“In-- wu..~u-‘..—.-~-.....~...»- -. ,. .fi._,~,, -_,,_ w-“ V.“ -‘>_H~“.. TOXECETY 0F CYCLOPHOSPHAMTDE ' EURIPTG DEVELOPMENT Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY ROBERT DOWNS SHORT JR. 1973 This is to certify that the thesis entitled TOXICITY 0F CYCLOPHOSPHAMIDE DURING DEVELOPMENT presented by Robert Downs Short, Jr. has been accepted towards fulfillment of the requirements for Ph.D. degree in PharmacoTogy U v Major professor / Date ‘3 /6{/73 0-7639 .. 1": Y Michigan Sign University magma DY 7’ HUNG 8 SONY BOOK BINDERY INE. LIBRARY BINDE RS J gammy. mcmsg WI 1 ABSTRACT TOXICITY OF CYCLOPHDSPHAMIDE DURING DEVELOPMENT BY Robert Downs Short Jr. Mammalian development represents an integrated sequence of growth and differentiation. Agents which interfere with these processes produce abnormal development. The purpose of this study was to examine the molecular basis of cyclophosphamide induced disruption of embryonic and postnatal development in mice. Cell division, an important aspect of normal growth, was reduced in embryos after the mothers received a teratogenic dose of cyclo- phosphamide (20 mg/kg i.p.) on day 11 of gestation. The population of dividing cells, which was estimated by measuring the incorporation of 1(.Cmthymidine into DNA, was reduced 12 hours after drug treatment. This treatment produced alkylation of embryonic macromolecules, as measured by the presence of radioactivity in a lipid-extracted acid precipitate, after mothers received 1l‘c—cyclophosphamide. Radioactivity in the acid precipitate declined in the order liver, placenta, and embryo and was present for at least 10 hours after treatment. 14C» cyclophosphamide radioactivity disappeared from the nucleic acid frac- tion of embryonic tissue more slowly than from.the nucleic acid fraction of maternal liver. The population of dividing cells was also reduced in the liver, brain, and carcass of neonatal mice after a dose of Robert Downs Short Jr. cyclophosphamide (80 mg/kgs.c. at one day of age) which reduced the growth of these tissues. The incorporation of 14C-thymidine into DNA was reduced 1 day after cyclophosphamide to 8, 26, and 251 of control in the liver, brain, and carcass, respectively, and this effect lasted for at least 3 days. The prolonged inhibition of DNA synthesis in neonatal relative to embryonic mice was attributed to a prolonged exposure of these animals to alkylating cyclophosphamide metabolites which were slowly formed from the poorly eliminated parent compound. The transcription and translation of genetic information are important developmental processes. RNA synthesis, as measured by 1('Cuuridine incorporation into RNA, was reduced in the liver and brain of mice treated at 1 day of age with 80 mg/kg s.c. cyclophosphamide. Liver RNA synthesis was reduced to 36 and 641 of control at l and 5 ,days after cyclophosphamide, respectively. Brain RNA synthesis was reduced to 78 and 642 of control at l and 4 days after treatment, respectively. Protein synthesis, as measured by 1['C-leucine incapora- tion into protein, was not affected in a manner which would indicate that drug treatment altered this process. The qualitative aspects of protein synthesis were studied by comparing the electrophoretic pattern of proteins synthesized in vitro by livers from control and cyclophosphamide treated neonates. Drug treatment did not affect the synthesis of specific liver proteins at l and 3 days after treatment. Drug treatment, in addition, did not affect the enzymatic activity of glucose-6-phosphatase, which was increased in neonatal liver shortly after birth, at l and 3 days after treatment. Drug toxicity, as esti- mated by these parameters, was not associated with quantitative or qualitative changes in protein synthesis. Robert Downs Short Jr. The protein content of a tissue is dependent on both the synthesis and degradation of protein. Cyclophosphamide treatment of neonatal mice increased the turnover of liver, brain, and carcass proteins, as measured by the reduced half-life of proteins labeled in utero with 14C-guanido-arginine. An increased degradation of radioactive proteins was attributed to the cytotoxicity of cyclophosphamide and catabolic processes associated with cell death. Doses of cyclophosphamide which disrupted embryonic and postnatal development interfered with biochemical processes essential for normal growth and development. Abnormal development was attributed to a reduced population of proliferating cells, as measured by a reduced incorporation of 1l‘C-thymidine into DNA, and increased cell death, as measured by a reduction in both DNA concentration (mg DNA/g tissue) and DNA content (mg DNA/organ). TOXICITY OF CYCLOPHOSPHAMIDE DURING DEVELOPMENT BY Robert Downs Short Jr. A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pharmacology 1973 ACKNOWLEDGMENTS The author wishes to thank the graduate comittee members, Drs. Theodore M. Brody, Jay I. Goodman, Jerry B. Hook, James E. Trosko, and especially James E. Gibson, for their assistance and guidance through- out the various phases of this project. These studies were supported by NIH Research Grant ES 00560 and general research support from the College of Veterinary Medicine and College of Human Medicine, Michigan State University. The author, in addition, received financial support from NIH Training Grant GM01761. ii TABLE OF CONTENTS ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . LIST OF TABLES. . . . . . . . . . . . . . . . . . . . . . LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . Historical Considerations. . . . . . . . . . . . . Phases of Development. . . . . . . . . . . . . . . Processes of Development . . . Principles of Teratology . . . . . . . . . . . Examples of Agents that Disrupt Development. . Cyclophosphamide . . . . . . . . . . . . . . . Toxicity of Cyclophosphamide During Development Purpose. . . . . . . . . . . . . . . . . . . . mmODS O O O O O O O O O I C O C O O O O O O O O O O O 0 Animals. . . . . . . . . . . . . . . . . . . . . Biosynthesis of DNA by Embryonic Mice. . . . . . . Alkylation of Embryonic Macromolecules . . . . . . Postnatal Toxicity . . . . . . . . . . . . . . . . Biosynthesis of Macromolecules by Neonatal Mice. . Protein Turnover in Neonatal Mice. . . . . . . . . Qualitative Changes in Neonatal Protein Synthesis. Recommendations. . . . . . . . . . . . . . . . . . Statistics . . . . . . . . . . . . . . . . . . . . RE SULTS O O O O I O O O O I O O O O O O O 0 O O O 0 O O O Biosynthesis of DNA by Embryonic Mice. . . . . . . Alkylation of Embryonic Macromolecules . . . . . . Postnatal Toxicity . . . . . . . . . . . . . . . Biosynthesis of Macromolecules by Neonatal Mice. . Protein Turnover in Neonatal Mice. . . . . . . . . Qualitative Changes in Neonatal Protein Synthesis. DISCUSSION. 0 I O O O O O O O I O 0 O O O O O O O O 0 O 0 Introduction . . . . . . . . . . . . . . . . . . . Inhibition of Cell Division. . . . . . . . . . . . Interference with Differential Gene Expression . . Conclusions. . . . . . . . . . . . . . . . . . . . BIBL IMMYO O O O O I O O O O O O O O O O O O O O O O O Page ii iv vi 102 106 111 Table 10 11 12 13 LIST OF TABLES 14C-thymidine incorporation after cyclophosphamide trea tmnt O O O O O O O O O I O O O O O O O I O O O O O 0 DNA content after cyclophosphamide treatment . . . . . . 14C-cycIOphosphamide radioactivity in DNA, RNA, and protein fractions from.maternal liver, placenta, and emb ryo O O 0 O I O O O O O O O O O C O O O I O O O O I O 14C-cyclophosphamide radioactivity in the nucleic acid extract from maternal liver, placenta, and embryo . 14C-cyclophosphamide radioactivity in the protein and nucleic acid fraction from.maternal liver, placenta, and emb ryo O O O O O I O O O O O O C O O O I C O O I O O C 0 Organ weights and organ to body weight ratios of liver and brain from mice treated with cyclophosphamide at one day 0 f age 0 O O O O O O O I O O 0 O O O I I O C C 0 Concentration and content of DNA and RNA in neonatal liver after treatment with cyclophosphamide at one day Of age I O O O O I O O O O O O O O O I O O O O O O O I 0 Concentration and content of DNA and RNA in neonatal brain after treatment with cyclophosphamide at one day Of age 0 O O 0 O O C O O O O O O O 0 O O O O I O O O O 0 Concentration of DNA and RNA in neonatal carcass after treatment with cyclophosphamide at one day of age. . . . DNA specific activity in neonatal mice after hydroxyurea t reatment O O O C 0 0 O O O O I O O O O O O O O I O C O 0 Protein synthesis in mice treated with cyclophosphamide. Equations for the calculated least squares regression line of log protein specific activity as a function of days after cyclophosphamide. . . . . . . . . . . . . . . Half-life of protein from neonatal mice after cyclo- phosphamide as estimated by the change in protein BPeCific aetiVitYe O O O O O O I O O O O O O O O I O O 0 iv Page 44 47 49 56 57 65 73 74 75 76 78 8O 81 Table 14 15 16 17 18 * Page Equations for the calculated least squares regression line of log protein radioactivity per organ vs. days af ter treatmnt O O O I O O O O O O O O I O O O O O O O O O 82 Half-life of protein from neonatal mice after cyclo- phosphamide as estimated by the change in organ protein radioactivity. . . . . . . . . . . . . . . . . . . . . . . 83 Half-life (days) of proteins from neonatal mice after cyclophosphamide . . . . . . . . . . . . . . . . . . . . . 84 Soluble and particulate protein in the liver and brain of neonatal mice after cyclophosphamide. . . . . . . . . . 86 G1ucose-6-phosphatase activity in the livers of treated neonatal mice relative to the enzyme activity in maternal liver. . . . . . . . . . . . . . . . . . . . . 95 Figure 10 11 LIST OF FIGURES 14C-thymidine incorporation into embryonic DNA as a function of time after cyclophosphamide. . . . . . . . . . Acid soluble radioactivity in the maternal liver, placenta, and emb O'as a function of time after 500 uc/kg (20 mg/kg) 4C-cyclophosphamide . . . . . . . . . . Radioactivity in the acid washed lipid extracted acid precipitate from.maternal liver, placenta, and embryo as a function of'time after 500 TIC/kg (20 mg/kg) 14C~cyclophosphamide . . . . . . . . . . . . . . . . . . . The ratio of acid precipitable to acid soluble radio- activity in the maternal liver, placenta, and embryo as a function of time after 500 uc/kg (20 mg/kg) 14C—cyclophosphamide . . . . . . . . . . . . . . . . . . . Adult mice that received either 80 mg/kg cyclophosphamide 24 to 48 hours after birth or normal saline. . . . . . . . Adult mice that received either 80 mg/kg cyclophosphamide 24 to 48 hours after birth or normal saline. . . . . . . . DNA, RNA and protein synthesis in the liver of cycle- phosphamide treated animals expressed as a percent of contr01 values 0 O O O I O O O O O O O 0 0 0 0 C 0 O O o 0 DNA, RNA and protein synthesis in the brain of cyclo- phosphamide treated animals expressed as a percent of contrOI values 0 O O O O O O O O O C O O 0 O O 6 O O O 0 0 DNA, RNA and protein synthesis in the carcass of cyclo- phosphamide treated animals expressed as a percent of cont r0 1 values 0 O 0 0 O O O I O O O O O O I O O O I O I O Electrophoretic analysis of proteins synthesized in vitro by neonatal brain (3Hvleucine) and liver (14C- leuc inc) 0 C O O O O O O I O C O C O O O O O O C C O O O . Electrophoretic analysis of proteins synthesized in vitro by livers from cyclophosphamide treated mice (3H-leucine) or livers from control mice (14C-1eucine) one day after treatment. . . . . . . . . . . . . . . . . . vi Page 46 52 54 59 62 64 68 7O 72 89 91 Figure 12 316 ml: pha tre Figure Page 12 Electrophoretic analysis of proteins synthesized in vitro by livers from control ( 3H—leucine) or cyclophos- phamide treated (1 40-1eucine)'mice 3 days after treatment. . . . . . . . . . . . . . . . . . . . . . . . . 93 vii INTRODUCTION Historical Considerations Embryology, the branch of science dealing with animal development, studies the ontogenetic development of an individual rather than the phylogenetic development, or evolution, of a species. Embryological studies were conducted as early as 340 8.0. when Aristotle described development of the chick (reviewed by Balinsky, 1970). The first coherent presentation of embryological observation, however, was not made until 1828, when von Baer summarized existing data and reported on his original observations. A purely descriptive approach to embryology, however, failed to answer many of the basic questions concerning developmental processes involved in the transformation of a fertilized egg into an adult. Roux (1850-1924) introduced experi- mental methodology to embryology in order to resolve opposing theories of development. The principles and techniques of molecular biology have extended the scope of embryology to an understanding of basic processes associated with genomic control of early embryogenesis (reviewed by Davidson, 1968). Teratology is the study of congenital malformations produced by disruption of normal developmental processes. Embryological principles were applied to the study of malformations in the 18008 as teratolo- gists searched for the causes of anomalies (reviewed by Barrow, 1971). Various physical insults were applied to nonrmammalian species in order to gain insight into the mechanisms of normal and abnormal development. 1 2 Mammalian experimental teratology did not begin until the late 19403 and 19508 when fetal development was demonstrated to be disrupted by vitamin deficiency, irradiation, hormones and nutritional deprivation. After the thalidomide episode animal screening programs were devised to meet the need of predicting the teratogenic potential of various agents in humans. The greatest disadvantage of extrapolating animal observations to humans, however, is the interspecific variability of responses. The predictive value of animal observations can be improved only by embryological and pharmacological investigations into the mechanisms by which development is disrupted (Tuchmann-Duplessis, 1972). Phases of Development The development of an adult organism from a single cell may be divided into six phases (Balinsky, 1970). The first phase of develop- ment, gametogenesis, involves the growth and maturation of egg and sperm. The gametes fuse during the second phase of development and the quiescent egg is activated to continue its developmental program. Cleavage, the third phase of development, encompasses a period of rapid cell division without a significant change in embryonic size or cellular differentiation. The embryo, at the end of cleavage, is referred to as a blastula and consists of a layer of cells, the blastodenm, sur- rounding a cavity, the blastocoele. Gastrulation is the fourth develop- ment phase which encompasses the formation of germinal layers endoderm, mesoderm.and ectoderm from the blastoderm. Primary organ rudiments are derived from the germinal layers during organogenesis, the fifth phase of development. Whole organ systems are formed from primary organ rudiments which subdivide to produce secondary organ rudiments that give rise to subordinated and simpler organs. The sixth phase of Y1. L‘IIi f- F..I-.I.. r I MA 3 development is a period of growth and histological differentiation. The organ rudiments grow during this period and acquire the structure and biochemical prOperties characteristic of adult tissues. Organs grow by increasing both the number and size of cells and tissue specific characteristics are established as a result of differential gene expression. Processes of Development Processes involved in the development of an organism include cell division, differential gene expression, changes in the surface proper- ties and shapes of cells, and cellular metabolism. Cell division underlies all the phenomena of multicellular development since cell differentiation usually begins as mitosis ends (Berril, 1971). Cell multiplication, in addition, is an important aspect of morphogenesis. Cell division is one phase of the cell cycle which may be divided into mitosis (M), a period prior to DNA synthesis (01). a time of DNA synthesis (8), and a period between DNA synthesis and mitosis (02). DNA synthesis in eukaryotic cells is dependent on chemical substances located in the cytoplasm, protein synthesis, and structural properties of the chromo- some (Mitchison, 1971). The differentiated state is a function of differential gene expres- sion. The production of unique cellular phenotypes from a common genotype occurs either by genetic control mechanisms at the transcription and translation level or epigenetic control mechanisms such as protein ‘modification (Market and Ursprung, 1971). The differentiated state, however, ultimately depends on the transcription of genetic information which determines protein structure and the resulting cellular character- istics (Davidson, 1968). N. I‘ll lll‘klll‘l III‘ II. 4 Changes in the surface properties and shape of cells are an important aspect of differentiation and morphogenesis. The adhesive properties of cell membranes probably determine whether cells remain in a tissue or move about in the body (Markert and Ursprung, 1971). As development progresses cells establish connections which maintain tissue integrity and permit intracellular communication. Changes in cell shapes are important in morphogenic cell movements during gastrulation (reviewed by Wessels et aZ., 1971). Development is a dynamic rather than a static process. Cellular metabolism during development, therefore, must provide not only for maintaining the tissue but also must prepare the tissue to conduct adult physiological functions. Cellular metabolismtmay be subdivided into catabolic and anabolic processes (Lehninger, 1970). Catabolism is the enzymatic degradation of relatively large nutrient molecules, which are derived either from the environment or from.the cells' own nutrient storage depots into smaller simpler molecules. Anabolism is the enzymatic synthesis of relatively large molecular components of the cell from simpler precursors. Principles of Teratology Congenital malformations have been described in many mammals. There are embryological and non-embryological factors which determine an organism's susceptibility to an agent (Wilson, 1965). Pharmacologic factors include the agent, dose, and maternal physiology. A teratogen may produce specific effects related to metabolic events occurring during animal development. All doses of a teratogen, however, do not disrupt development. Low doses generally permit normal development while higher doses kill all of the embryos and even the mother if ex is de ta ex of Th 38 vs ge: St. bii hat co: or; pr. is X F-b 5 extended high enough. There is a teratogenic zone in which the dose is sufficient to interfere with specific developmental events without destroying the whole embryo. The mother is important not only in main- taining the embryo but also in determining the maternal-embryonic exchange of teratogens. Embryonic factors which influence the action of teratogens are genetic composition and gestational age at treatment. The genotype determines the inherent susceptibility of an embryo to agents at a given time in development. Susceptibility to teratogens varies during the course of gestation. There is generally no terato— genic response to agents administered during cleavage and blastula stages in mammalian embryos. The beginning of teratogenic susceptie bility occurs about the time of germ layer formation. Most organs have a period of particular susceptibility to teratogens which probably coincides with the early and critical developmental events for that organ. Agents which disrupt development may initially act at any of the processes essential for normal development, e.g., cell division, dif- ferential gene expression, changes in the surface properties and shapes of cells, and general cellular metabolism. Examples of Agents that Disrupt Development Inhibitors of Cell Division. Cytosine arabinoside (l—B-D—arabino- furanosyl cytosine), a structural analog of cytidine, inhibited cellular proliferation and DNA synthesis in a variety of systems (reviewed by Chaube and Murphy, 1968). Cytosine arabinoside reduced DNA.synthesis either by inhibition of cytidine diphosphate reduction to deoxycytidine diphosphate by ribonucleotide diphosphate reductase (Chu and Fisher, 1962) or by inhibition of DNA polymerization as a result of the V“‘”“h *"‘“‘I fort Ina: the eff: aduu witi esta DNA redL l hc on c aft: IIEE IE 1‘ a c; dUe 6 formation of cytosine arabinoside triphosphate (Furth and Cohen, 1968; Inagak et al., 1969). Cytosine arabinoside produced anomalies which were dependent on the gestational age (reviewed by Chaube and Murphy, 1968). Teratogenic effects were observed when 20 to 80 mg/kg cytosine arabinoside was administered on gestational days 10 to 12 but not when rats were treated with 5 to 500 mg/kg on days 5 to 9 of gestation. Ritter et al. (1971) established a correlation between the degree and duration of inhibited DNA synthesis and severity of malformations in rats. DNA synthesis was reduced in a dose related fashion to less than 25% of control values at 1 hour after treatment with 25, 50, 100 or 200 mg/kg cytosine arabinoside on day 12 of gestation. Recovery to control values occurred by 15 hours after lower doses but remained depressed for at least 27 hours after treatment with 200 mg/kg. An apparent cause and effect relationship between inhibited DNA synthesis and abnormal development was not estab- lished since a 25 mg/kg dose of cytosine arabinoside, which produced a cumulative 78% reduction in DNA synthesis over 24 hours, failed to result in anomalies. These results suggested that embryos either had a capacity to repair drug induced damage or that malformations were due to factors other than inhibition of DNA synthesis. The teratogenicity of hydroxyurea was investigated in order to clarify the role of reduced DNA synthesis in abnormal development (Scott et al., 1971). Hydroxyurea inhibited DNA synthesis in mouse embryo cells and reduced the formation of deoxyribonucleotides by ribonucleotide-diphosphate reductase (Skoog and Nordenskjbld, 1971). Hydroxyurea produced a dose related incidence of malformations in pregnant rats treated with 500, 750 or 1000 mg/kg on day 12 of gesta- tion. Reduced DNA synthesis, which was nearly complete at 5 hours and 7 lasted for 23 to 29 hours, was correlated with the teratogenic response. A 250 mg/kg dose of hydroxyurea produced a degree of DNA inhibition which did not result in malformations. The effects on DNA synthesis were similar but not identical to those produced by cytosine arabinoside (Ritter et al., 1971). The majority of defects were similar for both agents, e.g., hydrocephalus, micrognathia, ectrodactyly, short kinky tail, cleft, hydroureter, hydronephrosis, fused or wavy ribs, anal atresia, diaphragmatic hernia, and hypoplastic lungs. Hydroxyurea, however, produced defects not observed after cytosine arabinoside, e.g., cardiac and aortic defects, exophthalmia, and cranial dysplasia. Cytosine arabinoside and hydroxyurea both induced cellular damage in rat embryos (Scott et al., 1971). Cell death, however, was a normal developmental process which influenced both tissue differentiation and organ shaping (reviewed by Menkes et al., 1970). Massive cellular death in the appendages, for example, was involved in the normal formation of the knee joint, elbow, muscles, bones of the hand and foot, digits and general shape of the embryonic limb. Limb malformations, after hydroxyurea, were attributed to extensive death of cells in the forelimb buds (Scott et al., 1971). Normal cell death differs from drug induced cell death because the former is a regulated specific process while the latter is more general- ized. The production of specific types of anomalies by drug treatment, therefore, requires some degree of selectivity in drug action. Agents which inhibit DNA synthesis, for example, may have selective toxicity in rapidly dividing cells. Qualitative aspects of DNA synthesis, in addition, may confer a degree of selectivity to drug action since hydroxyurea had more cytotoxicity in continuously dividing cells than in cells induced to divide (Farber and Baserga, 1969). Rapidly ‘NIN‘IIIII NN‘II 8 proliferating tissues, therefore, may show the largest amount of drug- induced cellular death which may result in malformations depending on the regenerative capacity of the tissue. Cytosine arabinoside and hydroxyurea have been proposed to disrupt normal development by reducing the proliferating cell populations of specific tissues (Scott et aZ., 1971). These tissues at later stages of development may, therefore, have an insufficient number or type of cells to conduct the genetically prescribed morphogenetic role of the affected tissue or primordium. The proposal, however, fails to provide an explanation as to how cell death and reduced rates of proliferation lead to malformations. Inhibitors of Gene Expression. Actinomycin D, an antibiotic which reduced the transcription of RNA from DNA (Wheeler and Bennet, 1962), was teratogenic in a variety of species (reviewed by Chaube and Murphy, 1968). Developmental defects were produced in the echinoderm, acetabu- 1aria, chick, frog and rat. Actinomycin D was teratogenic in rats during the early period of organogenesis. Rats treated with actinomycin D from day 2 to 6 of gestation had an increased incidence of resorptions but no fetal anomalies. Anomalies were produced, however, in rats treated on gestational days 7 to 9 but not 10 or 11 with acinomycin D in doses of 50 to 75 mg/kg and 0.25 to 5.0 mg/kg, respectively (TuchmannrDuplessis, 1969; Chaube and Murphy, 1968). Sea urchin development continued normally in the presence of actinomycin D to the blastula stage when a drug induced inhibition of synthesis of specific proteins was detected (Terman, 1970). Normal development continued to the blastula stage as a result of sustained protein synthesis which was directed by maternally formed long-lived i"'”3*t messenge mamalia of the I trast o actinom Au D, admi: DNA and bition I mental I tiatton acinomyI treatmet irectec events a developI ClI Elie by EXpreSs. Clef: 13‘ (KaltEX “Flam of PalaI Tri day 11,5 402 in C increaSet palate we 9 messenger RNA. Cleavage and early blastular morphogenesis during mammalian development, however, were dependent on the early activity of the embryonic genome (Mintz, 1964). Mammalian development, in con- trast to sea urchin development, was disrupted at earlier stages by actinomycin D. Autoradiographic studies in rats demonstrated that 3H-actinomycin D, administered on gestational days 7 through 11, was bound to embryonic DNA and inhibited RNA synthesis (Jordan and Wilson, 1970). The inhi- bition of RNA synthesis was postulated to disrupt closely timed develop- mental events which occurred during the period of germ layer differen- tiation and early organogenesis. Since the teratogenic effects of acinomycin D occurred relatively early during organogenesis drug treatment may affect the synthesis of messenger RNA molecules which directed protein synthesis at later stages of development. The actual events associated with the disruption of both cell function and normal development, however, are unknown. Glucocorticoids have been postulated to produce cleft palate in mice by an action at the transcription level which interfered with gene expression (Zimmerman et al., 1970). The frequency of cortisone induced cleft palate was dependent on the genotype, dose, and gestational age (Kalter, 1954)° These teratogenic properties of cortisone may be explained, in part, by strain differences in drug metabolism and time of palate formation. Triamcinolone acetonide, administered at a dose of 5 mg/kg on day 11.5 of gestation, produced 1002 incidence of cleft palate in A/J, 402 in C3H, and 02 in CBA mice (Zimmerman and Bowen, 1972a). The increased resistance of CBA over A/J and C3H mice to drug induced cleft palate was attributed to a greater maternal metabolism of triamcinolone 10 by this strain. The CBA embryos, as a result, were exposed to lower levels of the parent compound which was shown to be the teratogenic form of triamcinolone (Zimmerman and Bowen, 1972a, 1972b). The greater resistance of C3H over A/J mice to drug-induced cleft palate formation, however, was probably related to a developmental rather than metabolic factor. The sensitivity of A/J mice, in contrast to CBmeice, may be due to a more delayed palate formation which has been correlated with increased teratogenic susceptibility (Zimmerman and Bowen, 1972a; Trasler, 1965). Palate formation in mice involved both movement of the shelves of the secondary palate from a vertical position beside the tongue to a horizontal position above the tongue and fusion at the midline (Walker and Fraser, 1956). Glucocorticoids, which delayed shelf movement, produced cleft palate by a desynchronization of developmental events (walker and Fraser, 1957). If shelf movement was delayed sufficiently then head growth would have carried the shelves too far apart to meet and fuse at the midline. Palatal shelves changed their orientation between day 14.5 and 15.5 of gestation in mice (walker and Fraser, 1956). There was, there- fore, about a three day latent period between the time of optimal drug sensitivity and palate closure. The delay between drug administration and effect led to the proposal that glucocorticoid-induced cleft palate was produced by a reduced synthesis of messenger RNA molecules which directed the synthesis of proteins required for palate closure (Zimmerman et al., 1970). Triamcinolone, in support of this theory, inhibited total embryonic RNA synthesis 6 to 24 hours after treatment. The coadministration of cycloheximide, in addition, partially reversed the tendency toward cleft palate formation. These observations suggested 11 that triamcinolone disrupted the normal pattern of RNA and protein synthesis which subsequently delayed the movement of palatal shelves from a vertical to a horizontal position. Glucocorticoids, in addition to disrupting development during the phase of organogenesis, also affected the subsequent postnatal phase of growth and histological differentiation. Rats that received 1 mg of cortisol at 2 days of age grew at a reduced rate, were more nervous and developed a blotchy skin with areas of hyperpigmentation and sparse hair (Schapiro, 1965). Corticosterone administered to mice at doses of 0.08 to 0.6 mg/day between 2 and 7 days of age produced a significant reduction in liver, brain, and body weights at the end of treatment (Howard, 1965). Corticosterone treatment, in addition, reduced the accumulation of total liver and brain DNA (Howard, 1965, 1968). These effects were evident by 7 days and persisted for at least 8 months. A pellet of corticosterone, 1.3 mg/g body weight, implanted in 2-day- old rats, produced a reduced body weight with no change in pituitary and liver weight relative to body weight or liver DNA and RNA/DNA ratios (Taylor and Howard, 1971). Treated animals, in addition, had a short snout and a 4 to 52 reduction in nose to anus length. Cortisol treatment at doses of 0.2 mg/day for the first 4 days after birth reduced the incorporation of 14C—thymidine into brain DNA but did not change DNA concentration or RNA/DNA and protein/DNA ratios (Cotterrell, 1972). The corticosterone induced biochemical lesions were associated with impaired brain function and behavior (Howard and Granoff, 1968a). Treated mice, for example, had a permanent impairment of fine adjust- ment mechanisms of motor control. The effect of cortisol on the nervous system was age dependent (Vernadakis and Woodbury, 1971). Acute or chronic drug treatment in ——’ V72.— rats theI on d theI OXYS trat impa plas 1972 tion thrc 1106 pro; difi whiI am of Var Ce} [QC Si: 12 rats between 1 and 7 days after birth produced a delayed maturation of the nervous system and reduced brain excitability. These parameters, on the other hand, were increased when cortisol was administered after the 8th postnatal day. These effects were accompanied by changes in oxygen consumption and intracellular to extracellular potassium concen- trations. Rats treated with glucocorticoids at 2 days of age had impaired glucose homeostasis and suppressed circadian periodicity of plasma corticosteroids at maturity (Taylor and Howard, 1971; Krieger, 1972). Treatment at 12 to 14 days of age, on the other hand, produced no effect on circadian variations of corticosteroids. These observa- tions indicated that postnatal development in the brain progressed through various developmental stages which differed in susceptibility to drug treatment. ‘Aggnts that Affect Cell Surfaces and Shapes. Changes in the properties of cell surfaces represent an important aspect of embryonic differentiation and morphogenesis (Market and Ursprung, 1971). Agents which affect cell surfaces have the potential to disrupt development, although there is little teratological data available for this class of compounds. Lectins are protein molecules that specifically bind to various carbohydrates on the cell surface and cause agglutination (reviewed by Sharon and Lis, 1972). Concanavalin A is a powerful non- specific lectin which forms a precipitate with numerous polysaccharides. The concanavalin A binding properties of embryonic and adult cells have been shown to be different (Moscona, 1971). Dissociated embryonic cells were readily agglutinated by concanavalin A.while adult cells required an unmasking of receptor sites by trypsin digestion before agglutination occurred. The presence of exposed concanavalin A binding sites on embryonic cell surfaces suggested that these sites functioned duri norm cell been tubu and meat sySI forI DEL but wit Sir 2-1 po: of me- Pt in 13 in morphogenetic cell contacts, cell mobility, and tissue organization during embryonic development. .Agents which block these receptors during normal embryonic development may contribute to the production of anomalies. Microfilaments are contractile elements of cells which influence cell shape (reviewed by weasells et al., 1971). These structures have been shown to play important roles in morphogenic cell movements during tubular gland formation in the oviduct, formation of salivary epithelium and invagination during gastrulation. Cytochalasin B disrupted develop- mental events which depended on the integrity of the microfilament system. Agents which affect the function of the microfilaments, there- fore, may have teratogenic effects. .Agents that Affect Cellular Metabolism. The integrity of cellular metabolism is important not only in maintaining the viability of cells but also in permitting phenotypic expression of cellular characteristics. Agents which disrupt metabolic functions, therefore, possess a terato- genic potential. A glucose antimetabolite, 2-deoxy-D—glucose (2-DG), disrupted chick development as evidenced by a reduced hatchability 'without an increased incidence of anomalies (Landauer and Clark, 1962). Since glucose treatment reversed these effects it was proposed that 2-DG interfered with the utilization of glucose. Rat embryos, in sup- port of this proposal, treated with 2-DG had a reduced incorporation of radioactivity from 14C-U’-g1ucose into both protein and nucleic acids (Tanimura and Shepard, 1970). An adequate diet is an essential requirement for normal develop- ‘ment. Diets deficient in vitamins A, riboflavin, 312, pantothenic acid, pteroylglutamic acid, or E produced an increased incidence of anomalies in rats (reviewed by Johnson, 1965). The critical period of sensitivity to def inc 24 CH an: gr 0U Ii t0 bz TE 14 to the teratogenic diet, as revealed by antimetabolite induced vitamdn deficiency, was the period of differentiation and organogenesis. Malnutrition, induced by food deprivation, contributed to an increased incidence of congenital defects. MHoe deprived of food for 24 or 30 hours on days 7 to 10 of gestation had vertebral, rib and cranial defects (Runner and Miller, 1956). The highest incidence of anomalies was produced by fasts on day 9 of gestation. Forty-hour fasts, in addition, were usually incompatible with pregnancy. Nutritional deprivation of neonatal animals affected subsequent growth and differentiation. Mice separated from their mothers for 16 out of 24 hours from 3 to 7 days of age had reduced body, brain, and liver weights at 7 days of age (Howard, 1965). Brain growth, relative to liver and body growth, was spared during this period since the brain to body weight ratio increased while the liver to body weight ratio was not affected. The concept of a brain sparing effect during malnutrition was observed by others (reviewed by Dobbin, 1970). Reduced cerebral and cerebellar weights and total brain DNA was observed at 9 months of age in mice isolated overnight from their mothers between 2 and 16 days of age (Howard and Granoff, 1968b). There was no impair- ment in voluntary running, maze learning, or visual discrimination in isolated mice despite impaired brain growth. Malnutrition was reported in other studies to delay behavioral development and affect learning ability (reviewed by Dubbing, 1970). The magnitude of nutritional deprivation and temporal aspects of its administration determined both the extent of damage and recovera- bility of tissue following the initiation of an adequate diet (Winick and Ndble, 1966). Organ damage and recoverability, after malnutrition, were dependent on the growth phase of organs at the time of malnutrition. d q a in w in 1161 the the 15 Developmental growth may be divided into three phases which occurred in characteristic patterns for each organ (Winick and Noble, 1965). During the first phase, a period of rapid DNA synthesis, cell number increased while cell size remained constant (hyperplasia). As the rate of cell division slowed down, during the second phase, the size of individual cells increased and organs grew by increased cell number and cell size (hyperplasia and hypertrophy). During the third phase organs grew as a result of the enlargement of individual cells (hypertrophy). Organs from undernourished rats had a reduced weight, protein, RNA, and DNA content, which was not restored by refeeding, when food was restricted during the period of rapid DNA synthesis (Winick and Noble, 1966). The effect of malnutrition on brain development, for example, was reversed by feeding when food deprivation occurred both at 21 to 41 and 64 to 86 days of age but not at 0 to 21 days of age when the organs grew by hyperplasia. The developmental increase in protein synthesis, which occurred during hypertrophy, may be regulated, in part, both by hormones and by changes in the intracellular amino acid pools. Enzymic differentiation referred to the processes involved in prenatal and early postnatal development whereby the organs of an animal acquired their characteristic quantitative pattern of enzymes (reviewed by Greengard, 1969). The accumulation of various liver enzymes (e.g., glucose-6-phosphatase) immediately after birth may be initiated by glucagon and epinephrine which may act through the formation of cyclic AMP. Changes in the intracellular amino acid pool were proposed to regulate the develop- mental increase in protein synthesis (Miller, 1970). An increase in the amdno acid pool, for example, may increase polysomal stability and the binding of ribosomes to the endoplasmic reticulum. Since growth re 1'3 tr CV pr n E th s ,5 - .HLQ“... Ifi'EIEw _ _ r. :12 V wh de ti V8 c1 r1 0f 5 16 retardation produced by malnutrition was correlated with a protein, rather than caloric deficiency, a reduced amino acid pool may have cone tributed to impaired organ growth. Cyclophosphamide Bacgground. Cyclphosphamide (N,Nbbis-(2-chloroethy1)-N',0- propylene phosphoric acid ester diamine monohydrate, CytoxanR, EndoxanR, NSC 26271) is a clinically useful antineoplastic agent of the alkylating type with potent toxic properties in developing systems (see below). The molecule, which consists of nor-nitrogen mustard to which a cyclic phosphoramide ring structure has been attached, was designed by Arnold and Bourseaux (1958) to require metabolic activa- tion for the production of alkylating metabolites. The parent compound was proposed to be activated primarily within tumors by the hydrolytic cleavage of the P-N bond between nitrogen mustard and the phosphamide ring by phosphatases and phosphamidases. Metabolism. The requirement of bioactivation for the production of cytotoxicity was demonstrated by Foley et a2. (1961). Cell growth was inhibited when tissue cultures were treated with sera from rats administered cyclophosphamide but not by cyclophosphamide added directly to the sera. Homogenates of livers but not neoplastic tissue from cyclophosphamide treated rats produced cytotoxicity in tissue culture. Alkylating cyclophosphamide metabolites were demonstrated in serwm, urine, and bile of animals and humans by means of the nitrobenzyl pyridine (NBP) colorimetric assay (Brock and Hohorst, 1963). The maximum NBP activity in rat serum was reached 15 minutes after cyclo- phosphamide (1 g/kg) and dropped to 502 of maximum after 4 hours, and NBP activity was not detectable 24 hours after treatment. mili‘ , . Cyc to a les 1963). activat: recluire< form) a: to NBP-z (beta- 1 phenoba: suggest‘ primaril describe 17 Cyclophosphamide activation occurred primarily in the liver and, to a lesser extent, in the adrenal cortex and lung (Brock and Hohorst, 1963). Liver fractionation studies indicated that the cyclophosphamide activating system was located primarily in the microsomal fraction and required NADPH (nicotinamide adenine dinucleotide phosphate, reduced form) and molecular oxygen. The in viva conversion of cyclophosphamide to NBP-alkylating metabolites, in addition, was decreased by SKF SZS-A (beta- diethylaminoethyl diphenylpropylacetate) and increased by phenobarbital pretreatment (Rauen and Kramer, 1964). These studies suggested that the metabolic activation of cyclophosphamide occurred primarily in the same mixed function oxidase system of liver microsomes described by Conney (1967) for the metabolism of many drugs and steroids. These conclusions concerning cyclophosphamide activation have been confirmed and extended. Activation occurred primarily, if not exclusively, in rat liver microsomes and required NADPH and molecular oxygen (Cohen and Jao, 1970). This process was competitively inhibited in vitro by compounds such as hexobarbital and steroids which were metabolized by the mixed function oxidase system. Cyclophosphamdde metabolism, in addition, was inhibited by carbon monoxide, which bound to the micro- somal P-450 hemoprotein as well as p-hydroxymercuribenzoate and cyto- chrome C which disrupted the flow of charge along the microsomal electron transport chain. Sladek (1971), in addition, demonstrated that cyclo- phosphamide combined with the microsomal P-450 hemoprotein to produce a Type I spectral interaction (A max. in the range of 385-390 my and a A min. in the range 418-247 mu). Arnold and Bourseaux (1958) originally proposed that cyclophos- phamide was hydrolyzed by tumor phosphatases and phosphamidases to produce nor-nitrogen mustard. Studies have reported that nor-nitrogen 18 mustard was a cyclophosphamide metabolite present in rat urine (Rauen and Dirschka, 1964) and generated in vitro by a drug metabolizing system (Rauen and Norpoth, 1965). Hohorst at al. (1966), on the other hand, was unable to detect nor-nitrogen mustard as a metabolite of cyclophosphamide.4 Friedman (1967) confirmed that free nor-nitrogen mustard was not detectable in rat tissue at various times after drug treatment . Hill et al. (1972) proposed a pathway for the metabolic activation of cyclophosphamide in mouse liver. The carbon adjacent to the ring nitrogen was oxidized by the cytochrome P-450 linked enzyme to form 4~hydroxycyclophosphamide which was subsequently oxidized to products containing carbonyl groups. Aldophosphamide, generated by the micro- somal system, inhibited clone formation of human epidermoid carcinoma No. 2 cells and produced toxicity in L1210 cells. Carboxyphosphamide was produced from.a1dophosphamide by a soluble enzyme which had the properties of aldehyde oxidase. This metabolite produced little toxicity in either human epidermoid carcinoma No. 2 cells or L1210 cells. The spontaneous cyclization of carboxyphosphamide produced 4-ketocyclo- phosphamide which had weak cytotoxic effects against human epidermoid carcinoma No. 2 cells and L1210 cells (Struck et al., 1971). Struck at al. (1971) used infrared spectroscopy, nuclear magnetic resonance and mass spectrosc0py to identify carboxyphosphamide and 4-ketocyclophosphamide in urine from dogs. They estimated that 40 to 501 and 25 to 302 of the total 24 hour urine radioactivity in dogs and humans, respectively, after 14C-cyclophosphamidewas due to carboxy- phosphamide. Unmetabolized cyclophosphamide in urine from dogs and humans, on the other hand, was 7% and 162, respectively, of the adminis- tered dose and approximately 11% and 272 of the urinary radioactivity. 19 Bakke et a2. (1972) identified eight cyclophsophamide metabolites in the urine of sheep after a single oral dose of radioactive cyclo- phosphamide. Cyclophosphamide was labeled either with 14C at the 4 position in the ring or with 38 in the side chain on the carbons having chlorine substitution. Metabolites represented approximately 95% of the urinary radioactivity from ring labeled cyclophosphamide and 881 of the radioactivity from side chain labeled cyclophosphamide. There was, in addition, no evidence that nor-nitrogen mustard was a urinary metabolite of cyc10phosphamide. These observations suggested that the ring portion of the molecule was not cleaved from the alkylating portion during metabolic activation in sheep. Pharmacological Considerations. The antineoplastic activity of cyclophosphamide was evident in various animal tumor systems. Drug therapy in rodents was beneficial in the treatment of leukemia (Lane, 1959; Burchenal et al., 1960), lymphosarcoma (Lane, 1960), mammary carcinoma (Lane, 1960), and sarcomas (Burchenal at aZ., 1960). Since cyclophosphamide resistant L1210 leukemia cells were also resistant to other alkylating agents but not antimetabolites, it was concluded that cytotoxicity was produced by an alkylating mechanism (Lane, 1967). The antitumor activity of cyclophosphamide, however, was not altered either by SKF 525-A pretreatment of mice bearing L1210 leukemia cells (Hart and Adamaon, 1969) or by phenobarbital pretreatment of rats with Walker 256 carcinoma cells (Sladek, 1972). Phenobarbital and SKF 525-A pre- treatment, in addition, did not produce any therapeutic advantage over cyclophosphamide alone in the treatment of mice bearing L1210 leukemia cells or Lewis lung carcinoma (Field at al., 1972). The failure of drug pretreatments to provide a correlation between cytotoxicity and alkylating metabolites may be due to their failure to significantly 20 alter the total exposure of cancerous cells to alkylating metabolites. Phenobarbital pretreatment, for example, produced only a slight change in the total recovery of a cyclophosphamide dose as urinary metabolites (Jao at al., 1972). Cyclophosphamide has found wide clinical use in the treatment of a wide variety of malignant and nondmalignant diseases in humans. Drug therapy was effective in the treatment of acute childhood leukemia resistant to antimetabolite therapy (Brubaker at al., 1962; Pernhach at al., 1962), chronic lymphatic leukemia (Hall, 1963), recurrent and. metastatic ovarian cancer (Decker at aZ., 1967; Beck and Boyes, 1968) and advanced lung cancer (Bergsagel at al., 1968). Cyclophosphamide, in addition, has been used as an immunosuppressive agent in lipoid nephrosis (west et al., 1966), nephrotic syndrome in children (Drummond at al., 1968), and whole organ transplantation (Starzl at al., 1971). Cyclophosphamide is administered in doses of 2 to 8 mg/kg orally or intravenously for 6 days followed by maintenance doses of 50 to 300 mg daily or 5 mg/kg intravenously twice weekly. High dose cyclophosphamide therapy (1 to 6 doses of 60 and 120 mg/kg intravenously) with marrow infusion has been successfully employed in the treatment of a variety of malignant diseases (Buckner at aZ., 1972). Cyclophosphamide did not have selective toxicity for tumor tissues. Rats had a 30 day oral LD of 170 mg/kg for cyclophosphamide (Dixon, 50 1968). Nice, on the other hand, had a 28 day subcutaneous LD50 of about 400 mg/kg (Hart and Adamaon, 1969). Agents which inhibited metabolic activation reduced cyclophosphamide lethality in both mice (Hart and Adamaon, 1969) and rats (Dixon, 1968). Increased metabolic activation, on the other hand, was associated with a hastened onset of lethality but no change in the LDso (Hart and Adamaon, 1969). *A-H‘I—I 4'1". ~.._ .- . .‘ . 21 Cyclophosphamide produced a progressive wasting in rats treated with single subcutaneous doses of 65, 85 or 110 mg/kg (Stenram and Hordlinder, 1968). The animals died primarily between 45 and 91 days after treatment presumably as a result of toxic effects on the hemato- poietic system. Drug treatment produced bone marrow depression (Lane, 1959) with minimal effects on platelets (Coggins at al., 1960; Bisman at aZ., 1961). Cyclophosphamide induced leukopenia was correlated with the presence of alkylating metabolites in.mice pretreated with agents to increase or decrease metabolic activation (Hayes at al., 1972). Cyclophosphamide, in addition, produced hemorrhagic cystitis (Kaufman, 1963; George, 1963; Specter et al., 1965), alopecia (Coggins at al., 1960; Bisman at al., 1961), and urinary bladder epithelial necrosis which was more pronounced in male rats (Ross and Lavin, 1970). Toxicity of Cyclophosphamide DuringgDevelopment Gametogenesis. Cyclophosphamide disrupted various phases of normal ontogenetic development. Humans who received 50 to 100 mg/day of cyclophosphamide had low sperm counts in their seminal fluid (Fairley at al., 1972). The sperm count fell to zero in some patients after 4 months of treatment and all patients who received 6 months of therapy showed azoospermia. The germinal epithelia of mice treated with 97 ms/kg of cyclophosphamide showed a shift in metaphase ratios and an increased incidence of chromosomal aberrations (Rathenberg and Muller, 1972). These observations suggested that cyclophosphamide was poten- tially mutagenic. The mutagenicity of cyclophosphamide was confirmed by the dominant lethal test in mice (Epstein et al., 1972). 22 Prenatal Development. Alkylating agents disrupted normal embryonic development in a variety of species. Nitrogen mustard, for example, produced anomalies in amphibians (Gillette and Bodenstein, 1946), mice (Danforth and Center, 1954; Nishimura and Takagaki, 1959), and rats (Haskin, 1948). Fetuses from.rats treated with nitrogen mustard at doses of 0.5 or 1.0 mg/kg on days 12, l3, 14 or 15 of gestation were reduced in size and weight and had cranial faults, underdeveloped jaws, cleft palates, fusion of digits on front and hind limbs, abnormal posture of hind limbs, and shortness of the tail. Murphy and Karnofsky (1956) confirmed these results and showed that increased embryonic and fetal lethality was produced by treatment earlier in gestation. Five polyfunctional alkylating agents and erays were found to produce similar teratogenic effects in rats (Murphy, 1959). Cyclophosphamide was teratogenic in a variety of mammalian species. Rabbits treated subcutaneously with 50 mg of cyclophosphamide on days 10 through 13 of gestation produced fetuses with reduced size and increased incidences of resorptions, cleft palates, and hypognathus (Gerlinger, 1964; Gerlinger and Clavert, 1965a). Higher doses of cyclophosphamide produced complete fetal resorption and lethality in one dam. A 50 mg/rabbit dose of cyclophosphamide on day 10 or 13 of gestation produced abnormal gonadal differentiation which was more pronounced in females (Gerlinger and Clavert, 1956b). Cyclophosphamide embryotoxicity was evident in early rabbit embryonic development while teratogenicity was confined to the later period of organogenesis (Fritz and Hess, 1971). Cyclophosphamide teratogenicity was studied in the rat by Brock and Kreybig (1964) and extended by Kreybig (1965). Pregnant rats treated intraperitoneally with single doses of 20 to 40 mg/kg 23 cyclophosphamide on days 10 through 15 of gestation produced fetuses with exencephaly, microcephaly, syndactyly, polydactyly, and missing ribs. The critical period for teratogenicity was days 12 and 13 while resorptions were observed on days 11 through 14 of gestation. Genetic differences in the teratogenic response to cyclophosphamide were not evident in the anomalies produced in six strains of rats (Kreybig, 1965). The in utero cytological and histological effects of cyclophosphamide were studied in rats treated on day 13 (Kreybig and Schmidt, 1966) and days 15 and 16 (Kreybig and Schmidt, 1967) of gestation. General cellular damage such as karyorrhexis, karyopyknosis, hyperchromatosis, and abnormal immigration of embryonic cells were evident 24 hours after treatment. Necrotic substances diminished and repair occurred during the subsequent 24 hour period. Mice, in addition, were susceptible to the teratogenic effects of cyclophosphamide. Embryonic development of DM/MK mice was disrupted by cyclophosphamide, but this strain appeared to be more resistant than other previously studied species (Shoji and thu, 1965). Swiss webster mice, on the other hand, were more susceptible to cyclophosphamide teratogenicity (Gibson and Becker, 1968a). Cyclophosphamide administered intraperitoneally at single dose of 5 and 10 mg/kg on days 10 through 15 of gestation produced decreased growth and increased resorptions. Single dose of 20 mg/kg administered on the same schedule produced a variety of teratogenic effects in addition to a reduction in growth and an increase in lethality. The anomalies produced included gross defects (cleft palate, exencephaly, digital defects and kinky tail), skeletal anomalies (polydactyly, syndactyly, ectrodactyly, adactyly, fusion of the long bones, curvature of the long bones, and missing ribs), and soft tissue malformations (open eyes, aphakia, microphakia, hydronephrosis, 24 and hydrocephalus). The maximum teratogenic response was produced by treating mice intraperitoneally with 20 mg/kg cyclophosphamide on day 11 of gestation. The role of metabolic activation in the production of cyclophos- phamide teratogenicity was investigated in Swiss-Webster mice pretreated with either SKF 525-A or phenobarbital (Gibson and Becker, 1968b). SKF 525-A pretreatment significantly increased fetal weight loss and the incidence of resorptions, open eyes, polydactyly, aphakia, hydro- nephrosis, absence or non-ossification of ribs, bone fusion, and bone curvature. Phenobarbital pretreatment, on the other hand, signifi- cantly reversed the cyclophosphamide induced weight loss as well as the incidence of limb defects, hydrocephalus, exencephaly, cleft palate, aphakia, and hydronephrosis. As a result of these studies it was pro- posed (Gibson and Becker, 1968b) that cyclophosphamide teratogenicity was associated with the parent compound rather than with alkylating metabolites. The proposal that the cyclophosphamide parent compound had toxic properties during development received support from additional studies. The placental transport of ring labeled JAG-cyclophosphamide, under conditions of altered metabolic activation, showed that both terato- genicity and embryonic levels of radioactivity were decreased by pheno- barbital and increased by Sid" 525-A pretreatment (Gibson and Becker, 1971a). These observations indicated that there was a placental barrier to the transport of cyclophosphamide metabolites and suggested that teratogenicity was associated with the parent compound. Embryos and Placentas, in addition, had a poorly developed in vitro ability to metabolically activate the parent compound (Short and Gibson, 1971a). Gibson and Becker (1971b) studied the teratogenicity of structural 25 analogs of cyclophosphamide which were proposed as metabolites or degra- dation products (Foley at aZ., 1961). ‘Minimal teratogenic doses, which also produced signs of maternal toxicity, failed to parallel the response to a teratogenic dose of cyclophosphamide. The above observations sup- ported the proposal (Gibson and Becker, 1968b) that the parent compound had unique toxic properties during embryonic development. Recent evidence, however, has implicated cyclophosphamide metabo- lites and the alkylation of DNA in the disruption of normal embryonic development (Murthy at al., 1971). The binding of 3H-cyclophosphamide to embryonic DNA, but not RNA or protein, was correlated with terato- genicity by measuring drug binding to macromolecules after various doses of cyclophosphamide and modification of its metabolic activation. Jusko (1972), as a result of an analysis of the dose embryolethality data of Gibson and Becker (1968a), classified cyclophosphamide as a Type 1 teratogen. Type 1 teratogens had no minimal embryopathic dose while Type 2 teratogens had minimal embryopathic doses. Since nitrogen mustard and cyclophosphamide were Type 1 teratogens this may indicate that embryos lack a sufficient capability to repair alkylated DNA. Postnatal Development. Cyclophosphamide disrupted development during the postnatal phase of growth and differentiation. ‘Mice treated with cyclophosphamide 24 to 48 hours after birth had delayed development of hair, short snouts, ears and tails, and reduced body weights at maturity (Nordlinder, 1969). Cyclophosphamide neonatal toxicity, in addition, was dose related and not duplicated by equimolar doses of nor-nitrogen mustard, an agent with inherent alkylating activity (Short and Gibson, 1971b). Since alkylating activity per se did not disrupt development and cyclophosphamide neonatal toxicity occurred at a time 26 when mice had a poorly developed activating ability (Short and Gibson, 1971a), it was proposed that the parent compound may have toxic proper- ties during postnatal as well as embryonic development. Recent evidence, however, indicated that neonatal toxicity was produced by an alkylating mechanism.(Bus at al., 1973). Phenobarbital pretreatment increased and SKF 5252A decreased various signs of cyclophosphamide toxicity. The slow elimination of radioactivity from.plasma following treatment of neonatal mice with 14C-cyclophosphamide suggested that the poorly developed activating system (Short and Gibson, 1971a) possessed a suf- ficient capacity to generate alkylating metabolites over a prolonged period of time. ,Qytotoxicity of Alkylating Agents. Alkylating agents are highly reactive electrophilic compounds which are capable of substituting an alkyl group for a hydrogen ion on nucleophilic centers of molecules. Nucleophilic centers which are reactive with alkylating agents include organic and inorganic anions, amino groups, sulfhydryl groups, and sulfide groups (Ross, 1958). Alkylating agents have been shown to react with protein (Alexander and Cousens, 1958). Nitrogen.mustards, esters of methane sulfonic acid, and epoxides reacted with carbonyl and histidine groups, in bovine serum albumin although only a fraction of the groups were accessible to the reagents. The most reactive group to all of the alkylating agents was the sulfhydryl group in denatured egg albumin. These groups, however, were not available for reaction in native protein. Alkylating agents react with nucleic acids. Alkylation of DNA occurred primarily at the N7 position of guanine (Lawley and Brookes, 1967). The cytotoxicity of alkylating agents was greater for 27 difunctional than monofunctional alkylating agents (Brookes and Lawley, 1961). Since the extent and duration of DNA alkylation has been cor- related with toxicity in Escherichia coli (Lawley and Brookes, 1965), the increased toxicity of difunctional alkylating agents may be due to the nature of their interaction with DNA. Monofunctional alkylating agents yielded 7 alkylguanines while difunctional agents produced, in addition, di(guanin-7-yl) derivatives (Lawley and Brookes, 1967). Approximately 252 of sulfur mustard that reacted with DNA gave the diguaninyl product which was suggestive of an interstrand cross linking. The increased cytotoxicity of difunctional alkylating agents, therefore, may be due to the formation of interstrand cross-links in the DNA duplex. A variety of systems have the ability to remove alkylated bases from DNA and restore DNA integrity. Venitt (1968) demonstrated strain differences in E. coli sensitivity to sulfur mustard which were dependent on the cells' ability to remove alkylated bases from DNA. DNA repair has also been demonstrated in the HeLa and Chinese hamater cells follow- ing treatment with a variety of alkylating agents (Roberts et al., 1971a, 1971b). Extensive biochemical lesions occurred in cells as a result of damage produced by alkylating agents. The primary lesion produced by sulfur mustard was an inhibition of DNA, but not RNA or protein, synthe- sis in HeLa cells (Roberts at al., 1968). Nitrogen mustard, on the other hand, reduced both DNA and RNA synthesis in hamster plasmacytomas (Wheeler and Alexander, 1969). The template activity of calf thymus DNA for E. coli DNA and RNA polymerase as well as the coding capacity of artificial messenger RNA was reduced by nor-nitrogen mustard treat- ment (Johnson and Ruddon, 1967; Ruddon and Johnson, 1968). A reduced 28 replication of cellular DNA induced by alkylating agents led to a mitotic delay as a result of cross-linking of the DNA duplex which prevented strand separation during replication (Roberts et al., 1968). Cells con- tinued to synthesize RNA and protein at normal rates, however, and grew into giant cells with abnormal amounts of RNA and protein. Radioactivity after the administration of ring-labeled cyclophos- phamide to hamsters bearing plasmacytomas was located in the nuclear, mitochondrial, and microsomal fraction of the plasmacytoma (Wheeler and Alexander, 1964). Single doses of cyclophosphamide inhibited the in viva synthesis of DNA.snd RNA 24 to 48 hours after treatment in the same tumor system (Wheeler and Alexander, 1969). The reduction in DNA synthesis was accompanied by an inhibition of in vitro DNA synthesis in a systemwwhich contained denatured salmon sperm.DNA and soluble plasma- cytoma enzymes. Several studies suggested that cyclophosphamide damaged DNA.was repaired. Dose fractionation studies in mice with L1210 leukemia cells indicated that sublethal damage was repaired between 3 and 12 hours (Dewys and Right, 1969). An increased priming capacity of plasma- cytoma DNA for DNA synthesis, in addition, was attributed to a cellular response of attempted DNA repair (Tomisek at al., 1971). Purpose The purpose of this study was to examine the molecular basis of drug-induced disruption of development in mice. Cyclophosphamide induced disruption of embryonic and neonatal development of mice was the model system studied. METHODS Animals Virgin Swiss-webster mice were obtained from Spartan Research Animals (Haslett, Michigan) and housed at 70 to 75'F in stainless steel cages with wire mesh bottoms. The animals were maintained on a 12 hour light-dark cycle, which began with lights on at 8 a.m., in order to synchronize estrus cycles. Mice were given free access to food (Purina Lab Chow) and water. Timed pregnancies were obtained by the daily mating of 150 to 200 females. One male was placed in a cage with 5 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 42 of the females were 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. Neonatal Swiss-Webster mice were obtained from.timed pregnancies started in this laboratory. Pregnant mice were individually housed prior to delivery in clear plastic shoe-box cages that contained a base of Bed-O-Cobs (Anderson Cob Mill, Maumee, Ohio). The mothers delivered spontaneously and were given free access to food and water. The litters were weaned at 3 to 4 weeks of age. 29 30 Biosynthesis of DNA by Embryonic Mice Cyclophosphamide was administered intraperitoneally (i.p.) at a terato- genic dose of 20 mg/kg on day 11 of gestation (Gibson and Becker, 1968a). Treated mice received the drug in a volume of 10 ml distilled water/kg body weight and controls received an equal volume of the vehicle. Thymidine-Z-MC (55.7 to 58.9 mc/mM, ICN Tracer Lab, Irvine, Calif.) was administered at a dose of 100 ‘uc/kg at various times after cyclo- phosphamide. Mice were sacrificed 1 hour after 14C-thymidine by cervical dislocation and a laparotomy was performed. The uterus was removed and the embryos and placentae were obtained through incisions in the uterine wall. JAG-thymidine incorporation into a DNA fraction obtained from maternal liver, placentae, and embryos by method A or B was used as an estimation of DNA synthesis. Method A (3 to 24 hours after cyclophosphamide). The tissues were removed, weighed, and homogenized in 5 m1 of 0.2N perchloric acid (PCA). The supernatant was neutralized with potassium hydroxide and saved for liquid scintillation counting. The precipitate was washed with succes- sive 5 ml portions of 0.2N PCA, which contained 4 mM non-radioactive thymidine to reduce the level of unincorporated radioactivity, 0.2N PCA, 951 ethanol saturated with sodium acetate, a 3:1 ethanol-ether solution, and ether. The precipitate was heated in 4 ml of 0.3N potassium hydroxide for 1 hour, acidified with 3 ml of 302 trichloro- acetic acid (TCA), stored at 4°C for at least 8 hours, washed with 5% TCA and heated for 20 minutes at 90°C in 5 ml of 5% TCA. The hot TCA extract was combined with an additional 5% TCA wash of the precipitate and this solution was used for the determination of radioactivity and DNA. The specific activity of this fraction was expressed as disinte- grations per minute (DPM)/mg DNA. 31 Radioactivity in 1 ml of the PCA or TCA extract was measured in 15 ml of a dioxane base counting solution (60 g naphthalene, 4 g 2,5- diphenyloxazole (FPO), 200 mg 1,4 bis [2-(4-methyl-5-phenyloxazolyl)] benzene (POPOP), 100 ml absolute methanol, 20 ml ethylene glycol, and dioxane to a final volume of 1 liter) using a Beckman LS-lOO system with a 137Cs external standard. DNA was measured in the TCA extract by the diphenylamine method described 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. Acetaldehyde (0.1 ml of a 16 mg/ml solution) was added to 20 ml of the diphenylamine reagent prior to use. A sample of the DNA extract was brought to a volume of 1 ml with 52 TCA prior to the addition of 2 m1 of the diphenylamine reagent. The colored product, which was formed by the reaction of deoxyribose with diphenylamine, developed for 16 to 20 hours at room temperature. The optical density of the solution was determined at 600 nm and DNA was quantified with a standard curve prepared from calf thymus DNA (Sigma, St. Louis, Mo.). Method B (24 to 72 hours after cyclophosphamide). Tissues were removed, weighed and homogenized in 5 ml of 101 TCA. The precipitate was washed with 101 TCA that contained 4 mM non-radioactive thymidine, 102 TCA, and heated for 5 min at 60°C in a 3:1 alcohol-ether solution. The precipitate was heated for 15 minutes at 90°C in 5 ml of 5: TCA. The supernatant was saved and combined with one additional TCA wash. Radioactivity and DNA were measured in the TCA extract as described in method A. This method was used to show that cyclophosphamide did not affect embryonic DNA synthesis at 24, 48 or 72 hours after treatment 32 (Short, 1971). The methodology and results of that study are included in this thesis for completeness. Alkylation of Embryonic Macromolecules Radioactive ring labeled cyclophosphamide -5-MC (7.6 to 8.2 mc/mM, New England Nuclear, Boston, Mass.) was diluted with non-radioactive cyclophosphamide to give 250 tic/10 mg cyclophosphamide (6.5 uc/mM). The radioactive purity of an aqueous solution for injection was veri- fied by thin-layer chromatography on activated silica-gel plates with a solvent system of n-butanol-acetic acid-H20 (6:2:2). Radioactivity in fractions of the silica gel plates was measured in PCS (Amersham/ Searle, Arlington Heights, 111.) after the addition of 1 ml water using a Packard Liquid Scintillation Spectrometer Model 3380 equipped with an Absolute Activity Analyzer. A teratogenic 20 mg/kg dose of cyclophosphamide, which contained 500 uc/kg of 1l‘C—-cyclophosphamide, was administered i.p. in a volume of 10 ml water/kg to pregnant mice on day 11 of gestation (Gibson and Becker, 1968a). Mice were sacrificed by cervical dislocation at various times after treatment. Radioactivity bound to maternal liver, placenta, and embryonic macromolecules, which were isolated by one of two methods, was used as a measure of cyclophosphamide alkylation. Method A. DNA, RNA, and protein fractions were isolated from the various tissues by a modified phenol extraction (Caldwell and Henderson, 1970). All tissues were homogenized in 5 volumes of buffer A [154 mM sodium chloride, 10 mM sodium phosphate pH 7.4, and 1 mM ethylenediamine- tetraacetic acid, disodium salt (EDTA)] in a loosely fitting, motor driven, teflon-glass Potter-Elvehjem homogenizer. The homogenate was Shaken (280 oscillations/minute) with an equal volume of freshly 33 distilled buffer A saturated phenol for 10 minutes at room.temperature. The emulsion was centrifuged for 20 minutes at 25,000 g and the aqueous upper layer was saved. An equal volume of buffer A was added to the lower phase plus interphase and the mixture was shaken and centrifuged as before. The two aqueous phases were combined and re-extracted with freshly distilled phenol and residual phenol was removed with four ether extractions. Ether was removed under a stream of nitrogen gas. RNA.was precipitated from.the aqueous solution as an insoluble salt after the addition of cetyltrimethylamonium bromide (CTMA) to a final concentration of 6 mg CTMA/ml. The precipitate was extrated three times with 0.1M sodium acetate in 702 ethanol and dissolved in 150 mM sodium chloride 15 mM sodium citrate pH 7 (SSC). RNA was precipitated with CTMA, washed three times with ethanolic sodium acetate, and dis- solved in SSC. The lower phase and interphase were mixed with 1 m1 of 0.1M sodium citrate in 2.5M sodium chloride and 4 m1 of 0.25% sodium dodecylsulfate and shaken for 10 minutes prior to centrifugation. The viscous aqueous layer was extracted with ether and residual ether was removed with a stream of nitrogen gas° DNA and bound RNA were precipitated by CTMA and dissolved in SSC. The remaining interfacial material was mixed with 0.6N perchloric acid (PCA) and heated for 30 minutes at 70°C. The precipitate was washed twice with PCA, 952 ethanol saturated with sodium acetate, and ether. The acid precipitate was dissolved in 0.1N sodium.hydroxide overnight at 40°C. Radioactivity in 2 or 0.1 ml aliquots of the nucleic acid or pro- tein.fraction was measured by liquid scintillation counting (Packard Model 3380) after the addition of 15 ml of PCS (Mersham/Searle). The 34 radioactivity in each fraction was expressed as DPMlg wet tissue weight. The yield of nucleic acids by the phenol extraction was compared to the yield obtained following an acid precipitation. Liver, placental, and embryonic tissue were homogenized in 5 volumes of buffer A and divided into two equal portions. DNA and RNA were extracted from.one portion by the previously described phenol method. The other portion was acidified with an equal volume ff 0.6N PCA. RNA was extracted by heating the precipitate at 40°C for 1 hour in 0.3N sodium hydroxide. The alkaline solution was acidified with 30% trichloroacetic acid (TCA) and the precipitate was washed with 5% TCA. DNA was extracted by heat- ing the precipitate for 20 minutes in 5% TCA on a boiling water bath. DNA was measured by the previously described method of Burton. RNA.was measured by the orcinol procedure (Ceriotti, 1955). The orcinol reagent was freshly prepared by mixing 200 mg orcinol, 10 ml of 4 mM copper chloride (CuClz) in concentrated hydrochloric acid, and a suf- ficient volume of concentrated hydrochloric acid to give 100 m1 of the reagent. A sample of the TCA extract was added to an equal volume of the orcindl reagent, heated for 40 minutes on a boiling water bath, and cooled. The optical density of the solution was measured at 675 nm and RNA was quantified using yeast RNA (Sigma). Method B. Approximately 100 to 300 mg of tissue was homogenized in 5 ml of 0.2N PCA that contained 1 mM non-radioactive cyclophosphamide to minimize reversible binding. The precipitate was washed with 3 successive 5 ml portions of 0.2N PCA. Lipids were extracted from the precipitate with 5 ml washes of 1M sodium acetate in absolute methanol, methanol:chloroform (2:1), absolute ethanol, and two ether washes. 35 Nucleic acids were extracted by heating the precipitate in 5 ml of 0.6N PCA for 30 minutes at 70°C and the precipitate was washed with 0.2N PCA and dissolved in 0.1N sodium hydroxide overnight at 40°C. Radioactivity in the initial acid soluble, hot acid extract, and protein fraction were measured by liquid scintillation counting (Packard Model 3380) after the addition of PCS (Amersham/Searle). A portion of the ether washed precipitate was transferred to a tared counting vial, dried, weighed, and solubilized in 1 m1 of Soluene (Packard, Downers Grove, Ill.) at 40°C for at least 8 hours. Radio- activity was measured after the addition'of a toluene base counting solution (5 g PPO, 200 mg POPOP, and 1 liter toluene). DNA was esti- mated by the previously described method of Burton (1956). Protein was measured by the Lowry method (1951). A 0.5 ml aliquot of the protein solution was mixed with 2.5 m1 of reagent C (0.5 m1 of 22 CuSO °5H 0, 0.5 ml 22 NaK tartrate and 50 ml 21 Na in 0.1N NaOH) 4 2 2°03 and allowed to stand for 10 minutes. Phenol reagent (0.25 ml of a lN solution) was rapidly added and the solution was mixed and allowed to stand for 30 minutes. The optical density of the solution was measured at 500 or 700 nm and the protein was quantified using bovine serum albumin (Sigma) as the standard. A nomograph (California Corporation for Biochemical Research, Los Angeles, Calif.) of the optical density at 260 and 280 was used to measure total nucleic acids and protein in the nucleic acid extract. Acid soluble radioactivity was expressed in terms of wet tissue weight and bound radioactivity was expressed in terms of the amount of acid precipitate, nucleic acid, DNA, or protein present. 36 Postnatal Toxicity Neonatal Swiss-Webster mice were obtained from timed pregnancies started in this laboratory. Cyclophosphamide was administered subcu- taneously (s.c.) 24 to 48 hours after birth at a dose of 80 mg/kg in a volume of 20 ml normal saline/kg. This treatment significantly reduced the growth rate and increased lethality (Short and Gibson, 1971b). Control animals received an equal volume of normal saline. The animals were allowed to develop and some were weaned at 3 to 4 weeks of age. Body and organ weights were measured at various times after treatment. Biosynthesis of Macromolecules by Neonatal Mice Drug induced biochemical lesions were studied in neonatal mice treated, as previously described, with 80 mg/kg s.c. cyclophosphamide 24 to 48 hours after birth. 14C-labelled precursors of DNA, RNA or protein were administered s.c. at a dose of 100 uc/kg in a volume of 10 ml/kg at various intervals after cyclophosphamide and the animals were sacrificed by decapitation at the indicated times. The liver and brain from 1 to 3 mice in a litter were removed and pooled to represent a single experimental unit. The tissue was weighed and homogenized in 5 ml of 0.2N perchloric acid (PCA). The bodies were homogenized in distilled water with a Polytron homogenizer (Brinkmann, Westbury, N.Y.). A portion of the homogenate was made 0.3N in PCA and stored at least 15 minutes on ice. C-Thjmidine Incorporation. Thymidine-Z-MC (50 mc/mM, ICN l4 Tracer Lab) was administered 1 hour prior to sacrifice. After tissue homogenization the precipitate was washed with successive 5 ml washes of 0.2N PCA containing 4 mM non-radioactive thymidine, 0.2N PCA, 952 ethanol saturated with sodium acetate, ethanol:ether (3:1) and ether. 37 The precipitate was hes-ted at 37°C for 1 hour in 4 ml of 0.3N sodium hydroxide, acidified with 3 ml of 30% trichloroacetic acid (TCA), and stored on ice for 10 to 15 minutes. The precipitate was washed twice with 5 m1 of 5% TCA and then heated twice in 5 m1 of 5% TCA for 15 minutes on a boiling water bath. The supernatants, after heating, were saved for the measurement of radioactivity and DNA. Radioactivity was measured in the previously described dioxane-base counting solution using the Packard 3380 Liquid Scintillation Spectrometer. DNA was measured by the diphenylamine method of Burton (1956) . The specific activity was expressed as DPM/mg DNA. 14C-Uridine Incorporation. Uridine-Z-MC (50.8 mc/mM, ICN Tracer Lab) was administered 1 hour prior to sacrifice. After tissue homogeni- zation the precipitate was washed with successive 5 m1 portions of 0.2N PCA, 0.2N PCA containing 4 mM non-radioactive urid‘ine, two 0.2N PCA washes, 952 ethanol saturated with sodium acetate, ethanol:ether (3:1), and ether. The precipitate was heated at 37°C for 1 hour in 4 ml of 0.3N sodium hydroxide, acidified with 3 m1 of 302 TCA and stored on ice for 10 to 15 minutes. The precipitate was washed with 2.5 m1 of 101 TCA and the supernatant was combined with the supernatant after alkaline hydrolysis for the measurement of radioactivity and RNA. Radioactivity was measured in the dioxane-base counting solution using the Packard 3380. RNA was measured by the orcinol procedure (Ceriotti, 1955). The results were expressed as DPM/mg RNA. 14C-Leucine Incorporation. L—Leucine (UL)14C (197 mc/mM ICN Tracer Lab) was administered 30 minutes prior to sacrifice. After tissue homogenization the precipitate was washed with successive 5 m1 washes of 0.2N PCA containing 40 um leucine and 0.2N PCA. The 38 precipitate was washed with successive 5 ml portions of 952 ethanol saturated with sodium acetate, ethanol:ether (3:1) and ether. The pre- cipitate was dissolved in 0.3N sodium hydroxide overnight at 40°C prior to the measurement of protein and radioactivity. An aliquot of the alkaline protein solution was dissolved in Hyamine Hydroxide (Packard) and heated for at least 8 hours at 40°C prior to the addition of the dioxane-base counting solution. Protein was measured by the Lowry method (1951). The specific activity was expressed as DPM/mg protein. Protein Turnover in Neonatal Mice Proteins in neonatal mice were labeled with ll‘C-arginine prior to cyclophosphamide treatment by administering the radioactive precursor to prepant mice prior to parturition. Pregnant Swiss-Webster mice received 100 uc/kg i.p. of L-arginine (guanido-MC) (59 mc/mM. ICN Tracer Labs) at 10 p.m. on day 18 and at 10 a.m. and 10 p.m. on day 19 of gestation. Litters were delivered spontaneously and received either 80 mg/kg cyclophosphamide or normal saline one day after birth. This treatment significantly affected subsequent growth and development (Short and Gibson, 1971b). Neonatal mice were sacrificed at 1, 3, 5, and 10 days after cyclo- phosphamide by decapitation. The liver and brain from 1 to 2 mice were weighed and homogenized in 4 volumes of 154 mM potassium chloride and 10 mM tris(hydroxymethyl)aminomethane (TRIS, Sigma) pH 7.3 (KCl-TRIS) using a loosely fitting, motor driven, teflon-glass Potter-Elvehjem homogenizer. The bodies were homogenized in 4 volumes of KCl-TRIS with a Polytron homogenizer (Brinkmann). The homogenates were separated into a soluble and particulate fraction by centrifugation at 40,000 revolu- tions per minute in a 50 titanium rotor (Beckman) for 1 hour (105,000 g 39 x 1 hour) in a Beckman Model L3-50 ultracentrifuge. The supernatant was saved and the precipitate was resuspended in 5 m1 of KCl-TRIS and centrifuged at 105,000 g for 1 hour. The supernatants were combined and made 0.3N with concentrated perchloric acid (PCA). The acid precipi- tate .was washed with successive 5 m1 portions of 0.2N PCA, 0.2N PCA, 951 ethanol saturated with sodium acetate, ethanol:ether (3:1) and ether. The precipitate was finally dissolved in 5 ml of 0.3N sodium hydroxide. The pellet and walls of the centrifuge tube were carefully washed with distilled water after the second centrifugation and the pellet was dissolved in 5 ml of 0.3N sodium hydroxide. Radioactivity in the alkaline protein solution was measuredby liquid scintillation counting in a Packard 3380 after the addition of PCS-solubilizer (Amersham/ Searle). Protein was measured by the Lowry method (1951). The specific activity was expressed as DPM/mg protein. @alitative Changes in Neonatal Protein Synthesis Electrophoretic Analysis of Protein Synthesis. Litters of day-old Swiss-Webster mice re-eived either 80 mg/kg cyclophosphamide or normal saline subcutaneously in a volume of 20 mllkg. Mice were sacrificed by decapitation and livers and brains were removed and transferred directly to Hank's solution (Altman and Dittmer, 1964) which was maintained at 37°C in a glass stoppered Erlenmeyer flask. Hank's solution contained either L-leucine (UL)14C (1 uc/ml) or L-leucine (4,5)311 (2 uc/ml). for 2 minutes and incubated 2 for 1 hour at 37°C. Tissues labeled with 3H and 1l‘C-leucine were The flasks were gased with 952 02:52 CO removed from the incubation media, mixed, and homogenized in 0.01M TRIS pH 7.3 using a Polytron homogenizer. The homogenate was centri- fuged at 20,000 g for 30 minutes in a Sorvall RCZ-B centrifuge. The 40 supernatant was made 12 in regard to both sodium dodecyl sulfate (SDS) and 2—mercaptoethanol and heated at 60°C for 10 minutes. The supernatant was cooled to room temperature and made 0.3M with iodoacetamide. The mixture was dialyzed overnight at room temperature with two changes of 0.01M TRIS and 0.1% SDS buffer. Dialysis tubing was boiled in 0.01M ethylenediaminetetraacetic acid (tetrasodium.salt) and 0.1M.sodium carbonate for 30 minute periods until the solution was clear, washed in distilled water, and stored at 0°C in 501 ethanol before use. The protein was concentrated by dialyzing the sample for 2 hours against 40% sucrose that contained 0.1% SDS and 0.01M TRIS pH 7.3. Polyacrylamide gel electrophoresis was conducted in a discontinuous system.modified from one described by umbreit et a1. (1972). The modi- fications were reducing the running gel acrylamide concentration to 52 and including 0.1! SDS in both the buffer reservoir and gels. Gel solutions were deaerated under vacuum for 10 seconds, added to tubes (0.6 cm i.d.), and photopolymerized under fluorescent light for 30 minutes. The stacking gel was 0.6 cm and consisted of 2.52 acrylamide. The running gel was 6 cm and contained 52 acrylamide. The upper and lower buffer reservoir was a TRIS-glycine buffer pH 8.3. Aliquots of the protein sample (20 to 50 ul) were electrophoresed at 2 mA/tube for 2 to 3 hours at room.temperatures using an ISCO polyacrylamide electro- phoresis apparatus. The gels, after electrophoresis, were preserved in 102 TCA. Gels were frozen in a bath of dry ice and acetone and sliced with a hand-held razor blade cutter. The fractions were solubilized in Soluene (Packard) according to the method developed by Terman (1970). Radioactivity from 3R and 14Cwas measured simultaneously in a Packard 3380 liquid 41 scintillation spectrometer after the addition of the toluene base count- ing solution previously described. Develgpggntal Changes in Glucose-6-Phosphatase Activity. Litters of day-old Swiss-webster mice received either 80 mg/kg cyclophosphamide or normal saline s.c. in a volume of 20 ml/kg. Glucosen6-phosphatase activity was assayed by the method described by Greengard at al. (1967) which was used to show developmental changes in enzyme activity in rat liver. Neonatal and adult mice were sacrificed by decapitation and cervical dislocation, respectively. Livers were~removed,‘weighed, and homogenized in 10 volumes of 0.25M sucrose. All incubations were con- ducted for 10 minutes at 37°C in room air using a Dubnoff metabolic shaking apparatus. The incubation media consisted of 0.25 ml of homogenate in 20 mM glucose-6-phosphate and 50 mM TRISdmaleate buffer pH 6.5 for a final volume of 1 ml. Reactions were stopped by the addi- tion of 1 ml of 1.2N perchloric acid (PCA). Inorganic phosphate was measured in the supernatant as described below and protein was measured (Lowry et aZ., 1951) in the acid precipitate which was solubilized in sodium.hydroxide. All incubations were performed in duplicate and blank values were obtained by adding the homogenate to media that was acidified at the end of a 10 minute incubation. Inorganic phosphate was measured in 1 m1 of the supernatant by the addition of 1.5 ml of sodium molybdate solution. The yellow color was extracted into 5 ml of butyl acetate and the absorbsnce was measured at 390 nm, The values were quantified using a standard curve prepared from.potassium.phosphate. The sodium molybdate solution was prepared by adding 14.5 g of sodium molybdate to a solution of 210 ml of 702 PCA and 90 g of sodium hydroxide which was titrated to a cresol red: 42 thymol blue (25 mg:150 mg in 100 ml water) light blue and adding suf- ficient water to give a final volume of 1 liter. Recommendations Various methods were used in these studies to measure the incorpora- tion into or binding of radioactive molecules to macromolecules. The review by Munro and Fleck (1966) provided a valuable source of informa- tion for devising and improving the methodology used to isolate DNA, RNA, and protein fractions. The procedure used to measure the biosyn- thesis of macromolecules by neonatal mice is a modification of a pro- cedure suggested by Munro and Fleck (1966) and is recommended as a suitable method for further studies. Statistics Statistical evaluation of the data was performed by Student's t-test (Steel and Torrie, 1960). Equations for lines were calculated by linear regression analysis (Mendenhall, 1971). The level of sig- nificance was chosen as P<0.05. RESULTS Biosynthesis of DNA.by Embryonic Mice DNA synthesis and cell multiplication are important aspects of normal development. DNA synthesis, as measured by 1l‘C—thymidine incorporation into DNA, was greater in embryos and placentae than in maternal liver (Table 1). The more rapid rate of DNA synthesis in developing tissue indicated that an increase in cell number in con- trast to an increase in cell size was an important factor in growth. A dose of cyclophosphamide, which disrupted normal embryonic develop- ment, also reduced DNA synthesis in mouse embryos. Cyclophosphamide- treated embryos had a significantly impaired ability to incorporate 1l'C-thymidine into DNA 12 hours after treatment (Figure l). A reduced DNA concentration (mg DNA/g wet weight of embryo) was subsequently observed at 12, 24, 48, and 72 hours after drug treatment (Table 1). DNA synthesis was reduced in the placenta at 24, 48, and 72 hours after cyclophosphamide (Table l) but the concentration of placental DNAuwas reduced only at 12, 24, and 48 hours after treatment (Table 2). The specific activity of DNA, after a pulse of 14C-thymidine, was sig- nificantly reduced in maternal livers at 6, 12, and 72 hours after cyclophosphamide (Table l) and the DNA concentration*was reduced at 24 and 48 hours after treatment (Table 2). The initial acid soluble radioactivity of tissues from treated mice was equal to or greater than the same fraction from control tissue. Cyclophosphamide, therefore, did not affect the placental transport of radioactivity and thereby 43 44 Table l. ll‘C-thymidine incorporation after cyclophosphamide treatment a Hours after Method of DPM/mLDNA x 10"3 cyclophos- measuring in- Liver Placenta Embryo phamideb corpora tionc Cont . Treat Cont . Treat Cont . Treat 3 A 5.2d 4.5 29 25 27 25 .19-7 $9.3 :t3-9 3:2.0 ;:5.1 ;12.3 6 A 5.1 3.5° 36 29 32 23 ‘19.4 19.6 ;17.3 ;:5.6 19u3 .i1.6 12 A 5.1 3.3° 23 19 26 17° ‘i0.6 119.4 12.2 39.3 11.9 5:1.8 24 A 10 5.6 44 30° 28 27 :9-3 (19.9 ;:3.0 ;:2.4 14.8 (19.1 24 n 4.9 6.0 43 31° 19 26 39.4 ;tp.8 3:2.6 ;:1.8 1919 14.3 48 B 4.3 2.6 27 15° 18 13 :t .3 :t .5 ;:2.0 ;:3.6 12.1 l:4.o 72 n 3.9 2.0e 22 14° 19 17 1;.3 19.4 ;t2.o 11.2 19.2 $9.6 llOne-hour pulse (100 uc/kg) JAG-thymidine. l)Cyclophosphamide (20 mg/kg) was administered to pregnant females on day 11 of gestation. cSee Methods. dMean 1; SE for 3-6 observations. eSignificantly different from control (P<0.05). -11! 33'5" W78 are WA .._.- f. 45 Figure l. l4C-thymidine incorporation into embryonic DNA as a function of time after cyclophosphamide. Pregnant mice were treated with cyclophosphamide (20 mg/kg i.p.) on day 11 of gestation and a 1 hour pulse of 14C-thymidine (100 uc/kg i.p.) was administered at various intervals after drug treatment. The values are the mean;1 SE of the specific activity of embryonic DNA for 3-6 determinations. An asterisk indicates a significant difference from control (P<0.05). 46 0-0 Control H Cyclophosphamide 36 .. . V 32 Y DPM/Ing on 7: 10° 2 s 3 6 12 24 48 72 HOURS AFTER CYCIOPHOSPIIAMIDE Figure 1 Table 2. Hours after Method of 47 DNA content after cyclophosphamide treatment ngNA3per g wet weight tissue cyclophos- isolating Liver Placenta Embryo phamide“ DNAID Cont. Treat Cont. Treat Cont. Treat 3 A 2.4° 2.5 3.1 3.1 4.9 4.5 330.2 330.3 330.2 330.3 30.9 30.5 6 A 3.2 3.0 3.2 2.9 3.9 4.3 330.2 330.2 330.2 330.1 330.4 30.2 12 A 3.4 3.3 3.6 3.1‘1 4.3 3.7d 30.2 330.1 330.1 330.2 30.2 330.1 24 A 3.4 3.0 3.7 3.3d 5.1 3.3d 30.2 330.1 330.1 330.1 330.3 330.2 24 n 2.9 2.3d 4.0 3.7 5.0 3.7‘1 330 330.2 330.4 330.3 330.0 330.2 40 3 2.6 1.6d 4.1 3.4d 6.3 4.3d 30.2 330.1 330.1 330.2 330.2 30.2 72 3 2.6 2.9 2.9 3.0 6.4 5.5‘1 330.1 30.2 30.1 30.2 +0.2 330.3 aCyclophosphamide (20 mg/kg) was administered to pregnant females (or: day 11 of gestation. b See Methods. cMean 3; SE for 3-6 observations. d Significantly different from control (P<0.05). 48 limit precursor availability for incorporation. These observations indicated that cyclophosphamide significantly reduced DNA synthesis, as measured by ll‘C—-thy1l.idine incorporation, in both' embryonic and adult tis sues . Alkylation of Embryonic Macromolecules Drug binding to macromolecules was studied to determine if a terato- genic dose of cyclophosphamide produced'alkylation of embryonic macro- molecules. The temporal aspects of drug binding were measured to evaluate changes in this parameter with time. Alkylation in both embryonic and adult tissues was compared 'because‘the adult had a proven ability to form alkylating metabolites from the parent compound. Drug binding to total macromolecules rather than subcellular fractions was measrued to avoid the requirement of obtaining high yields of‘pure subcellular frac- tions from a heterogeneous mixture of embryonic tissues. Radioactivity in DNA, RNA, and protein fractions from maternal liver, placenta, and embryo, after lac—cyclophosphamide, are shown in Table 3. Negligible amounts of radioactivity were observed in the DNA and RNA fractions relative'to radioactivity in the protein fraction. As a result of these low levels of radioactivity the yield of macro- molecules was measured in order to determine how accurately this obser- vation reflected cyclophosphamide binding to DNA and RNA. The phenol method used in the present study was reported to yield 75 to 802 of the DNA. 70 to 802 of total RNA, and 90 to 952 of the‘protein in Ehrlich “Cites tumor cells (Caldwell and Henderson, 1969). The yield of m1(2-‘l-eic acids from maternal liver, placenta, and embryos by the phenol Pr°¢edure relative to the yield of DNA and RNA from an acid precipitate were 65 to 802 for RNA but only 25 to 302 for DNA. The phenol procedure as“ 1n the present study was concluded to be unsatisfactory for measuring 49 Table 3. 1l’C-cyclophosphamide radioactivity in DNA, RNA, and protein fractions from maternal liver, placenta, and embryo DPM/Eissuea ’ b 3 3__ Tissue DNA RNA Protein x 10 Liver 10c 36 242 :6 :13 :30 Placenta 48 408 52 1:25 3:123 :16 Embryo 61 184 42 :35 _-_1~_125 :17 a'500 uc/kg 1l’C—cyclophosphamide was injected i.p. on day 11 of gestation. bDNA, and protein fractions were isolated by a phenol method (see Methods). cMean 1; SE for determinations made at 0.5, 1.5, 3, and 5 hours af ter 14C-cyclophosphamide . 50 drug binding to nucleic acids in an embryonic system as a result of both the small amount of tissue available and loss of macromolecules during the isolation procedure. Temporal aspects of the binding of radioactivity to macromolecules from the maternal liver, placenta, and embryo, after lac-cyclophosphamide, were studied by an acid precipitation procedure. This method provided readily detectable levels of radioactivity. Acid soluble radioactivity was present in all of the tissues studied for at least 10 hours after 14C-cyclophosphamide (Figure 2). The liver had the highest level of acid soluble radioactivity at 0.5, 1.5, 3.0, 5.0 and 10 hours after treatment. The data indicated that radioactivity after a teratogenic dose of 1l‘C—cyclophosphamide crossed the placenta and remained in in acid soluble form in embryos for at least 10 hours after drug treatment. This study, in addition, demonstrated that acid soluble radioactivity was eliminated from all of the tissues studied and the elimination occurred in a biphasic fashion. The acid washed-lipid extracted acid precipitate of maternal liver, placenta, and embryo contained radioactivity after JAG-cyclophosphamide treatment (Figure 3). The specific activity of the acid precipitate decreased in the order liver, placenta, and embryo at all of the times studied. Radioactivity in the acid precipitate reached maximal values by 1.5 hours after treatment and declined with time. The decline in aPeecific activity appeared to plateau towards the end of the observa- tion period. Radioactive molecules were solubilized from the acid-washed-lipid- 61": reacted precipitate of tissue by hot acid (Table 4). The acid extract contained both DNA, as measured by the diphenylamine reagent, and ”0t elm and nucleic acids, as measured by the optical density at 280 51 Figure 2. Acid soluble radioactivity in the maternal liver, placenta, and embryo as a function of time after 500 uc/kg (20 mg/ kg) 1l‘C-cyclophosphamide. The values plotted are the mean t_SE for at least 4 observations. DPM/g ussu: x 103 2000 1000 'IC 52 i3. ......... LIVER '. _ PLACENTA ___ EMBRYO 015 1.5 310 510 10.0 nouns an“ ‘4c-cvcmPHosmamo: Figure 2 4“”- 53 Figure 3. Radioactivity in the acid washed lipid extracted acid precipitate from maternal liver, placenta, and embryo as a function of time after 500 uc/kg (20 mg/kg) 4C-cyclophosphamide. The values plotted are the mean :_SE for 3 to 5 observations. 54 Sigubbs at: A:50... 92 ca 8 ngu I I I <22mQ= .......... find . oo— -08.. .. coon alumnus am Bull was Figure 3 1w!” Ell; .. .rrl. E45" ‘1 55 and 260 nm (Tables 4 and 5). The expression of radioactivity in terms of the amount of extracted DNA indicated that the specific activity of this fraction decreased in the order liver, placenta, and embryo at all of the times studied (Table 4). Radioactivity of this fraction expressed in terms of milligrams total nucleic acid was lower than the specific activity of the same fraction expressed in terms of milligrams DNA at all of the times studied in liver and placenta but not embryo. The decline in specific activity of the nucleic acid extract (DPM/mg DNA) between 0.5 and 5.0 hours after 1l'C-cyclophosphamide expressed as a percent of the initial values was 28, 38, and 832 for the liver, pla- centa, and embryo, respectively. 1l‘C-cyclophosphamide radioactivity was associated with the protein fraction (Table 5). The specific activity of this fraction decreased in the order liver, placenta, and embryo. Radioactivity in both the ' protein and nucleic acid fraction, expressed in terms of milligrams protein, indicated that the specific activity was higher in the former. The specific activity of the protein fraction declined to 75, 61, and 662 of the initial values observed in liver, placenta, and embryo, respectively, between 0.5 and 5.0 hours after treatment. The data in Figure 4 depict the temporal changes in the ratio of acid precipitable to acid soluble radioactivity after l('Cucyclophosphamide in maternal liver, placenta, and embryo. This value is used as a measure of the ratio of bound to free radioactivity. This ratio decreased in the order liver, placenta, and embryo. This observation, in conjunc- tion with previously discussed macromolecule specific activity data, indicated that the alkylation of tissue macromolecules also decreased in the order liver, placenta, and embryo. These differences may be 56 Table 4. lac-cyclophosphamide radioactivity in the nucleic acid extract from maternal liver, placenta, and embryo8 Hours after DPM/mg DNAc x 10"2 DPM/mg NAd x 10.2 14C-cyclob Liver Placenta Embryo Liver Placenta Embryo 0.5 126.3° 27.3 3.0 9.3 3.9 1.9 309.1. 33.3 30.5 330.3 30.6 330.5 1.5 77.1 19.0 1.1 6.6 2.4 4.3 3325.0 3310.3 330.1 330.6 330.3 334.4 3.0 66.3 19.5 2.6 4.0 2.2 3.3 3340.1 30.1 330.6 30.3 330.1 332.4 5.0 34.9 10.4 2.5 3.3 2.0 2.7 :26-3 +2.3 330.6 30.4 +0.3 331.2 aThe nucleic acid extract was obtained by heating a lipid extracted acid precipitate in 0.6N perchloric acid for 30 minutes at 70°C. h500 uc/kg of 1l‘C--cyclophosphamide (20 mg/kg) was administered 1- p. on day 11 of gestation. “DNA was measured by the diphenylamine method of Burton (1956) . dTotal nucleic acids were determined from the optical density at: 260 and 280 nm. eMean : SE for at least 3 determinations. 57 Table 5. JAG-cyclophosphamide radioactivity in the protein and nucleic acid fraction from maternal liver, placenta, and embryo V?— Protein Fractionb 2 Nucleic Acid Fraction‘l Hours after DPMlgg proteinc x 10' ijmroteine x 10"2 14C—cyclo‘ Liver Placenta Embryo Liver Placenta Embryo 0.5 14.9f 4.9 2.9 0.30 0.45 0.14 10.8 330.5 330.6 10.07 10.03 310.02 1.5 17.4 4.1 3.3 0.57 0.24 0.09 ilez .1004 i006 1:0313 £0004 $0.04 3.0 13.9 4.5 2.4 0.37 0.21 0.13 11.2 30.8 310.4 10.06 10.01 10.01 5.0 11.2 3.0 1.9 0.34 0.18 0.13 310.7 $0.3 +0.2 +0.04 10.02 10.01 ‘500 uc/kg of lac-cyclophosphamide (20 mg/kg) was administered i-p. on day 11 of gestation. I’An acid precipitate washed with lipid solvents'and heated with O- 6N perchloric acid at 70°C for 30 minutes. f:‘Protein was measured by the Lowry method (1951). dThe hot 0.6N perchloric acid extract in b. eProtein was determined from the optical density at 260 and 280 nm. fMean -_+-_ SE for at least 3 determinations. “LT—El . a" '"P v ‘ I-.“ 1‘1"” w-mpapc I". . 58 Figure 4. The ratio of acid precipitable to acid soluble radio- activity in the maternal liver, placenta, and embryo as a function of time after 500 uc/kg (20 mg/kg) 14C-cyclophosphamide. The values plotted are the mean _-l_- SE for at least 3 observations. DPM Im ACID PRECIPITATE om [9 x 103 ACID sown: 59‘ 13- 0.5 1.5 3.0 5.0 HOURS AFTER '4c-CYCLOPHosn-IAMIDE Figure 4 10.0 60 attributed to reduced levels of either alkylating metabolites or target sites for alkylation in the placenta and embryo relative to the liver. Postnatal Toxicity Neonatal mice treated subcutaneously with 80 mg/kg cyclophosphamide 24 to 48 hours after birth matured into abnormal adults. Treated mice, at 7 weeks of age, were reduced in size and had short snouts, ears, and tails (Figures 5 and 6). Organ weights and organ to body weight ratios were measured at various times after drug treatment for the liver and brain (Table 6). There was no significant change in these values in drug treated mice for at least 1 week. Liver and brain weights were Significantly reduced 35 days after treatment when the body weights were only 602 of control. The organ to body weight ratio, however, was changed only for the brain where an increase was observed. Cyclophos- phamide treatment, therefore, interfered with body, brain, and liver growth. The liver weight to body weight ratio, however, indicated that drug treatment did not produce a selective reduction in liver growth since body weights were reduced to the same extent as liver weights. Brain growth, relative to body growth, was not as severely affected in cyclophosphamide treated neonates since the brain weight to body weight ratio was increased over control values. Biosznthesis of Macromolecules .52 Neonatal Mice Neonatal mice actively synthesized DNA, RNA, and protein as unsecured by ll‘C-precursor incorporation into the respective macromole- cules. Neonatal mice, therefore, were actively involved in cell repli- Cation and RNA and- protein synthesis during the period of drug exposure. Cyclophosphamide. induced biochemical lesions were present in neonatal 61 Figure 5. Adult mice that received either 80 mg/kg cyclophosphamide 24 to 48 hours after birth or normal saline. The mouse on the left was a control and the 2 mice on the right were drug treated. 62 E _ _2 .5 a _ _. s _o_ _s 2: v ___—___n____::__~_:_:___ _:::: aces: Figure 5 63 Figure 6. Adult mice that received either 80 mg/kg cyclophosphamide 24 to 48 hours after birth or normal saline. The mouse on the bottom was a control and the mouse on the top was drug treated. 64 1.096;} ___—____________—___;_____==___________==__=__=:__=_______:___:___ NEE:— 23:: H 91.52 Figure 6 65 Table 6. Organ weights and organ to body weight ratios of liver and brain from.mice treated with cyclophosphamide at one day of age Liver Brain mg tissueli g tissue/l mg tissue/I g tissue7' Days after mouse 100 EN mouse 100 g DH treatmenta Cont Treat Cont Treat Cont Treat Cont Treat 3 105b 93 3.6 3.3 152 135 5.2 5.4 :6 3:1 330.2 10.2 :5 338 110.2 10.7 5 151 123 3. 7 3.8 189 177 4.6 5.6 :5 :23 10.2 10.4 :6 115 _-_i-O.2 10.3 7 130 107 2. 8 3.8 217 179 4. 7 6. 8 ill :26 :0. 4 :0 . 1 :16 :26 _-_|-_0. 2 :1. 0 35° 1440 720d 5.6 5.4 440 370d 1.7 2.3‘1 jdl. :60 +0.2 330 :00 330 3:0 3:0 ‘Cyclophosphamide (80 mg/kg s.c.) or normal saline was injected 24 to 48 hours after birth. b cTreated body weights were 602 of control at this time. (1 Mean 3; SE for at least 3 determinations. Significantly different from control (P<0.05). 66 mice treated subcutaneously with 80 mg/kg of the drug 24 to 48 hours lac-thymidine incorporation, after birth. NA synthesis, as measured by was significantly inhibited in neonatal liver (Figure 7), brain (Figure 8), and carcass (Figure 9) 24 hours after cyclophosphamide. DNA synthesis at this time was 8, 26, and 252 of control, respectively. DNA synthesis remained reduced in the livers and brains from drug treated mice for at JAG-thymidine incorporation into DNA of carcasses from least 5 days. treated mice returned to control values 5 days after treatment. The DNA concentration (mg DNA/g wet weight) and DNA content (mg DNA/organ) were significantly reduced in liver between 1 and 5 days after treat- ment (Table 7). DNA concentration was reduced in the brain and carcass at l and 3 and at 2 and 3 days after treatment, respectively (Tables 8 and 9). Hydroxyurea, an inhibitor of DNA synthesis, was used as an internal control to distinguish between real incorporation and nonspecific binding of JAG-thymidine to DNA (Morley and Kingdom, 1972). Hydroxy- urea was administered subcutaneously at a dose of 500 mg/kg to neonatal mice one hour before the 1l‘C-thymidine pulse. This treatment reduced the specific activity of DNA to less than 12 of the value found in controls and abolished differences between control and treated mice 24 hours after cyclophosphamide (Table 10). This observation indicated that the radioactivity associated with DNA isolated from neonatal mice was a result of incorporation rather than nonspecific binding. Day-old mice treated with 500 mg/kg hydroxyurea appeared normal and grew normally for at least 4 weeks despite the dramatic initial inhibition of DNA synthesis. The incorporation of 1l‘C—uridine into total RNA was reduced in neonatal liver between 1 and 5 days after cyclOphosphamide (Figure 7). 67 Figure 7. DNA, RNA and protein synthesis in the liver of cyclo- phosphamide treated animls expressed as a percent of control values. Mice received either 80 mg/kg cyclophosphamide or normal saline one day after birth and 100 ‘uc/kg 4C-labelled thymidine, uridine, or leucine at various times afterwards. Precursor incorporation into DNA, RNA and protein was measured at various times after treatment. The specific activity of macromolecules from treated animals was expressed as a percent of the values simultaneously determined in controls. The average control specific activities were: 39,000 DPM/mg DNA, 4000 DPM/ms RNA, and 1000 DPM/mg protein. The values represent the mean 1 SE for 3 to 6 determinations. Open symbols indicate values that are significantly different from control (P<0.05). 68 1-4—1 3.5350333". cut? 953 O \ \ \\ / \ . . 4 \ ‘ \ ‘ A a. _- - - a . e 9325.6 (.8 0... 2352.5 (2.. III 4 m.mw:.—.2>m 23.23: ‘I‘ rev reap i. GNP All/\llOV OHIOBdS 'IOHLNOO :10 11030836! Figure 7 ran—mi“: ..—.x,£ 31;... 3_ 69 Figure 8. DNA, RNA and protein synthesis in the brain of cycle- phosphamide treated animals expressed as a percent of control values. See legend to Figure 7 for method. The average control specific activities were: 3000 DPM/mg DNA, 1000 DPM/mg RNA, and 500 DPM/Ig protein. 70 wo.2U mwhm< m>0 mwhm< ”>40 m w m N p c r\ b _ e e w , 3 ll. 5 I .p 2325.6 <20 eve \ \ \ / / £8255 <22 mum .q\ / 93255 2552.. {L1 \ \M / I 3 x m \ \ , , 8 x . \ A. \ \ \ \ \ I I I — 8 q I I t I H \ \ I r k \ 1.11 I I .. . 1 ,2 . . H _ T 8. \ \ — H .. .. - - H L . eu- AllAllOV OlleadS WOHlNOO :10 11030836 Figure 9 73 Table 7. Concentration and content of DNA and RNA in neonatal liver after treatment with cyclophosphamide at one day of age Concentration Content mean / wet wei ht 1 SE (mean mglggggn 1 SE) Time after DNA RNA DNA RNA 3 treatmenta Cont Treat Cont Treat Cont Treat Cont Treat 3 hr 10.5 9.4 ---- -- 0.80 0.68 -- --- i0. 4 1:00 7 ---‘ n"' :00 02 :0 e 05 “- ---- 6 hr 9.6 8.2 14.4 13.5 0.76 0.67 1.2 1.4 330.2 330.3 330.2 330.4 330.03 330.031 30 330.1 1 day 3.9 5. b 12.1 12.0 0.71 0.43b 1.1 1.0 30.3 30.4 31.0 30.3 30.02 30.03 30.1 2 days 9.6 5.4b 12.6 13.2 0.91 0.40b 1.2 1.2 330.4 330.2 330.5 330.7 330.05 330.03 330.1 330.1 3 days 9.6 4.3b --- -~- 0.61 0.31b ---- --- 330.6 330.3 ---- ---- 330.02 330.04 ~-- --—- 4 days -- --- 13.6 13.3 ---- --- 1.4 1.4 ---- ---- 30. 2 30.1 ---- --- 30.1 30.1 5 days 6.4 4.6 15.4 15.3 0.32 0.43b 1.7 1.3 +0.4 +0.2 10.4 10.1 +0.10 10.05 10.2 +0.3 aMice received cyclophosphamide 80 mg/kg or normal saline. bSignificantly different from control (P<0.05). 74 Table 8. Concentration and content of DNA and RNA in neonatal brain after treatment with cyclophosphamide at one day of age Concentration Content (mean gglg wet weight t SE) (mean mglgggan t SE) Time after DNA RNA DNA RNA treatment“ Cont Treat Cont Treat Cont Treat Cont Treat 3 hr 4.0 4.2 --- --- --- --- 10 . 2 10 . l ---- ---- ---- --— 6 hr 4.8 4.6 4.6 4.5 0.41 0.37 0.47 0.47 10.1 10.1 10 10.2 10.02 10 1.03 1.03 1 day 3.7 3.5b 4.0 3.3 0.39 0.33 0.40 0.37 10.1 10.1 10.2 10.1 10.01 10.01 10 1.02 2 days 3.3 3.4 3.9 3.8 0.40 0.38 0.53 0.45 10.1 10.3 10.1 10.2 10.03 10.02 10.03 10.04 3 days 2.7 2.3b ---- —--— 0.23 0.21 -- --- 10.1 10. 1 --- -- 10. 02 10. 01 --- -- ---- ---- :00 1 _tOe 1 --- --- i0. 03 :0 5 days 2.1 1.9 3.9 3.7 0.42 0.50 0.70 0.67 +0.1 +0.1 +0.1 +0.1 +0.02 +0.10 +0.06 310.06 aMice received cyclophosphamide 80 mg/kg or normal saline. b Significantly different from control (P<0.05). 75 Table 9. Concentration of DNA and RNA in neonatal carcass after treatment with cyclophosphamide at one day of age Concentration (mean qglg wet weight 1 SE) Time after DNA RNA Y treatment“ Cont Treat Cont Treat 3 hr 4.7 4.9 --- --- 310.1 10.1 ---- --- 6 hr 2.9 4.8 5.8 6.1 310.9 10.1 10.2 310.2 1 day 4.3 4.0 4.6 4.6 10.2 10.1 10.1 10.1 2 days 3.4 3.61’ 5.7 5.5 10.3 10.2 10.2 10.2 3 days 3.3 2.7b ---- ---- 10.3 310.1 -- --- 4 days --- --- 6.3 5.3b --- --- :0-2 :0 5 days 4.1 2.3 6.4 5.51) +0.2 +0.1 +0.1 10.2 aMice received cyclophosphamide 80 mg/kg or normal saline. b Significantly different from control (P<0.05). 76 Table 10. DNA specific activity in neonatal mice after hydroxyurea treatmenta 'DNA Specific Activity Relativeb Dru/53 DNA DPMng DNA __ Tissue Control Control Treated” Liver 0.272cl 106 156 10.03 3110 102 Placental 0. 832 24 22 310.27 310 10 Embryo 0.542 78 58 10 . 05 17 112 aDag-old mice received 500 mg/kg hydroxyurea s.c. 1 hour before a 1 hour 1 C-thymidine pulse. bRelative to mice that did not receive hydroxyurea. cCyclophosphamide 80 mg/kg s.c. 24 hours before hydroxyurea. dMean 1 SE of 3 determinations. 77 RNA synthesis ranged from 36 to 642 of control between 1 and 5 days after treatment in the liver of treated mice. Liver RNA.concsntration and content were not affected by drug treatment (Table 7). RNA synthesis was reduced in brain to 78 and 642 of control on days 1 and 4 after treatment (Figure 8). RNA concentration was increased in brain only at 4 days after treatment (Table 9). RNA synthesis was not changed in carcass but the RNA concentration was reduced 4 and 5 days after treatment . The incorporation of IAC-leucine into total protein did not reveal a consistent pattern which would indicate that drug treatment altered protein synthesis. Protein synthesis was increased in the liver 3 days after treatment (Figure 7) and reduced in the brain at 3 hours and 5 days after treatment (Figure 8). Protein synthesis, in addition, was not significantly changed in the liver or brain at 7, 9, 12, or 21 days after treatment (Table 11). Protein Turnover in Neonatal Mice The differentiated state is ultimately dependent on the acquisition of adult tissue specific proteins. Agents which interfere with this process may disrupt normal development. The protein content of a tissue is dependent both on the synthesis of new proteins and the degradation of existing proteins. Since protein synthesis, as measured by 16C- 1eucine incorporation into protein, was not affected in cyclophosphamide treated neonatal mice, the possibility remained that drug treatment altered the rate of protein degradation. The specific activity of neonatal proteins labeled in utero with 1l‘C—guanido-L-arginine was measured as a function of time after cyclo- phosphamide in mice treated with 80 mg/kg s.c. of the drug one day after 78 Table 11. Protein synthesis in mdce treated with cyclophosphamide a Relative specific activity of protein8 Days after treatmentb Liver Brain 7 120c 86 ;120 :9 9 82 74 ;:7 {1;3 12 106 102 i7 i3 21 132 118 i?5 :11 836) minute pulse of 100 uc/kg ll‘C-ileucine. bCyclophosphamide 80 mg/kg 24 to 48 hours after birth. cMean;:.SE (2 to 3 observations) of protein specific activity expressed as a percent of control. 79 birth. Protein turnover, as measured by the decline in protein specific activity, refers to the combined synthesis and degradation of tissue constituents (Arias at aZ., 1969). Cyclophosphamide significantly affected protein turnover in neonatal mice (Tables 12 and 13). The half-life of total soluble protein, which was the time required for the protein specific activity one day after treatment to decline by 502, was significantly decreased in the liver, brain, and carcass of drug treated neonates (Table 13). A reduced half-life of particulate protein was measured in the liver and brain after cyclophosphamide treatment. Isotope re-utilization which occurs as a result of protein degra- dation is a problem in estimating protein half-life. This problemAwsa manimized in protein turnover studies by the use of 14C—guanido-L- arginine which was metabolized by arginase to give lac-urea‘with the result that the label was not re-utilized for protein synthesis (Arias at al., 1969). The half-life of total organ protein radioactivity, therefore, is an estimation of protein degradation assuming minimum re-utilization of the isotope. Cyclophosphamide significantly affected protein degradation in neonatal mice (Tables 14 and 15). The half-life of soluble and particulate protein was decreased in the liver but not the brain after drug treatment. Protein half-lifes as calculated from the decay in protein specific activity and total organ protein radioactivity are summarized in Table 16. The temporal decrease in protein specific activity is attributed both to the synthesis of new proteins and the degradation of existing proteins. The decrease in total organ protein radioactivity, on the other hand, is due to protein degradation. The difference between these two calculations of half-life are dramatically illustrated in the liver. 80 Table 12. Equations for the calculated least squares regression line of log protein specific activity as a function of days after cyclophosphamide d e Equation r Protein Fractionc Control _ Treated Control Treated Liver '1 Soluble y-3.lO-0.l3x y-3.4l-O.l9x 0.98 0.89 Particulate y-3.l7-0.12x y'3.50-0.19x 0.98 0.86 Brain Soluble y-3.28-0.09x y-3.53-0.l3x 0.98 0.82 Particulate y-3.28-0.08x y-3.S9-0.1Zx 0.98 0.77 i Carcass iJ Soluble y-3.29-0.l3x y-3.58-O.l7x 0.99 0.86 Particulate y-3.53-0.l4x y-3.70-0.l7x 0.98 0.78 ‘114 C-guanido~L-arginine was administered i.p. t0 the mother prior to birth. bCyclophoaphamide (80 mglkg s.c.) or normal saline was administered to neonatal mice 1 day after birth. cSoluble protein was obtained from the 105,000 g supernatant and particulate protein was obtained from the pellet. The specific activity (DEN/mg protein) of each fraction was measured as a function of days after cyclophosphamide treatment. dThe equations were calculated by least squares regression analysis for 3 observations made at l, 3, 5 and 10 days for control liver and brain (N-12) or 1, 3 and 5 days for carcasses and treated liver and brain (N-9). eRegression coefficient. 81 Table 13. Half-life of protein from neonatal mice after cyclophos- phamide as estimated by the change in protein specific activity‘:b Half-life (days) 95: Confidence Interval of Sloped Protein Fractionc Control Treated Control Treated Liver Soluble 2.4 1.5 -0.125 to -0.135 -0.162 to -0.221 Brain Particulate 4.1 2.5 -0.074 to -0.080 -0.095 to -0.156 Carcass Soluble 2.4 1.7 -0.124 to -0.l36 -0.143 to -0.203 Particulate 2.2 1.9 -0.127 to -0.144 -0.127 to -0.206 aProtein half-life was estimated from.the least squares regres- sion line of log protein specific activity vs. days after treatment (Table 12). bDay-old mice received either 80 mg/kg s.c. cyclophosphamide or normal saline. cSoluble protein was obtained from.the 105,000 g supernatant and particulate protein was obtained from the pellet. dEquations for the lines are given in Table 12. 82 Table 14. Equations for the calculated least squares regression line of log protein radioactivity per organ vs. days after treatment“b Equationd re Protein Fractionc Control Treated Control Treated Liver Soluble y-3.70-0.06x y'4.06-0.16x 0.80 0.86 Particulate y-4.05-0.051 y-4.34-0.15: 0.80 0.84 Brain Soluble y-3.79-0.07x y-4.03-0.10x 0.76 0.73 Particulate y-4.43-0.13x y-4.33-0.09x 0.62 0.68 a14-guanido-L-argininewas administered 1.P. to the mother prior to birth. bCyclophosphamide (80 mg/kg s.c.) or normal saline was adminis- tered to neonatal mice 1 day after birth. cSoluble protein was obtained from the 105,000 g supernatant and particulate protein was obtained from the pellet. Incorporated radio- activity per organ was measured as a function of days after cyclophos- phamide treatment. dThe equations were calculated by least squares regression analysis for 3 observations made at l, 3 and 5 days (Nb9) after treatment . eRegression coefficient. 83 Table 15. Half-life of protein from neonatal mice after cyclophosphamide as estimated by the change in organ protein radioactivity“b Half-life (days) 95% Confidence Interval of Slope Protein Fractionc Control Treated Control Treated Liver Particulate 6.4 2.0 -0.0377 to -0.0589 -0.1512 to -0.2090 Brain Soluble 4.0 4.0 0.0541 to 0.0905 -0.0704 to -0.1256 Particulate 2.2 2.2 0.0834 to 0.1853 -0.0587 to -0.ll46 aProtein half-life was estimated from.the least squares regression line of log organ radioactivity vs. days after treatment (Table 14). b normal saline. Day-old mice received either 80 mg/kg s.c. cyclophosphamide or cSoluble protein was obtained from the 105,000 g supernatant and particulate protein'was obtained from.the pellet. quuations for the lines are given in Table 14. 84 Table 16. Half-life (days) of proteins from neonatal mice after cyclophosphamide Tissue a Protein Specific Activityb Organ Protein Radioactivityc Fraction Control Treated Control Treated Liver d d Soluble 2.4 1.5d 6.0 1.8d Particulate 2.6 1.6 6.4 2.0 Brain d Soluble 3.7 2.3d 4.0 4.0 Particulate 4.1 2.5 2.2 2.2 Carcass d Particulate 2.2 1.9 --- -- aSoluble protein was obtained from the 105,000 g supernatant and particulate protein was obtained from the pellet. b Table 13. cTable 15. d952 confidence limits for the slope of the least squares regression line for control and treated observations do not overlap. 85 The half-life of liver soluble protein in control mice was increased from.2.4 to 6.0 days when protein synthesis was excluded from the calculations. The yield of soluble and particulate protein from the liver and brains of neonatal mice were measured as a function of days after treatment during the half-life studies (Table 17). There was a sig- nificant reduction in both soluble and particulate total organ protein in the liver at 5 but not at l or 3 days after cyclophosphamide. There was no significant change in brain total organ protein or the amount of protein per unit tissue weight in the liver or brain at any of the times studied. Cyclophosphamide treatment, therefore, affected the amount of protein present in neonatal mice but the detection of the effect was dependent on both the tissue and the expression of the data. Qualitative Changes in Neonatal Protein Synthesis Differential gene expression, which results in characteristic tissue specific protein patterns, is an important developmental process. Agents which disrupt the developmentally programmed expression of genetic information may produce abnormal development. Although cyclo- phosphamide did not affect protein synthesis quantitatively, as measured by 1l‘C-leucine incorporation into total protein, drug treatment may have affected the synthesis of specific proteins. Drug induced qualitative changes in protein synthesis, as a result, may have been obscured when protein synthesis was examined quantitatively. Electrophoretic Analysis of Protein Synthesis. The proposal that postnatal development was disrupted as a result of a change in the synthesis of specific proteins was evaluated by double label isotope 86 Table 17. Soluble and particulate protein in the liver and brain of neonatal mice after cyclophosphamide Days Tissuea afterb AProtein/Organ 3g Protein“ Tissue Fraction Treat Control Treated Control Treated Liver Soluble 1 4.7 i 0.6“ 4.7 i: 0.4 59 1; 2 52 i 2 3 7.610.!» 6.01:0.4cl 64:4 66:5 5 12.2 1 1.3 6.4 1 0.5 67 j; 2 74;!- 3 Particulate 1 8.0 i 0.8 7.0 i 0.3 99 j; 3 921- 2 3 12.6 i 0.5 10.5: 1.0d 106 j; 4 114 i 3 5 19.71- l.4 10.2 _-l_; 0.5 109 j; 1 120 j; 9 Brain Soluble 1 3.1 -_l-_ 0.2 3.5 _-l_-_ 0.3 28 i l 30 -_|-_ 2 3 4.2:0.2 4.11:0.1 27:1 28:1 5 5.0il.0 4.91- 0.2 3111 29:1. Particulate l 5.9 1 0.3 6.2 3; 0.2 54 i 3 54 i l 3 8.5 1; 0.8 6.7 i 0.3 54 i- 5 47 j; 1 5 8.9_-|_-_l.4 8.9106 57:3 54:2 aSoluble protein was obtained from the 105,000 g supernatant and particulate protein was obtained from the pellet. l’Day-old mice received either 80 mg/kg s.c. cyclophosphamide or normal saline. cMean 1; SE for 3 observations. dSignificantly different from control (P<0.05). 87 techniques and polyacrylamide gel electrophoresis. An in vitro incu- bation was used to label the newly synthesized proteins since preliminary experiments indicated that proteins from mice treated with 100 tic/RS of 14C(UL)-L-leucine (197 mc/mHole), 100 lie/kg of J'4C(UL)-L-amino acid mixture (1.1 mc/mg), or 500.uc/ks 3H-4,5-L-1eucine (50 uCImMole) for 1 hour periods did not have sufficient specific activity for poly~ acrylamide gel electrophoresis. The ability of the methodology employed in this study to detect qualitative changes in protein synthesis was evaluated by comparing protein synthesis in neonatal brains and livers. The distribution of the percent of total 3B and 14 C incorporated into protein in the gel fraction indicated that different proteins were synthesized in the liver and brain (Figure 10). These differences were more apparent when the 3H/MC ratio was plotted for each fraction. The magnitude of the dif- ferences observed in comparing liver and brain protein synthesis were used as a guide to evaluate possible qualitative changes in protein synthesis produced by cyclophosphamide. The distribution of gel radio- activity from.liver protein of control and cyclophosphamide treated mice were similar at l and 3 days after treatment (Figures 11 and 12). Apparent differences in the distribution of protein radioactivity were not maintained when the tissue was incubated with the other isotope. Qualitative changes in protein synthesis, as measured by the 3H/MC ratio observed when liver and brain protein synthesis was compared, were not evident in livers from cyclophosphamide treated mice. Cyclophos- phamide treatment therefore did not affect the synthesis of specific liver proteins as measured within the resolving power of the methodology employed in this study. ({‘M. 7 ' "TA :1“; ., ~' Hr '. '3. s43. W‘AJ'nf 88 Figure 10. Electrophoretic analysis of proteins synthesized in vitro by neonatal brain (3H-leucine) and liver (14C-leucine). The 3H/140 ratio was calculated on the basis of percent of total radio- activity present in each fraction. 89 «N «N ON 2. 0— S. E. O— m o v N on: "can H u u n eeeeeee in. H uuuuuuuuuu — If I ”n. . o .3 “claw ............. — o l .k\ A..e...s~ \/ ,e ..... ... I x / 3 .... ... s /e IN e\ /(\ I -.V. / _ xx . / \TW _ a . /.\.\// 2:5! L3: 2: 20529.32. I635- u: ....... :58.— 3 DE 20.5.1330: “fig-Ini F Ix o In 1- n N MIAILWOKIVU 1m l0 1N3”!!! O— On Figure 10 f1 l 4‘" all» 90 Figure 11. Electrophoretic analysis of proteins synthesized in vitro by livers from cyclophosphamide treated mice (3H-leucine) or livers from control mice ( AC-leucine) one day after treatment. The 3H/14C ratio was calculated on the basis of percent of total radio- activity present in each fraction. 91 50h 23232 20:35 20......8 QNONVNNNONw—O—SNo—mo lllo..|1 . \ a n - u . q . 4 - q ’.I IO’ \ / \\/ \ I \ [OI a/ P. .e \si 8 \ I \ IN I es oil’s-n a as so I ‘s e \ I ’ use. . saus- O l/ O a. O n. urns-ul- ~\ 0 \o . fi F... . u. x _ u... \ :nl .m>... ..oE,.20u 22. 20:59.30..- esueflbv. ...... ¢u>... 935.: 92. 20:59:32. 3.33-3.-. \ F ODNQWVMN ALIAILDVOKIVI W101 :IO lfllflld 889 Figure 11 1:». 92 Figure 12. Electrophoretic analysis of proteins synthesized in vitro by livers from control (3H-leucine) or cyclophosphamide treated (14C-leucine) mice 3 days after treatment. The 3H/14C ratio was calculated on the basis of percent of total radioactivity present in each fraction. 93 us)...” 3. m... as?” I. . ~32.— Dub