solid line correspond to the control litters and the open triangles and dashed line correspond to the treated group. The values represent the mean for 3 litters. 18 1.00 1"" 0.50 1 new wem/Aouu CONTROL aoov WEIGHT -|\\ —1 I 0.10 _ Q \ ‘Q ‘ Q ~\ ‘Q t— . _—1 '\ \ \ s \ s \ \ I-°\-4 l-°-1 \ I—d T r I I 10 20 30 40 DAYS AFTER BIRTH Figure 1 SO 19 Figure 2. Growth curves of cyclophosphamide and nor—nitrogen mustard treated litters. The values plotted were determined by the process described in Figure 1. The growth curves were obtained from litters that received either nor—nitrogen mustard 54 mg/kg, A; cyclo— phosphamide 45 mg/kg, B; or cyclOphosphamide 80 mg/kg, C. The indi- vidual control growth curves have been omitted. The growth curves for nor-nitrogen mustard and its control were identical. Each curve was determined with 3 treated and 3 control litters and values earlier than 10 days have been omitted for clarity. ZOF ADULT CONTROL BODY WEIGHT 100 50 20 10 DAYS AFTER BIRTH Figure 2 50 21 Table 1. Cumulative litter mortality produced by nor-nitrogen mustard or cyclophosphamide administration to day-old mice % Mortality Treatment Control Treated Cyclophosphamide a b 45 mg/kg 0 22 :15 . b Cyclophosphamide 15 44 80 mg/kg :fi4 114 Nor—nitrogen mustard 8 11 54 mg/kg +4 1; 5 3The values represent the average percent mortality :_8. E. for 3 litters 46 days after birth. bSignificantly different from control at P < 0.05 (Student's _t_ test). 22 absorption and elimination. Since both drugs were present in the urine it was concluded that the agents had been absorbed, to some degree, from the subcutaneous injection site. Nor-nitrogen mustard could not be detected in the plasma 1 hour after injection. Cyclophosphamide, on the other hand, was detected in both plasma and urine at the end of 4 hours but only in the urine at the end of 8 hours. These observations suggest that the 2 agents may have different pharmacokinetic properties. B. In vitro Cyclophosphamide Activation The effect of a pH range of 6.7 to 8.4 was evaluated on the ability of the 9000 g supernatant from mature virgin females to activate 12.8 mM cyclophosphamide after a 1 hour incubation. Maximum activity was observed at pH 7.4 (Table 2) and this pH was used in subsequent incuba- tions. The effect of varying concentration of G6P, MgSO4, and NADP on the ability of mature female livers to activate 12.8 mM cyclophosphamide after a 1 hour incubation was studied. The addition of 12.5 umoles G6P, 12.5 umoles MgSO and 1.5 mg. NADP to the media enabled the reaction to 4 proceed independently of these cofactors (Table 3). Subsequent incuba- tions were conducted using these amounts of cofactors. The kinetics of activation were determined using the 9000 g super- natant of liver from mature females and a substrate concentration range of 0.4 to 12.8 mM. The apparent Km was 0.92 mM and the Vmax was 1.7 umoles of alkylating activity/min/mg protein (Figure 3). The calculated line had a regression coefficient equal to 0.92. As a result of the kinetic analysis subsequent incubations were conducted using a substrate concentration of 12.8 mM and a 10 minute incubation unless otherwise indicated. 23 Table 2. Optimization of pH for the in vitro activation of cyclophosphamidea Cyclophosphamide alkylating activity relative to the activity atng 7.4b 6.7 7.4 7.9 8.4 Relative activity 0.67C 1.0 0.73 0.47 +0.07 +0.08 _-_I-_0.08 Table 3. Optimization of cofactors for the in vitro activation of cyclophosphamidea Cyclophosphamide alkylating activity relative to the activity produced by the addition of 1.5 mg NADP and 25 umoles G6Pb mg NADP umoles of G6P and Mg804 added added 6.25 25 37.5 1.0 0.56:0.06C 0.56:0.10 --- 1.5 0.78:0.18 1.0:0.21 0.93:0.06 2.0 1.0 i_0.l4 0.86i0.l8 0.71:0.14 2.5 0.71:0.14 0.73:0.10 0.71:0.04 89000 g supernatant of liver from mature virgin female. bOne hour incubation. CMean :_S. E. 24 .25 w.~a on «.0 mo swamp coaumuusoosoo mumuumnsm m paw monEom afiwuw> Eouw Ho>fia mo uamumsummdm w ooom wnu wsflmn Eoumkm voNflSHuao onu cw vodwsumump ouw3 cowum>fiuom mo mofiuocfix one .GOHum>fluom owwfimnamonaoao%o mo mowuocflx may .m muswflm 25 m shaman “\— O.N O... 2E9... .9: ..z=z Eu< a... < 3.52 =8 n.— u 3:5 SE 0.0" E: v \ O O.— o >\_ O.N ZO:<>:U< un=z<=tm02m0du>u 26 The activating ability of livers from phenobarbital and SKF 525-A pretreated female mice was investigated in this system. These experiments were conducted to provide an in vitra confirmation of the in viva obser- vations of Gibson and Becker (1968b) which showed that these pretreatments altered the maternal metabolism of cyclophosphamide. A 5 minute incuba- tion was used in an effort to maintain a linear reaction rate for livers from phenobarbital pretreated mice. Table 4 shows that phenobarbital pretreated animals had only 35% of control activity. The phenobarbital pretreated mice, in addition, had a significantly increased liver/body weight ratio. It was concluded from these observations that the in viva alteration of cyclophosphamide metabolism could be duplicated using in vitra conditions and, therefore, was due to a direct action of the pre- treatment drugs on the liver mixed function oxidase system. In order to investigate the ontogeny of cyclophOSphamide activation it was necessary to determine if a sex difference in the activation reaction would be a complicating factor. The data in Table 5 indicated that there was no significant sex difference in cyclophosphamide activa— tion at 4 weeks of age or at maturity. Since there was no sex difference in cyclophosphamide activation it was possible to pool litters without separating the mice on the basis of sex. The data presented in Table 6 indicated that the whole fetus on Idays 15, 17, or 19 of gestation was not able to activate the parent compound to the same degree as the maternal liver. The corresponding placentae, in addition, did not possess activity equivalent to the maternal liver. The range of values for mg protein/incubate were: maternal liver (21—31), placenta (9-14), and fetus (8-12). 27 Table 4. Effect of phenobarbital and SKF 525-A on cyclophosphamide activation and liver/body weight ratio mimoles Alk. Act.a g liver Pretreatment mg protein 100 g body Control 9.0 i 0.9 (6)b 4.5 _-I_-_ 0.2 (6) Phenobarbital 21.0 i 2.3 (5)c 5.6 i 0.2 (6)C SKF 525—A 3.2 i 0.6 (6)° 4.3 i 0.1 (6) aFive min incubation. Alk. Act. = alkylating activity. bMean :_S. E. (number of observations). CSignificantly different from control (P < 0.05). 28 Table 5. Activation of cyclophosphamide by the 9000 g supernatant of livers from mature and 4dweek—o1d mice Animal mumoles alkylating_activity/mg proteina age female male 4 week 16.0 i 2.2 (6)b 20.0 i 1.8 (6) mature 18.0 i 1.6 (8) 18.0 i_3.2 (8) a . . . Ten min incubation. bThe values represent the mean :_S. E. (number of determinations). 29 Table 6. Activation of cyclophosphamide by the 9000 g supernatant of tissues obtained from pregnant mice Day of mumoles alkylating activity per mggproteina gestation Maternal liver Placenta Fetus 15 21.0 i 4.4 (3)b 1.7 i 0.2 (3) 1.8 i 1.0 (4) 17 14.0 _-+_-_ 1.7 (6) 7.5 i 1.9 (6) 3.0 -_I-_ 3.0 (4) 19 15.0 :_0.8 (7) 4.6 i 1.9 (7) 1.0 :_1.0 (5) a . . Ten min incubation. b . . The values represent the mean :_S. E. (number of determinations). 30 The data presented in Figure 4 show that the neonatal liver was not able to activate cyclophosphamide as readily as adult virgin female liver for at least 2 weeks after birth. There was an increase in activity between 14 and 21 days of age and activity equivalent to adult values was reached by 2l-day-old mice. There was, in addition, a sig- nificant difference in activity between control and nursing mothers at 3, 7, l4 and 21 days after giving birth. The range of values for mg protein/incubate were: maternal liver (22—39), mature female liver (21—33) and neonatal liver (11—20). The data in Table 7 show activation expressed in terms of tissue weight. This presentation of the data provides a more obvious basis for comparing the abilities of various tissues to metabolize cyclophos— phamide. This expression of the data does not alter the relationship between maternal and control females at 14 and 21 days postpartum. The neonatal livers at 21 days of age, however, have significantly lower activity than adults. The table shows, in addition, that the neonatal body, without the liver, is a relatively insignificant site of activa— tion for at least 5 days after birth. C. Biosynthesis of Macromolecules 1. Protein Synthesis The data in Figure 5 are a plot of the specific activities of total embryonic protein, after a 30—minute pulse of l4C—leucine, as a function of time after cyclophosphamide. Cyclophosphamide was administered on day 11 of gestation. As a result of this treatment the embryos had a significantly reduced capacity to incorporate the precursor at 72 and 96 hours after drug administration. There were, however, no significant differences in the Specific activities of control and drug treated proteins 31 .mfioauwsfiahmumw aim Mom .m .m.H.smmE onu mum muafiom 03H .woumsoos mnu ma vmaflaumuom mmsam> ou wsoammuuoo mmHmeHHu ammo mom “sauce was How paaflaumumw monam> ou odommouuoo mmHoHHo vmmoau .mGOflumcfiauauap on How aflououm we \xufl>fiuom wcfiumamxam mmHQsau 0.0.H ma mm3 moamsom Ham now mwmum>m vmcfinaoo one .GOHumnoooH sundae OH m “mums afimuoum wa\mufi>fiuom wafiumaxxam mmaoane omloa mm3 .msouw m wafiwmum>m Hmuwm .mmamaom uaowm How manam> mo mwsmu 05H .muo>fla stuouma was Hmumooms spas hamnomQMDHoafim nonfiaumuov muoa uo>HH mamaom ago How mooam> 65H .wmaoom mums muouufla mHo£3.Eoum mso>fiq .mmamamm assume Eoum muo>HH Ga wcnow hufi>fluom ago mo unmouoa m mm wommmuaxm mm3 mum>fia Hmahouma pom mafia doomsoas Ga zuH>Huom moflom> Iwuom unflamnmmonmoao%o oauwa as 05H .coaum>Huom owfiamnmmonmoaoho mo hsowouco 6:9 .6 muswfim 32 FN VF q ouswfim 12:0 cur? mg I . .... ......H. ......... . LI m.O m; AllAIlIN 3M 1V1!!! 33 Table 7. Activation of cyclophosphamide by the 9000 g supernatant of perinatal tissue obtained from mice umoles alkylating_activity per gram tissue x 10_23 Days after Maternal Neonatal Neonatal birth liver liver body Fetus Placenta -5 280 i 21 (6)b --- —-- 32 i 14 (6) 31 i 9 (6) —l 239 :_18 (7) —-— --- 37 :.20 (7) 33 ill4 (7) +1 270 $.18 (6) 41 i'l4 (6) 18 :_10 (5) --- --- +5 370 i 54 (5) 27 i 6 (4) 33 i 19 (3) —-- --- +14 460 _-I_-_ 4 (3) 47 i 3 (3) --- --- --- +21 400 :_39 (7) 189:34 (7) --- --- --- a a I Ten min incubation. bThe values represent the mean :_S. E. (number of determinations). The activity in mature virgin females was 300 :_23 (10). 34 . 14 . . . . . . Figure 5. C-LeuCine incorporation into embryonic protein as a function of time after cyclophosphamide. Cyclophosphamide was adminis- tered on day 11 of gestation and the precursor was administered to the mother 30 minutes prior to sacrifice and removal of the embryos. Pro— tein was isolated and counted. The values plotted represent the mean :_S. E. of the specific activities for at least 3 determinations. An asterisk indicates a significant difference from control (P < 0.05). 35 ‘4C-LEUCINE INCORPORATION (30 MIN. PULSE I 1,400 I4 I .000 600 DPM I mg. PROTEIN CONT. . I'"l"_|""'"l""'l— 3 24 4B 72 96 200 HOU RS AFTER TREATMENT Figure 5 36 at 3, 24, and 48 hours after treatment. The acid soluble precursor pool from the 2 groups of embryos, in addition, was not significantly dif- ferent at any of the times studied and, therefore, depletion of the pre— cursor could not account for the decrease in incorporation. Cyclophos- phamide treatment did not affect the ability of the placenta to incor- porate the precursor at any of the times studied. There was, however, a significant reduction in precursor incorporation in the liver at 96 hours after cyclophosphamide and in the kidney at 72 and 96 hours after treatment . The duration of the l4C—leucine pulse was varied 72 hours after drug treatment, in order to determine if a 30 minute pulse period was optimum for precursor incorporation by the embryos. The acid soluble pool was maximally labeled by 15 minutes in both groups of embryos. The incorporation of the precursor into proteins by control embryos reached a plateau at 30 minutes (Figure 6). The specific activity of proteins from treated embryos, however, was significantly different from controls at 30 and 60 minutes after precursor administration. The specific activity of protein was not significantly different between the 2 groups at 120 minutes after precursor administration. There were, in addition, no significant differences between either group in the acid soluble precursor pool at any of the times studied. Time dependent differences between the 2 groups in the placental incorporation of precursor, on the other hand, were not detected at any of the times studied. 2. RNA Synthesis Initial experiments using orotic acid as a precursor of RNA showed a placental barrier to its transport into the embryo. Since a 37 Figure 6. l4C-Leucine incorporation into embryonic protein as a function of time after precursor. The precursor was administered to the mother 72 hours after cyclophosphamide treatment. The mothers were sacrificed at the indicated times and the embryos removed. The protein was isolated and counted. The values plotted represent the mean :_S. E. of the specific activities for at least 3 determinations. An asterisk indicates a significant difference from control (P < 0.05). DPM Img. paonm 900 500 100 38 14 C- lEUC I N E INCORPORATION 0N DAY 14 15 30 60 120 MIN. AFTER PRECURSOR Figure 6 39 30 minute pulse did not produce any detectable labeling of RNA in embryos the time course of precursor incorporation was investigated. The mice used in this study were not pretreated and ranged bewteen 12 and 14 days of gestation. Figure 7 shows that both the liver and placenta have a greater ability than the embryo to incorporate the precursor into nucleic acid under the conditions of this study. The reduced incorporation of orotic acid was attributed to a placental barrier because the acid soluble precursor pool in the placenta and liver were approximately 10 and 100 times, respectively, the precursor pool of the embryo. A 4 hour pulse was selected in an effort to obtain satisfactory labeling of RNA in the embryo with a minimum precursor exposure. Using these conditions, it was determined that cyclOphosphamide did not significantly affect the ability of the embryo to synthesize RNA 24 hours after treatment. A 4 hour 14C—uridine pulse was initially used as a result of the previous experience with orotic acid. These experiments indicated that there was no significant difference in the ability of control and cyclo- phosphamide treated embryos to incorporate l4C-uridine 24 hours after drug treatment. The specific activity of the RNA, after a 4 hour pulse indicated the feasibility of a shorter pulse period. The precursor, therefore, was given for a 1 hour period in subsequent experiments. The ability of embryos to synthesize RNA.was not affected at 3 or 72 hours after drug treatment as determined with a 1 hour 14C—uridine pulse. The data in Figure 8 summarize the incorporation of l4C—orotic acid and l4C-uridine into RNA of cyclophOSphamide treated embryos as a percent of the incorporation determined in control embryos. There was an indi— cation that RNA synthesis was depressed; however, the differences between a treated group and its control were not statistically significant. The data in Figure 9 are the values of mg RNA/g of wet weight of embryo. 40 Figure 7. The incorporation of l4C-orotic acid into RNA. The precursor was administered to mice on days 12—14 of gestation. The mothers were sacrificed at the indicated times after precursor admin— istration. Radioactivity was determined and RNA measured in the hot TCA extract of the acid precipitate from the indicated tissues. The values plotted represent either a mean i_range for 2 determinations (1 and 2 hours) or the mean iLS° E. for 3 to 4 determinations. 41 ”c ononc ACID INCORPORATION DAY 12 -14 I J ”' I-----~~,| 10.0 ...4. ' 7— 3.0 'I/ LIVER /’ I P o DPM I mg. RNA x103 2" o ‘02 0.4 I2 4 8 12 HOURS AFTER PRECURSOR Figure 7 42 Figure 8. The incorporation of either 14C-uridine or 14C—orotic acid into RNA of embryos. Cyclophosphamide was administered on day 11 of gestation and the animals were sacrificed at the indicated times. l4C—orotic acid was given for 4Iunanand l4C-uridine was given for either 4 hours (24 hours after cyclophoSphamide) or 1 hour (3 and 72 hours after cyclophosphamide) prior to sacrifice. The values correspond to the mean :_range for 2 determinations or to the mean i_S. E. for 3 or more determinations. The ratio of treated/control determinations was: 3 hours after treatment (4/2), 24 hours after treatment (4/3 for uridine and 3/3 for orotic acid) and 72 hours after treatment (2/3). PERCENT OF CONTROL 43 RNA SYNTHESIS I: mc-ononc ACID @ ”c-URIDINE — O O O O 20 24 HOURS AFTER CYC LOPHOSPHAM IDE Figure 8 III 44 .mooflumswauouow m ummoa um pow .m .m.H name asp ucmmmummu moDHm> 6:9 .wOSumfi Hocflouo onu mo uomuuxo «09 no: asp CH vocaauouov mm3.¢nv