RNA SYNTHESIS AND DNA TEMPLATE ACTIVITY IN THE LIVERS OF RATS FED THE HEPATIC CARCINOGEN. 2-ACETYLAMINOFLUORENE Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSITY MICHAEL PATRICK ADAMS 1976 LIBRARY “/ 3 3 095. Mimi-5.: ; 234.210 I University ' THS PLACE IN RETURN Box to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 6/01 c-JCIRC/DateDue.p65«p.15 ABSTRACT RNA SYNTHESIS AND DNA TEMPLATE ACTIVITY IN THE LIVERS OF RATS FED THE HEPATIC CARCINOGEN 2-ACETYLAMINOFLUORENE By Michael P. Adams It is becoming increasingly evident that the mere presence or persistence of carcinogen-induced DNA base modifications does not auto- matically implicate these lesions as relevant to carcinogenesis. These lesions must be capable of producing some functional change in the DNA which will eventually result in initiation or promotion of carcino- genesis. One potential method of assessing the functional consequences of carcinogen-DNA interactions is to assess the ability of the DNA, in the target organ, to serve as a template for RNA synthesis. Significant decreases in hepatic DNA template activity have been reported after several weeks of feeding the hepatic carciongen 2-acetylaminofluorene (AAF), after a single injection of N-hydroxy-Z-acetylaminofluorene (N- OH—AAF) and after reacting DNA in gi££g_with N-acetoxy-Z-acetylamino- fluorene. 0n the other hand, several investigators have been unable to demonstrate inhibition of chromatin or DNA template activity after N-OH— AAF administration. The studies described in this thesis were designed to examine the feasibility of using DNA template activity as a func— tional measure of carcinogen-induced DNA damage. Michael P. Adams RNA polymerase I activity and the capacity for RNA synthesis were assessed, under conditions optimal for RNA polymerase I activity, in hepatic nuclei isolated from rats fed 0.05% (w/w) AAF for 4, 7 or 14 days. RNA polymerase I activity progressively increased with time on the carcinogenic diet, while the capacity for RNA synthesis remained quite constant. These results suggest a progressive inhibition of DNA template activity during the early stages of AAF—induced hepatocarcino- genesis. The "permanence" of the increase in polymerase I activity was examined by switching carcinogen-fed animals to a control diet for either 2 or 5 days prior to making an assessment of the above parameters. The "permanence" varied, depending upon the duration of carcinogen exposure. A more direct measure of DNA template activity was obtained by isolating hepatic DNA and transcribing the template in_vi££9_with E, ggli_RNA polymerase. Template activity was assessed, in animals main— tained on a schedule of carcinogen feeding similar to that described above, utilizing a method which separates the initiation step of RNA synthesis from the propagation step. Rate and extent of RNA synthesis, RNA chain length and the number of initiation sites were calculated from the resulting data. DNA was isolated from parenchymal cell (N-l) and non-parenchymal cell (N—2) nuclei. The rate of RNA synthesis on N-l DNA was inhibited 24% and 48% after 4 and 14 days, respectively, of AAF ingestion but was unchanged after 7 days. The rate of RNA synthesis on N—2 DNA was inhibited (40%) only after 14 days of AAF feeding. No change in the number of initiation sites could be demonstrated on either template at the end of 4, 7 or 14 days of AAF feeding. When animals were returned Michael P. Adams to the control diet for 2 days, statistically significant increases in the number of initiation sites were observed for N-l and N-2 DNA after 4 days and for N-l DNA after 14 days of AAF feeding. The results of these studies suggest that an early effect of AAF ingestion is an impairment of hepatic DNA template activity. The major mechanism of action appears to be an inhibition of the elongation step of RNA synthesis, although some alterations in the number of initiation sites were observed. RNA SYNTHESIS AND DNA TEMPLATE ACTIVITY IN THE LIVERS OF RATS FED THE HEPATIC CARCINOGEN 2-ACETYLAMINOFLUORENE BY Michael Patrick Adams AN ABSTRACT OF A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Pharmacology 1976 DEDICATION This thesis is dedicated to my loving wife Susie, without whose compassion and understanding I would never have received this degree, and to my parents, John and Shirlee Adams for their never—ending love and guidance which will always be an inspiration for me to pursue higher goals. ACKNOWLEDGEMENT I wish to thank all those professors, graduate students and technologists who contributed both professionally and socially to my two years of graduate education at Michigan State. Particular thanks are extended to Dr. Jay Goodman for his financial support throughout my graduate career and to Dr. Theodore Brody, Dr. Douglas Rickert and Dr. James Trosko for their willingness to serve on my graduate committee. ii TABLE OF CONTENTS DEDICATION ‘- ACKNOWLEDGEMENTS TABLE OF CONTENTS - LIST OF TABLES LIST OF FIGURES LIST OF ABBREVIATIONS -- INTRODUCTION 1.1 1.2 P‘P‘H‘F‘ 0\U1¢~u: MATERIALS 2.1 2.2 2.3 2.4 2.5 2.6 2.7 Chemical Carcinogenesis Metabolism of AAF and Interaction with Cellular Macromolecules DNA Repair Effects of AAF on RNA Synthesis Rationale ----- Research Objectives and Long Range Goals --------- AND METHODS ‘— Animals Carcinogenic Diet - Chemicals Isolation of Nuclei Assay for the Capacity for RNA Synthesis in Hepatic Nuclei Assay for RNA Polymerase Activity in Hepatic Nuclei Isolation of Hepatic Parenchymal and Non-Paren- chymal Cell Nuclei - Isolation of Hepatic DNA — Measurement of the Rate and Extent of RNA Synthe— sis on Rat Liver DNA Determination of the Number of Initiation Sites for RNA Synthesis -— Calculations and Statistics Determination of E, coli RNA Polymerase Activity Using Calf Thymus DNA ————— Miscellaneous Methods -- iii Page ii iii vi viii 12 16 17 19 l9 l9 19 20 21 22 23 24 25 27 28 30 30 Page RESULTS -- --- 31 3.1 Determination of the Capcity for RNA and Poly A Synthesis in Hepatic Nuclei 31 3.2 Determination of the Capacity for RNA Synthesis and RNA Polymerase I Activity in Carcinogen- Treated Animals -—- 39 3.3 Determination of the Rate and Extent of RNA Syn— thesis, RNA Chain Length and Number of Initiation Sites on DNA Isolated from Carcinogen—Treated Animals 49 DISCUSSION 69 SUMMARY AND CONCLUSIONS ---- 84 BIBLIOGRAPHY - ----- 87 iv Number 10 LIST OF TABLES Research Objectives and Long Term Goals ————— Effect of Varying the Amount of Nuclei in the Reaction Mixture Upon RNA Polymerase Activity-- Effect of Varying the Amount of Nuclei in the Reaction Mixture Upon RNA Poly- merase I Activity — -- Effect of Ribonuclease and Protease Concen— trations on Percent Contamination of DNA—-—- Variability of the Rate and Extent of RNA Synthesis Between DNA Preparations ---------- Percent Protein and Percent RNA Contami— nation on DNA Isolated from Parenchymal (N—l) and Non-Parenchymal (N-Z) Cell Nuclei —-- Relationship Between Number of Initiation Sites and Percent Contamination of DNA ------ Stability of E, coli RNA Polymerase---——---- Effect of a Single Injection of AAF on Rate and Extent of RNA Synthesis ————————————————— Effect of AAF Ingestion on the Extent of RNA Synthesis, Number of Initiation Sites and RNA Chain Length on DNA Isolated from Paren— chymal (N-l) and Non-Parenchymal (N-Z) Cell Nuclei — ----- Page 18 46 47 50 52 54 56 58 6O 67 Number 10 11 LIST OF FIGURES The effect of Varying the Amount of Nuclei in the Reaction Mixture Upon Incorporation of 3H-ATP into RNA and Poly A —— Determination of Optimal Ionic Conditions for Assay of RNA Synthesis by RNA Polymerase I In Hepatic Nuclei —— Determination of Optimal Ionic Conditions for Assay of RNA Synthesis by RNA Polymerase II in Hepatic Nuclei ————— Variability of the Capacity for RNA Synthesis Between Nuclei Preparations -- Effect of AAF Ingestion on RNA Synthesis and Poly A Synthesis by RNA Polymerases I and II in Hepatic Nuclei -- Effect of the Duration of AAF Ingestion on RNA Synthesis and Poly A Synthesis by Polymerases I and II in Hepatic Nuclei Effect of Actinomycin-D on RNA Synthesis by RNA Polymerases I and II in Hepatic Nuclei ------- Effects of Ethanol, Actinomycin-D and Omission of CTP and GTP on RNA Synthesis by RNA Poly- merases I and II in Hepatic Nuclei -- RNA Synthesis by RNA Polymerases I and II in Hepatic Nuclei in the Presence and Absence of Poly (dA—dT) - Effect of AAF Ingestion on RNA Synthesis and RNA Polymerase I Activity in Hepatic Nuclei ------ Effect of Varying the Concentration of E, coli RNA Polymerase and DNA in the Reaction Mixture Upon Incorporation of 3H-UTP Into RNA vi Page 32 33 34 36 37 4O 41 43 44 48 51 Number Page 12 Relationship Between Percent Contamination of DNA and Rate and Extent of RNA Synthesis --------- 55 13 Rate and Extent of RNA Synthesis on Control DNA Isolated from Parenchymal (N-l) and Non- Parenchymal (N—2) Cell Nuclei ---- 59 14 Effect of AAF Ingestion on the Rate of RNA Synthesis on Parenchymal (N-l) and Non- Parenchymal (N—2) Cell DNA 61 15 Determination of the Number of Initiation Sites on Control DNA Isolated from Parenchymal (N-l) and Non-Parenchymal (N-2) Cell Nuclei ------ 64 16 Determination of the Number of Initiation Sites on DNA Isolated from Parenchymal (N-l) and Non-Parenchymal (N-2) Cell Nuclei from AAF-treated Animals 65 vii AAF N-OH-AAF N-Ac-AAF ATP CTP DNA GTP poly A poly (dA-dT) RNA TCA UTP LIST OF ABBREVIATIONS 2-acetylaminofluorene N-hydroxy-Z-acetylaminofluorene N-acetoxy-Z-acetylaminofluorene adenosine 5' triphosphate cytidine 5' triphosphate deoxyribonucleic acid guanosine S'triphosphate poly adenylic acid a high molecular weight, doublt stranded copolymer composed of alternating dA and dT units ribonucleic 3 (id trichloroacetic acid uridine 5' triphosphate viii INTRODUCTION 1.1 Chemical Carcinogenesis In the past 50 years, a large number of chemicals have been identi— fied as carcinogenic in animals. Structures of these compounds vary considerably, from highly complex natural products such as aflatoxin and cycasin to metals such as chromium, nickel and beryllium (1). Because these carcinogens exhibit considerable variability as to species and tissue specificity, there is currently no unequivocal method for extra- polating the production of experimental cancers in animals to environ- mentally produced human cancer. Chemicals have been implicated in human cancer largely on the basis of epidemiological studies. Such studies usually examine a small number of workers exposed to high concentrations of chemicals in an industrial situation or differences in the incidence of specific cancers between isolated geographical regions (1). As a result, several chemicals have been identified as probable causes of human cancer, including asbestos, cigarette smoke and the industrial compounds 4-aminobiphenyl, benzedrine and 2-napthylamine (2). Although no definitive data exist on the proportion of human cancer that develops as a result of environmental chemical exposure, epidemiological studies suggest that a large percentage, perhaps 80%—90%, of all human cancers have environmental factors in their etiology (2). The magnitude of the 2 problem demonstrates the need for a greater understanding of the under- lying mechanism(s) of chemical carcinogenesis. Carcinogenesis appears to consist of two major stages, initiation and promotion. Initiation of carcinogenesis is thought to consist of an essentially irreversible, heritable but unexpressed change in one or more informational macromolecules (3). Initiation may occur with "sub- carcinogenic" doses, proda ing no apparent morphological change, such that the initiated state can be detected only by subsequent exposure to a promoting stimulus (4). The end result of initiation is the produc- tion of a potential cancer cell. If followed by an appropriate stimulus, the previously unexpressed change will become expressed and tumor pro- gression will begin. Tumor promotors do not appear to be merely weak carcinogens. Application of promotor alone does not consistently result in tumor production and their action is dependent upon the order of application, 3,3,, promotors applied prior to initiation are generally not effective (3). In additiOn, a promotor may be effective when applied several months after the initiation event (5). Although not well understood, the process of promotion appears to involve a pro- liferatory stimulus resulting in tissue hyperplasia (3). Although the model tissue for studying the two stages of carcino- genesis has been mouse skin, several studies have suggested that these processes also occur in the liver. Peraino §E_§E, (6) demonstrated that feeding of phenobarbital or dichlorodiphenyltrichloroethane (DDT) to rats previously maintained on a 0.02% AAF diet for only 18 days, markedly enhanced hepatoma formation. Amobarbital, structurally very similar to phenobarbital, had no promoting action (6). Craddock (7) 3 significantly enhanced hepatoma frequency by performing partial hepa- tectomies on rats previously treated with the carcinogen dimethylni— trosoamine. Furthermore, Armuth and Berenblum (8) demonstrated these processes in mice by following a "sub-carcinogenic" dose of dimethyl- nitrosoamine with repeated injections of phorbol. While no hepatomas appeared in the group receiving phorbol alone, the group receiving carcinogen followed by phorbol developed a high incidence of tumors (8). Of particular importance to the study of chemical carcinogenesis are the observations that; l) the carcinogen appears to produce a very early, possibly irreversible change in an information macromolecule; 2) the carcinogen need not be present during the entire latent period of tumor production, and 3) "sub-carcinogenic" doses of a chemical carcino- gen can produce high incidences of tumors if the appropriate promoting stimulus is applied. 1.2 Metabolism of AAF and Interaction with Cellular Macromolecules While metabolism of most compounds results in their transformation to less chemically reactive species which can be readily excreted (9), metabolism of many chemical carcinogens results in small amounts of highly reactive species capable of interacting with cellular components (10). Formation of such highly reactive compounds may require several metabolic steps; 1,3, transformation of a ”non-carcinogen" to a proxi— mate carcinogen and eventually to an ultimate carcinogen. A charac— teristic common to most and probably all ultimate carcinogens is the possession of strong electrophilic properties (10). These highly electrophilic reactants are capable of interacting non-enzymatically at a large number of nucleophilic sites, including those in glycogen, DNA, 4 RNA and proteins. The indiscriminate nature of this binding makes elucidation of the critical target(s) of chemical carcinogens a monu- mental task. One of the most extensively studied carcinogens, in terms of its metabolic fate, is the hepatocarcinogen AAF. Early studies identified a metabolite of AAF, N-OH-AAF (11,12), which when administered to animals, exhibited greater carcinogenicity (l3) and a higher degree of binding to nucleophiles (14,15) than the parent compound. The importance of the N— hydroxylation product in AAF-induced carcinogenesis has been demon- strated by several investigators. Irving gEHEE. (16) demonstrated that the male Sprague-Dawley rat, a species highly susceptible to AAF- induced hepatocarcinogenesis, can readily N—hydroxylate AAF in the liver, as evidenced by the high percentage of this metabolite excreted in the bile. However, the female Fischer rat, a species refractory to AAF- induced hepatic tumors, excretes very little N—OH—AAF in the bile and thus appears to have an impaired capacity for N-hydroxylation in the liver (17). This correlation between inability to N-hydroxylate AAF and lack of significant carcinogenicity also exists for the rabbit (16) and guinea pig (18). Substantial evidence has accumulated implicating esters of N-OH—AAF as ultimate reactive metabolites of AAF. Synthetic esters such as N—Ac- AAF bind avidly to proteins and nucleic acids EEHXEEEQ (10,19) and several of the reaction products are identical with those obtained after EE_XEyg_administration of AAF or N-OH—AAF (10,19). In the liver, the ultimate carcinogenic form of AAF is believed to be the sulfate ester of N-OH—AAF. DeBaun g3 3E. (15) demonstrated the importance of the sulfate 5 ester by showing a positive correlation between hepatic sulfotransferase activity and susceptibility to AAF-induced hepatomas in mice, hamsters guinea pigs, rabbits and male and female rats. This correlation was confirmed by Gutmann E£_§l° (17) in two species of female rats. In addition, reducing the liver sulfate pool by treatment with p-hydroxy- acetanilide (20) or decreasing liver sulfotransferase levels by hypo- physectomy, gonadectomy or thyroidectomy (15) can inhibit carcinogenesis by AAF. However, it is probable that other metabolites of AAF can contribute to carcinogenicity by this agent since some tissues exhi- biting AAF-induced tumors do not possess significant sulfotransferase activity (21). Binding of chemical carcinogens to protein was first reported many years ago (22). Early studies demonstrated that certain proteins exhibiting binding to azo dyes soon after carcinogen administration are absent from the resulting hepatoma (23). Subsequent EE_XEEEQ_studies with N—Ac-AAF resulted in the identification of methionine, cysteine, tryptophan and tyrosine as target amino acids (19). De Baun 22.21: (15) identified two methionyl—AAF adducts in rats fed AAF or N-OH-AAF which appeared identical to those formed 12.21532; thus supporting the exi— stence of these reactions 32.23293 In addition, this group also demon- strated that formation of methionyl—AAF derivatives Eg_ylyg correlated well with the hepatic carcinogenicity of N-OH-AAF in several species (15). Metabolites of AAF have been shown to bind to histones (24) and acidic nuclear proteins (24,25), both of which are thought to be in- volved in controlling gene expression. Interaction of a carcinogen with 6 a critical cellular protein could, theoretically, result in a permanent, heritable change in gene expression, possibly leading to carcinogenesis (26). Several studies suggest, however, that proteins may not be critical targets of chemical carcinogens. The amount of protein-bound carcinogen does not always correlate well with carcinogenicity in certain rodents (27,28) and the ability to participate in protein binding is not unique to chemical carcinogens (29,30). In addition many protein-bound AAF adducts exhibit a relatively short half-life compared to some AAF-DNA moieties (24), although this may simply reflect the slow turnover of the latter macromolecule in many organs. Another possible critical target of chemical carcinogens is RNA. N—Ac-AAF reacts well with RNA $2.XEE£2.(31) and this interaction has been shown to produce defects in amino acid acceptance capacities, codon recognition and ribosomal binding of several transfer RNA molecules (32). Covalent binding of AAF metabolites to RNA has been reported to occur after 13 ylyo_administration of AAF or N-OH—AAF (33-35) and the major reaction product formed is believed to be N—(guanosin—8—yl)-2-AAF (31,36). The extent of binding of AAF metabolites to RNA is greater in male rat liver than female rat liver (34,37); the latter being rela— tively non-susceptible to hepatic carcinogenesis by this agent. In addition, two non-carcinogenic derivatives of N-OH-AAF, 1-hydroxy-2- acetylaminofluorene (35) and N-hydroxy-3-acety1aminofluorene (38), exhibit negligible binding to this macromolecule. 7 Several studies suggest that RNA may not be a critical target of chemical carcinogens. Irving gE_§l, (35) demonstrated that detectable amounts of AAF metabolites are bound to RNA in tissues which are rela- tively non—susceptible to AAF-induced carcinogenesis, such as rabbit and guinea pig liver, after a single injection of AAF or N-OH-AAF. Matsu- shima gE_§E, (33) prefed various inhibitors of AAF-induced hepato— carcinogenesis (m-acetatoluidine, indole, acetanilide, chloramphenicol) to rats and examined the binding of AAF metabolites to the various hepatocellular fractions after a single dose of N—OH—AAF. While pre- feeding of these inhibitors generally decreased the initial binding of AAF metabolites to DNA, no such correlation could be demonstrated for RNA, Numerous studies have presented data implicating DNA as a critical target of chemical carcinogens. The degree of covalent binding of a chemical carcinogen to DNA correlates well with its ability to induce tumors. This relationship has been shown to exist for a series of aromatic hydrocarbons (39), a series of alkylating agents (40) and certain metabolites of 4—nitroquinolin-l-oxide (41). Furthermore, prefeeding of several inhibitors of AAF-induced carcinogenesis, in- cluding indole, chloramphenicol, m-acetotoluidide and acetanilide markedly decreased the extent of AAF moieties bound to DNA after a single injection of N-OH-AAF (33). Attempts have been made to identify specific structural alterations in DNA after carcinogen administration with the expectation that such knowledge would allow one to predict the functional consequences of such 8 an interaction on the molecular level. Experiments using double la- belling techniques (3H-acetyl group, 14C-fluorene ring) indicate that 70% to 80% of the carcinogen molecules bound to DNA after a single injection of AAF (42) or N-OH-AAF (14) are deacetylated. This is in contrast to results with RNA which suggest that 70% of the bound AAF molecules retain the acetyl group (14,35,43). Eighty percent of the acetylated AAF moieties bound to DNA have been identified as N-(deoxy- guanosin-8-yl)—2—AAF (31,42). Although not well characterized, the remaining 20% of the bound acetylated AAF moieties are believed to be 3- (guanin-Nz-yl)—2-AAF (19,42). Binding of AAF metabolites to adenine is not well understood and several investigators have failed to identify such base modifications (19,31,43). At least one study, however, has suggested the existence of modified adenine in DNA (44) and under certain circumstances, N-Ac-AAF can react with poly A.12.X}E£2.(45’46)- Failure to detect AAF modifications of adenine, cytidine or thymidine bases may be due to artifacts in the isolation procedure or to the limited sensitivity of the analytical methods used to identify these bases. Numerous biochemical and biophysical studies on AAF—modified DNA and oligonucleotides have led to an "insertion-denaturation" ("base displacement") theory (46-49). According to this model, binding of an AAF metabolite to the C-8 of guanine causes a rotation of the base around the glycosidic bond from anti to syn (47,50). In such an arrange- ment, a modified guanine base is displaced from its normal position by the fluorene moiety and the latter becomes stacked coplanar with ad- jacent bases. This model explains the local denaturation of DNA 9 suggested in several studies (44,48,51,52) and predicts general desta- bilization of double helical conformation and disruption of normal hydrogen bonding between affected bases (52). Several studies have demonstrated that certain AAF moieties bound to DNA can exhibit a prolonged biological half-life. Epstein 33.31. (53) demonstrated that hepatic DNA isolated from the hyperplastic nodules of AAF—fed rats exhibited an altered ultraviolet absorbance spectrum indicative of AAF-modified DNA, four weeks after discontinua- tion of the carcinogenic diet. These results were confirmed by Szafarz and Weisberger (54) who identified DNA-bound radioactivity in the livers of rats eight weeks after removal of a (9-14C) N-OH-AAF diet. Per— sistant AAF-DNA adducts have also been identified in rat liver after a single injection of AAF (14) or N—OH-AAF (42). Irving and Veazy (14) determined the rate of disappearance of radioactivity from ribosomal RNA and DNA after a single injection of AAF and demonstrated that radio- activity remained bound to the former for only 2 weeks and with the latter throughout the entire eight week study. Kriek (42) isolated two reaction products after a single injection of N—OH-AAF, one having a half-life of seven days and the other remaining bound to DNA for eight weeks. The persistent moiety, accounting for 20% of the total bound AAF molecules, was later identified as 3-(guanin-N2-yl)-2-AAF (19). 1.3 DNA Repair Most cells possess a mechanism whereby certain types of DNA damage can be removed and the affected region repaired to its original struc- ture. This phenomena, termed excision repair, has been extensively 10 studied in procaryotes (55) and mammalian cells in culture (56) and appears to involve at least five steps (58,57): 1) recognition of base damage; 2) production of a single strand break near the lesion by an endonuclease; 3) excision of the affected base(s); 4) re-insertion of a correct base sequence using the complementary strand as a template, and 5) annealing of the repaired region to the daughter strand by a ligase. The fidelity of this process has been assessed in mammalian cells and it appears to be an error-free process (60,59). Mutant cells deficient in excision repair have been isolated from patients with Xeroderma pigmen- Egggg; a disease characterized by an increased susceptibility to ultra- violet light-induced skin cancer (61). The hypothesis that this in— creased susceptibility is due to a deficiency in a step of the excision repair process is attractive, but is as yet unsubstantiated. Certain variants possessing all the clinical symptoms of this disease appear to have no detectable defect in excision repair (62). A second type of DNA repair which has been demonstrated in pro— caryotes and postulated to exist in eucaryotes is called post—replica- tion repair (57). Although the process is poorly understood, it is thought to occur when a DNA polymerase bypasses a DNA lesion and con— tinues synthesis distal to this site. A "gap" is created on the newly synthesized strand opposite the parental lesion and this "gap" is filled in at a later time by some unknown mechanism. This process does not actually repair the lesion but it does provide for cell survival by allowing the DNA polymerase to circumvent a potentially lethal block to DNA replication. The original lesion will be diluted out in subsequent 11 cell divisions. Painter (58) reviewed the available evidence for post- replication repair and concluded that this type of repair, as described by Lehman (63), does not exist in mammalian cells. In a recent article, Higgins gE_§E, (64) proposed a model of replication repair which in- volves a bypass of the DNA lesion without the production of a "gap". In this model, the DNA synthesized off the undamaged strand acts as a template for replication of the damaged region in the homologous paren- tal strand. Theoretically, this type of repair would be error free as opposed to the process of post—replication repair which, in procaryotes, is error prone (64). Several studies have suggested that there are some reaction products formed between chemical carcinogens and DNA which are not readily re- paired by the DNA excision repair process. Kriek (42) demonstrated the presence of two metabolites of AAF bound to DNA which exhibited markedly different biological half—lives. Whereas guanine modified by AAF at C-8 was removed with a half-life of 7 days, guanine modified by AAF at the 2—amino group remained in DNA eight weeks after a single injection. No persistent aminofluorene moieties could be identified. Utilizing molecular models, Kriek concluded that AAF substituted at the 2-amino group of guanine creates little distortion of the double helical con- formation and thus may not be removed by DNA excision repair (19). Goth and Rajewsky (65) examined the initial extentof binding and rate of removal of three base modifications, 06-ethylguanine (believed to cause mis—pairing during DNA replication [65]), N3-ethylguanine and N7- ethylguanine, produced by the alkylating agent N—ethyl—N—nitrosourea. 12 While the initial degree of alkylation to all three sites was essentially equal, the 06 modification was removed considerably slower than the others. Furthermore, the rate of removal of 06-ethylguanine was markedly slower in target tissue (brain) than in non-target tissue (liver) (65). These results suggest the importance of those reaction products not readily repaired by the excision repair process to carcinogenesis. However, in both of the studies cited above, the direct participation of the DNA excision repair process in preferential removal of a modified base was not demonstrated. The possibility exists that these modified bases were removed non-enzymatically due to instability or decomposition of the carcinogen-DNA adduct. The consequences of DNA lesions not readily repaired by the excision repair process have not been unequivo- cally determined, however, some appear to be potentially damaging (66) and may lead to cell death, mutation and possibly neoplasia. 1.4 Effects of AAF on RNA Synthesis A potential method of assessing the functional consequences of carcinogen-DNA interactions is to measure the ability of the modified DNA in the target organ to serve as a template for RNA synthesis. Since hepatic RNA synthesis is occurring continuously at a relatively high rate, there is no need to perform a partial hepatectomy as is often the case when assessing DNA synthesis in the liver. Although several investigators have attempted to assess the effects of AAF on RNA syn- thesis and DNA template activity, the results obtained have been va- riable, depending upon the mode of carcinogen administration and the methods utilized to assess the experimental parameters. 13 Most investigators have reported a decrease in rat liver RNA synthesis several hours after a single injection of N—OH-AAF (67-72). However, there is disagreement as to whether this inhibition is due to an impairment in the capacity of DNA to serve as a template or to an inhibition of RNA polymerase activity. Initial studies by Grunberger EE.§l: (69) indicated a decrease in hepatic nucleolar RNA synthesis with no change in the activity of RNA polymerase I, as measured on the synthetic template poly (dA-dT). They interpreted this data as indicative of an impaired DNA template activity. Herzog gE_§E, (71), using an assay procedure similar to that used by Grunberger EEH§l° (69), demonstrated a decreased RNA polymerase activity after AAF administra— tion which they claimed could account for the observed inhibition in RNA synthesis. In addition, Herzog 22.2l: could not demonstrate a decreased template activity on hepatic chromatin isolated from car- cinogen—treated animals (71). In agreement with Herzog gE_§E, were the results of Glazer gE_§E, (67) and Zieve (68). Neither of these in- vestigators could demonstrate an impaired template activity on DNA isolated from animals given a single injection of N—OH—AAF. In an attempt to resolve the apparent discrepancy, Yu and Grunberger (73) very recently confirmed their previous results (69) by demonstrating an impaired nucleolar template activity after a single dose of N-OH- AAF. However, their data also suggested the possibility of a direct alteration of RNA polymerase II by AAF metabolites, as was recently suggested by Glazer (74). Thus, AAF may have two mechanisms of RNA synthesis inhibition after acute administration; 1) alteration of l4 nucleolar template activity, and 2) selective inhibition of RNA poly- merase II. Troll_gE_§E. (75) were the first to isolate hepatic DNA from animals fed an AAF diet for several weeks and assess template activity using a bacterial RNA polymerase. They demonstrated that the ability of the DNA to serve as a template for RNA synthesis after AAF ingestion was substantially decreased. The degree of DNA template activity inhibition was somewhat variable, depending upon the duration of carcinogen feeding. Adams and Goodman (76) confirmed the results of Troll_gEH§E. (75), presenting data suggesting a decreased template activity in hepatic nuclei isolated from rats fed an AAF diet for 7 or 14 days. Troll 2E 3E. (51) also demonstrated that DNA modified Ea KEE£2_with N—Ac—AAF exhibits an impaired template activity. Extensive modification by N-Ac-AAF completely abolished the capacity of DNA to serve as a template for RNA synthesis. These results were confirmed by Zieve (77), Glazer §E_§l, (67) and Millette and Fink (78), who also reported over 90% inhibition of DNA template activity after reaction of the macro- molecule with high concentrations of N—Ac—AAF EE_XEE£93 In addition, Zieve reported template activity inhibition after reaction of DNA 12. yEE£2_with N-OH—AAF, although very high concentrations were required to produce minimal changes (77). Unfortunately, many studies assessing RNA synthesis on DNA isolated from carcinogen—treated animals have utilized relatively unsophisticated methods of measuring transcription. Some assay procedures (67,68,71, 77) measure a total amount of RNA synthesized after a fixed incubation period under conditions where results may be a complicated function of 15 the number of initiation sites available, the rate of chain propagation and RNA chain length. Such methods do not allow dissection of the mechanism of AAF-induced template activity inhibition. A few studies have attempted to distinguish among these mechanisms (74,78,77). The major effect of AAF upon transcription appears to be a decrease in RNA chain elongation (74,77,78), possibly due to premature chain termination (78). The number of initiation sites for RNA synthesis may also de- crease with extensive DNA modification (77,78). It should be emphasized that these studies utilized DNA modified 12.Xi££2.With N-Ac-AAF. No studies have been performed examining the mechanism of template activity inhibition of DNA modified ifl.XlXE.With AAF. The study of precancerous liver offers a particularly difficult problem in that the organ undergoes major histological changes following carcinogen feeding (79-82). In addition, the liver contains several cell types, some of which may change biochemically (83,84) as an animal progresses on a carcinogenic diet. Consequently, many studies examining effects of AAF on liver biochemistry measure para- meters of a very diverse population of cells, most of which were pro- bably not destined to become tumorogenic. In order to understand critical events leading to neoplasia, techniques should be developed to identify, isolate and study those cells which are direct precursors of the cancerous lesion (85). Experimental procedures have been developed to isolate nuclei from different hepatic cell types (86-88). A rela- tively simple procedure developed by Bushnell gE_§l, (86) permits a relative separation of parenchymal cell nuclei (hepatocytes) from non— parenchymal cell nuclei (non-hepatocytes). Such an approach is potentially 16 very useful due to the fact that some carcinogens, e.g., AAF, produce primarily hepatocellular tumors which appear to arise from parenchymal cells (79,80) and other carcinogens, e.g., 3'-methyl-4-dimethylamino- azobenzene, produce tumors presumably arising from both cell types (79,80). Identification of selective effects upon the template activity of parenchymal cell DNA during the very early stages of AAF-induced hepatocarcinogenesis could have particular relevance to carcinogenesis by this agent. 1.5 Rationale It has become axiomatic that chemicals induce cancer as a result of binding to one or more cellular macromolecules. Convincing evidence has been presented implicating DNA as a critical target of chemical carcinogens. Damage to DNA resulting in alteration in the capacity of the macromolecule to serve as a template for RNA synthesis could provide a mechanism for qualitative or quantitative changes in RNA production, eventually leading to synthesis of aberrant proteins. Abnormal pro- duction of certain critical proteins, such as those involved in nucleic acid biosynthesis or repression of genes controlling cell growth, could result in permanent, heritable alterations in gene expression. If the DNA damage producing the above sequence of events can be demonstrated very early in AAF-induced hepatocarcinogenesis, 3,3,, the time at which the initiation stage of carcinogenesis is thought to occur, such a phenomenon could have particular relevance as an initial molecular event leading to neoplasia. 17 1.6 Research Objectives and Long Range Goals Research objectives and goals are summarized in Table #1. 18 TABLE 1 Research Objectives and Long Term Goals 1. Objectives A. Does AAF feeding affect the ability of hepatic DNA to serve as a template for RNA synthesis? B. If there is a change in template activity; 1. How is it altered as an animal progresses on the carcinogenic diet? 2. How permanent are the changes? 3. Is there a particular species of RNA or a particular hepatic nuclei population which is selectively inhibited? 4. What is the molecular mechanism of the template activity inhibition? 11. Long-term Goals A. Can DNA template activity be used to assess the functional consequences of carcinogen-DNA interactions? B. Are changes in DNA template activity relevant to carcino- genesis, i.e., is this an initial molecular event leading to neoplasia? MATERIALS AND METHODS 2.1 Animals Male Sprague—Dawley rats (Spartan Research Animals Inc., Haslett, MI), weighing 175 i 25 g, were used in all experiments. Rats used for any given experiment were received on the same day and housed in a windowless room kept dark from 7 a.m. to 7 p.m. and light the other 12 hours of the day. Food and water were allowed 33_libitum until the time of sacrifice, 9 a.m. i 1 hour. 2.2 Carcinogenic Diet A control basal diet (79) (carcinogenic basal diet) containing U.S. Pharmacopiea XIV salt mix and supplemented with p-aminobenzoic acid, 0.11 g/kg diet; inositol, 0.11 g/kg diet; and dry vitamin E acetate, 0.24 g (121 international units)/kg diet was purchased from Teklad Mills, Madison, WI. This control diet was supplemented with 0.05% (w/w) AAF purchased from Eastman—Kodak or Aldrich Chemical Co. Animals were fed the control diet for 3-5 days before beginning the carcinogenic diet, and were always started on the control and carcinogenic diets at 8 a.m. i 1 hour. 2.3 Chemicals Ribonuclease A (Sigma Type XI-A), protease (Sigma Type VI), actino- mycin D and the sodium salts of cytidine triphosphate (Sigma Type III), 19 20 guanosine triphosphate (Sigma Type III), uridine triphosphate (Type I) and adenosine triphosphate were purchased from Sigma Chemical Co. Poly (dA-dT) and E, 3331_K-12 RNA polymerase were purchased from Miles Laboratories, calf thymus DNA from Worthington Biochemical Corp., and Multisol from Isolab Inc., Akron, Ohio. Uridine (5-3H)5' triphosphate and adenosine (2,8-3H)5' triphosphate were purchased from New England Nuclear. 2.4 Isolation of Hepatic Nuclei Rat liver nuclei were selected as the initial system in this study because 1) the liver is the major target organ for AAF-induced carci— nomas; 2) the procedure for nuclei isolation is relatively simple and well characterized; 3) the assay for RNA synthesis in hepatic nuclei has been extensively studied and parameters such as optimal ionic strength, optimal temperature of incubation and requirements for the polymerization reaction have been known for many years (89); and 4) the system has been used successfully by others attempting to assess carcinogen-induced changes in RNA synthesis (67,69,7l,73,90). Hepatic nuclei were isolated by the method of Yu and Feigelson (91). Animals were stunned by a blow to the head and their excised livers placed in a beaker of ice cold 0.34 M sucrose. Ten grams of liver were homogenized in 2.3 M sucrose; 3.3 M CaCl utilizing a glass 2 homogenizer fitted with a motor-driven teflon pestle. The homogenate was adjusted to a final volume of 100 ml, filtered through cheesecloth and centrifuged at 40,000 x g for 1 hour at 0°-5°C. The supernatant was discarded and the pellet resuspended in 10 ml of ice cold 0.34 M sucrose. I I .l I .i I . I. I p. I . l I I 1 21 An aliquot of this nuclei suspension (0.2 ml) was stored at -90°C for assessment of DNA content on a later date. The remainder of the sus- pension was kept on ice until needed (15 i 10 minutes). 2.5 Assay of the Capacity for RNAEEynthesis in Hepatic Nuclei A major disadvantage of measuring RNA synthesis in a nuclei pre- paration is that the total amount of RNA synthesized during a fixed incubation period is a complex function, depending upon the activity of endogenous RNA polymerases and the ability of the DNA to serve as a template for RNA synthesis, 3,3, template activity. Chemical carci— nogens could affect 33_X333_RNA synthesis by altering enzyme, template or both. Unfortunately, the ability of the DNA to serve as a template in the transcriptional process cannot be measured directly in hepatic nuclei. However, by concomitant assessment of the capacity for RNA synthesis and RNA polymerase activity, an indirect measure of changes in template activity can be obtained. The usefulness of this proce- dure has been demonstrated (76,69,73). Mammalian cells contain multiple RNA polymerases (89). RNA poly— merase I is nucleolar in origin and catalyzes the synthesis of ribosomal RNA while RNA polymerase II is nucleoplasmic in origin and synthesizes primarily messenger RNA (89). RNA polymerase III has been identified but its role in RNA synthesis is incompletely understood (89). To examine the possibility of AAF selectively inhibiting synthesis of a particular RNA species, a simple method was selected which separates RNA synthesis by the two major polymerases on the basis of ionic strength of the reaction mixture. RNA polymerase I is active when 22 . . . . ++ . assayed in a low salt media containing Mg and RNA polymerase II is active when assayed at high ionic strength in the presence of Mn++ (89). The basic reaction mixture employed was modified from Grunberger 33 33, (69) and contained 100 mM Tris-H01 (8.2); 2.0 mM dithiothreitol; 0.4 mM each GTP, CTP, UTP; 0.4 mM 3H-ATP (20 Ci/mole), and 50 ul nuclei (70 i 20 ug DNA), in a final volume of 0.2 ml. Conditions optimal for RNA synthesis by the nucleolar RNA polymerase I included the basic re— action mixture described above, plus 6.0 mM MgCl2 (89). Conditions optimal for RNA synthesis by the nucleoplasmic RNA polymerase 11 included the basic reaction mixture described above supplemented with 3.0 mM MnCl and 0.2 M (NH 2 SO4 (89). Reactions were initiated by the 4)2 addition of nuclei and incubation was for 3 minutes at 30°C (92), during which time the rate was linear and directly proportional to the amount of nuclei added. Reactions were stopped by the addition of 2.0 ml ice— cold 5% TCA. Bovine serum albumin, 0.5 ml of a 0.5% solution, was added and the precipitate collected by centrifugation for 5 minutes at 1000 x g. The precipiate was washed twice with 2.0 ml portions of cold 5% TCA and once with ethanol; ether (4:1 v/v), prior to being dissolved in 0.5 ml of 88% formic acid (93). Multisol, 10 ml, was added and the radio- activity measured with a Packard Model 3380 liquid scintillation spec— trometer . 2.6 Assay_for RNA Polymerase Activity in Hepatic Nuclei In the initial experiments, synthesis of poly A was selected as a potential method for assessing RNA polymerase activity, since the procedure is fast, simple and relatively inexpensive. Since AAF 23 metabolites do not appear to bind to thymine bases in DNA, thymine— rich regions should be relatively free of AAF modification and RNA synthesis in these regions should be somewhat independent of AAF— induced changes in template activity. Therefore, changes in poly A synthesis following AAF administration could potentially reflect changes in RNA polymerase activity relatively independently of changes in tem- plate activity. Conditions utilized for assessment of poly A synthesis by poly- merases I and II included the reaction mixtures described in section 2.5 for the respective polymerase; omitting GTP, CTP and UTP. All reactions were incubated 10 minutes, during which time the rate was linear and directly proportional to the amount of nuclei added. RNA polymerase I activity was assayed using the basic reaction mixture described in Section 2.5 containing 20 ul of nuclei (28.6 ug of DNA), 8 ug poly (dA—dT), 4 ug actinomycin—D and omitting GTP and CTP. All reactions were run for 10 minutes at 30°C during which time the rate was linear and directly proportional to the amount of nuclei present. The reaction was terminated, precipitate washed and radio- activity measured as in Section 2.5. 2.7 Isolation of Hepatic Parenchymal and Nonjparencymal Cell Nuclei Hepatic parenchymal (class N-l, hepatocytes) and non-parenchymal (class N—2, "stromal") cell nuclei were prepared by the procedure of Bushnell 33_33, (86). The validity of the procedure has been reviewed (95) and the usefulness of the procedure has been demonstrated (70,84, 95). 24 2.8 Isolation of Hepatic DNA The method of DNA isolation utilized was a modification (96) of the Marmur procedure (97). The Marmur procedure is widely used, well characterized and yields DNA reasonably free from nucleoprotein or RNA contamination. One disadvantage of the procedure is it involves the use of highly lipophilic agents such as chloroform and phenol. Since AAF is lipid soluble, it is possible that pieces of DNA con- taining multiple AAF modifications could become more lipid soluble and be lost in the organic phase during the isolation procedure. If this were to occur, there would be a loss of AAF modified DNA and data for template activity inhibition may represent minimum values. In other words, the true template inhibition may be greater than that actually measured in these experiments. DNA was prepared from parenchymal (class N-l) and non—parenchymal (class N-2) cell nuclei. Nuclei were suspended in 10 mM Tris-HCl (pH 7.9); 0.1 M NaCl; 5.0 mM EDTA; 0.5 M NaClO 1% sodium dodecyl sulfate 4; and the solution was incubated for 40 minutes at 37°C with constant shaking. The suspension was deproteinized three times with chloroform; 3% isoamyl alcohol. To the resulting aqueous phase, an equal volume of ice-cold 95% ethanol was added and the nucleic acids were wound on a glass rod and transferred to a flask containing 10 mM Tris-HCl (pH 7.9); 5.0 mM EDTA. This suspension was treated with ribonuclease (50 ug enzyme/ml) for one hour and protease (3 mg enzyme/ml) for an additional two hours at 37°C. The solution was then deproteinized, once with an equal volume of chloroform; 3% isoamyl alcohol and once with an equal volume of freshly distilled phenol. The DNA was precipitated from the 25 aqueous phase by adding ice-cold 95% ethanol, wound on a glass rod, resuspended in 5.0 mM EDTA; 10 mM NaOH and sonicated for 50 seconds at low power in a Branson sonicator. The solution was adjusted to pH 7 and dialyzed overnight against 5.0 mM EDTA in a cold room (0°-5°C). 2.9 Measurement of the Rate and Extent of RNA Synthesis on Rat Liver DNA The rate and extent of RNA synthesis was assessed by a method originally reported by Hyman and Davidson (98) and later fully charac- terized by Cedar (99) and Cedar and Felsenfeld (100). This method efficiently separates the initiation phase of RNA synthesis from the elongation phase. The high salt conditions of the assay appear to be as effective as the drug rifampicin in preventing reinitiation of RNA synthesis (100). Conditions of the assay are such that 1) sufficient time was allowed (15 minutes) for all possible initiation to occur (100), 2) RNA polymerase is present in excess such that all binding sites on the template should be saturated, 3) the high salt concentra- tion during chain propagation should prevent a second polymerase mole- cule from initiating on a site already used for transcription (99), and 4) each polymerase molecule Should be able to initiate only one chain (99). These conditions permit a quantitative estimation of RNA chain length and the number of initiation sites for RNA synthesis. The validity of the Hyman and Davidson method has been verified by others (101-105). E, 3333_RNA polymerase was the enzyme selected to transcribe the various DNA templates isolated from control and AAF-treated animals. There are several advantages to using this enzyme for in 26 y3333_template activity studies; 1) E, 3333 RNA polymerase initiates at one site per 1500 nucleotide pairs on DNA whereas the mammalian enzyme (RNA polymerase II) initiates at one site per 40,000 base pairs (99). Thus, the chances of detecting carcinogen-induced damage would be greater with the bacterial enzyme; 2) E, 3333_RNA polymerase appears to support initiation at every binding site on mammalian templates whereas calf thymus RNA polymerase II can support initiation at only a small number of binding sites (99,100); 3) the bacterial enzyme can be purchased from several manufacturers at a reasonably high purity and can be stored many months without loss of activity (Table 8). Several investigators have presented data suggesting that E, 3333_RNA polymerase does not initiate at the same sites as the mammalian enzyme (99,106). The bacterial polymerase, however, does synthesize hemoglobin messenger RNA on duck reticulocyte chromatin (96) but not from brain (107) or liver (96) chromatin. This suggests that E, 3333_RNA polymerase ex- hibits some degree of specificity as to initiation and termination sites on mammalian templates. Although there may be some advantages to using rat liver RNA polymerases, these initial studies were primarily con- cerned with the presence and recognition of template damage, as opposed to whether the damage is occurring in transcriptionally active regions of the genome. E, 3333_RNA polymerase, 0.25 or 0.35 unitsl, and 8 ng rat liver DNA were incubated for 15 minutes at 37°C in 10 mM Tris-HCl (pH 7.9); 1.0 1One unit of enzyme is that amount catalyzing the incorporation of 1 mole of 3H-UTP into acid insoluble material in the presence of calf thymus DNA in 10 minutes at 37°C. 27 3 mM MnCl 0.08 mM each ATP and GTP; and 0.02 mM H-UTP (500 Ci/mole) 2; in a final volume of 100 pl. During this low-salt incubation period, the enzyme binds to DNA and initiates transcription but chain elongation is blocked due to the absence of CTP (100). Ammonium sulfate, 1.6 M, was added to prevent reinitiation and chain propagation was begun by the addition of MgCl and CTP (final concentrations 5.0 mM and 0.063 mM, 2 respectively)(100). Reactions determining the rate of RNA synthesis were incubated for 10 minutes. Those reactions determining the extent of RNA synthesis were allowed to proceed until the change in reaction rate approached zero (usually 140 minutes) as described in Section 2.10. Reactions were terminated, precipitate washed and radioactivity measured as described in Section 2.5, omitting the ethanol—ether wash. 2.10 Determination of the Number of Initiation Sites for RNA Eynthesis A fixed amount of E, 3333_RNA polymerase was titrated with in- creasing amounts of rat liver DNA to obtain an estimate of the number of initiation sites for RNA synthesis (100). The same procedure and basic reaction mixture described in Section 2.9 was employed using 0.05 units enzyme and 8, 16, 23, 48 or 64 ng DNA. Reactions were stopped after 10 minutes incubation in high salt. Since the conditions of the assay are such that no re-initiation is occurring (101,102), the incorporation of nucleotides after 10 minutes can be used as an indication of the number of initiated RNA chains. As the amount of template is increased, more chains are initiated until a plateau is reached. Since subsequent addition of DNA results in no increase in 3H-UTP incorporation, the end point of the titration represents that amount of DNA which binds all 28 available RNA polymerase molecules. Thus, at the end point, the number of available RNA polymerase molecules is equal to the number of initiations, assuming that each initiation results in a propagating chain (100). 2.11 Calculations and Statistics a. RNA polymerase activity, as measured in these experiments, was expressed as the amount of RNA synthesized on the exogenous template poly (dA-dT). Enzyme activity in hepatic nuclei was calculated from the following equation: RNA Polymerase I Activity = A - B where A = capacity for RNA synthesis with both actinomycin-D and poly (dA-dT) present in the reaction mixture and B = capacity for RNA syn- thesis with actinomycin—D present but poly (dA-dT) absent from the reaction mixture. b. The extent of RNA synthesis is defined as the maximum number of nucleotides incorporated under the reaction conditions described in Section 2.9. It is equal to the incorporation of 3H—UTP x 4 (assuming rat liver DNA contains 25% adenine bases) and was found by fitting a regression line to the data on the plateau region of the mean rate curve for a given treatment group. If the slope was not different from zero (p<.05), all the data on the plateau were combined to yield a mean i S.E. representing the extent of RNA synthesis for a given treatment group. However, the control groups yielded statistically significant slopes. For these groups, analysis by a paired 3 test showed that incorporation of 3H-UTP at 120 minutes was greater than that at 80 29 minutes but equal to that at 140 minutes. Thus, incorporation at 120 minutes was defined as the extent of RNA synthesis for these groups. c. The number of molecules RNA polymerase in the reaction mixture can be calculated using the following formula: molecules RNA polymerase = (U x N)/(S x MW) where U = that number of enzyme units added to the reaction; N = Avo- gadros number 6.02 x (10)23; S = specific activity of RNA polymerasel, 400 units/mg; Mw = molecular weight of E, 3333_RNA polymerase, 475,000 daltons (108). d. The number of initiation sites for RNA synthesis/ng DNA was calculated from the data obtained by titrating the enzyme with template according to the following formula (100): Initiation sites/ng = P/(ng DNA) where P = number of enzyme molecules added to the reaction mixture; (ng DNA) = amount of DNA at the end point of the titration. The end point of the titration was that point at which further addition of DNA re— sulted in no statistically significant increase in 3H-UTP incorporation, as determined by a paired E_analysis (p<.05) on each individual DNA preparation. The number of base pairs/molecule polymerase (Table 10) can be obtained by multiplying (ng DNA)/initiation x (10)12 base pairs/ng DNA. e. The average RNA chain length can be calculated by dividing the extent of RNA synthesis by the total number of initiation sites. Chain length = extent/(Initiations/ngDNA x 8) 1The activity of E, coli RNA polymerase assayed by the method of Cedar and Felsenfeld (100) was one-third of that activity obtained when assayed by the Burgess method (109). Thus, the maximum specific activity of 1200 units/mg as described by Burgess was divided by 3 for the above calculations. 30 where extent = number of nucleotides incorporated (see Section 2.11 b); initiations/ng DNA = number of initiation sites (see Section 2.11 d.); 8‘= ng DNA used to determine extent of RNA synthesis. f. All other statistical analyses utilized the student's Eftest (p<.05) for unpaired observations unless otherwise noted. 2.12 Determination of E. coli RNA Polymerase Activity Using Calf Thymus DNA To examine the possibility of loss of activity occurring during storage of the enzyme, the activity of E, 3333_RNA polymerase was assayed every 8-12 weeks using calf thymus DNA as the template. No loss of activity could be demonstrated for any of the enzyme prepara- tions used in these experiments. The assay for RNA polymerase was that described by Burgess (109), substituting 3H-UTP for 14C—ATP. 2.13 Miscellaneous Methods DNA and protein were determined by the methods of Ceriotti (110) and Lowry 33_33, (111), respectively. For determination of RNA, each DNA sample was incubated at 37°C for one hour in 0.3 M KOH. Cold 5% TCA was added and the solution was centrifuged at 5,000 x g for 10 minutes. The absorbance of the supernatant was monitored at 260 nm in a Gilford UV spectrometer and RNA was calculated using 1 Absorbance unit = 32 ug RNA/ml (112). RESULTS 3.1 Determination of the Capacity for RNA and Poly A Synthesis in H3patic Nuclei The effect of varying the amount of nuclei in the reaction mixture upon RNA and poly A synthesis is shown in Figures 1A and 1B, respec- tively. The data in these figures demonstrates that incorporation of 3H—ATP into RNA and poly A is directly proportional to the amount of nuclei added to the reaction mixture. In addition, the four nucleotide triphosphates were present in excess in the reaction mixture (data not shown). The optimum ionic concentrations for the assessment of RNA syn- thesis by RNA polymerase I were determined by measuring the rate of RNA synthesis while varying the concentration of MgCl2 and (NH4)ZSO4 in the reaction mixture. The data in Figure 2A demonstrates that increasing the concentration of MgCl2 in the reaction mixture from 2.8 mM to 14 mM has little effect upon the rate of RNA synthesis. Addition of up to 20 mM (NH4)2804 also produced essentially no effect upon incorporation of 3H-ATP into RNA, as shown in Figure ZB. The optimal ionic concentrations for assessment of RNA polymerase II were also determined and the data is shown in Figure 3. Increasing the concentration of MnCl2 in the reaction mixture from 1.2 mM to 3 mM increased the incorporation of 3H—ATP after 6 minutes of reaction time. Increasing the concentration of MnCl to 6 mM resulted in no 2 31 3H-ATP INCORPORATION (DPM x 10"?) Figure l: 32 1 16 O~ 1 23,19 N -h ‘1 n O r d O- I ”I. o I l 0 IO 20 TIME (MIN) The Effect of Varying the Amount of Nuclei in the Reaction Mixture Upon Incorporation of 3H-ATP into RNA and Poly A. RNA (A) and poly A (B) synthesis were assessed under conditions optimal for RNA polymerase I as described in Section 2.5. Each point represents the mean of duplicate determinations using a single nuclei preparation. The amount of nuclei added to the reaction mixture was either 15 ul (0), 25 ul (I) or 50 ul (‘) where 50 ul equals approxi- mately 70 ug DNA. 33 all-ATP mconorumon IDPM x 104') I TIME (MIN) Figure 2: Determination of Optimal Ionic Conditions for Assay of RNA Synthesis by RNA Polymerase I in Hepatic Nuclei. RNA synthesis by polymerase I was assayed as described in Section 2.5 except A) MgC12 concentration was varied keeping (NH4)ZSO4 concentra- tion constant at 10 mM or B) (NH4)2804 concentration was varied keeping MgClz concentration constant at 7 mM. Each point represents the mean of duplicate determinations using a single nuclei preparation. Concentra- tions of MgClZ in A were 2.8 mM (0), 7 mM (I) or 14 mM (fi). Concen- trations of (NH4)2804 in B were 4 mM (I), 10 mM (I), 20 mM (‘) or no NH4SOz, (O). 34 I I I I I .03" I i2 Ix I: In I3 '2 9 .— '1 a: 2 «5" O u 12 4. G. h, o '1 I :2- a) 0 I o J . I O 3 5 I TIME (MIN) Figure 3: Determination of Optimal Ionic Conditions for Assay of RNA Synthesis by RNA Polymerase II in Hepatic Nuclei. RNA synthesis by polymerase II was assayed as described in Section 2.5 except A) MnClz concentration was varied keeping (NH4)ZSO4 concen- tration constant at 100 mM or B) (NH4)ZSO4 concentration was varied keeping MnClz concentration constant at 3 mM. Each point represents the mean of duplicate determinations using a single nuclei preparation. Concentrations of MnClz in A were 1.2 mM (‘), 3.0 mM (0) or 6.0 mM (I). Concentrations of (NH3)2804 in B were 4.0 mM (0), 10 mM (I), 20 mM (I) or no (“$2304 (0). 35 further increase in 3H—ATP incorporation. Addition of 40, 100 or 200 mM (NH4)ZSO4 resulted in a progressive increase in 3H-ATP incor— poration. Figures 4A and 4B show the results of an experiment designed to examine the variability of the capacity for RNA synthesis by poly- merases I and II. The data demonstrates that variability in the capacity for RNA synthesis between nuclei preparations obtained from 2 rats is low for both enzymes. Variability between two nuclei suspensions prepared from a single rat is also quite low. RNA and poly A synthesis by polymerase I and polymerase II in nuclei isolated from carcinogen—treated animals is shown in Figure 5. Feeding of the AAF diet for 4, 7 or 14 days resulted in little change in the capacity for RNA synthesis by RNA polymerase I (Figures 5A, 5B and 5C, respectively). However, removal of the carcinogenic diet for 2 days resulted in a 140% and 46% increase in the capacity for RNA synthesis by polymerase I in those animals ingesting the AAF diet for 7 or 14 days, respectively. Five days after returning to the control diet, RNA synthesis by polymerase I returned closer to control levels in nuclei isolated from animals ingesting the AAF diet for 7 days but remained elevated approximately 40% in those animals receiving the AAF diet 14 days. RNA synthesis by polymerase II was inhibited 54% after 4 days, 10% after 7 days and essentially uninhibited after 14 days of AAF ingestion. The inhibition seen after 4 days of AAF feeding returned to control values 2 days after returning to the control diet. 36 —3) 3H—A‘I’P mconporunou (DPM/pg DNA x10 33 I 1i 16- T2- 3 - 4.. °o :I 2. THHE UHIN) Figure 4: Variability of the Capacity for RNA Synthesis Between Nuclei Preparations. Nuclei were prepared from two rats. The nuclei from rat 1 was divided into two parts. RNA synthesis by polymerase I (A) and polymerase II (B) was assayed as described in Section 2.5. Each point represents the mean of duplicate determinations. Rat 1, part 1, (I); Rat 1, part 2 (fi); Rat 2 (I). 37 Figure 5: Effect of AAF Ingestion on RNA Synthesis and Poly A Synthesis by RNA Polymerases I and II in Hepatic Nuclei. RNA and poly A synthesis by RNA polymerase I and II were determined as in Section 2.5 in nuclei isolated from rats fed a 0.05% (w/w) AAF diet for 4 (A), 7 (B) or 14 (C) days. Groups of animals were killed at either the end of the period of carcinogen feeding (zero time) or at 2 and 5 days after being returned to the control diet. Each point repre- sents the mean 3 range of the data obtained from two animals. Control values (mean t SE of the data obtained from 8 animals) were 5,000i333 DPM/min/mg DNA for RNA synthesis by polymerase I (I); 7000i667 DPM/min/mg DNA for RNA synthesis by polymerase II (0); 27003600 DPM/min/mg DNA for poly A synthesis by polymerase I (‘); and 21003200 DPM/min/mg DNA for poly A synthesis by polymerase II ([5). ._.__ Fry—#— *wM—g 38 —I I90 - 150F- _ O m 55on p235: _ om. 73 110-- _ P _ ET 0 O 0 w 9 5 H 7 p — p O O O 5 7 3 3 2 2 .l .I 200 - 160 '- DAYS OFF THE AAF DIET Figure 5 39 However, removal of the carcinogenic diet for 2 days resulted in an 88% and 50% stimulation of RNA synthesis in those animals ingesting the AAF diet for 7 or 14 days, respectively. RNA synthesis by polymerase 11 remained elevated approximately 30% 5 days after returning to the control diet in those animals receiving the AAF diet 7 or 14 days. Poly A synthesis by RNA polymerase II did not always mirror RNA synthesis by this enzyme (Figure 5). Poly A synthesis by polymerase II was increased 62%, 105% and 106% in nucleic isolated from animals ingesting the AAF diet 4, 7 or 14 days, respectively. Poly A synthesis by this enzyme remained elevated above control levels throughout the entire 7 and 14 days series. Figure 6 shows the relationship between duration of AAF ingestion and RNA and poly A synthesis by RNA polymerases I and II. The data for 4, 7 and 14 days has been taken from Figure 5. The data in Figure 6 suggests that RNA and poly A synthesis by both polymerases reach a peak stimulation after 8 weeks of AAF feeding. An additional 4 weeks of carcinogen feeding results in all parameters except poly A synthesis by polymerase II returning close to control values. 3.2 Determination of the Capacity for RNA Synthesis and RNA Polymerase I Activity in Carcinogen-treated Animals Figures 7A and 7B show the effect of actinomycin-D on the initial rate of RNA synthesis by RNA polymerases I and II, respectively. Controls included addition of 20 ul of glass distilled water or 95% ethanol, the latter being the solvent used to suspend the actinomycin D. The data in Figure 7A suggests that after 6 minutes of reaction time, 40 .450 I .400 I 5 350A ’ e I a: I 5 300 r o I" u n- 250- :z 3 A a: 200- ‘ III I. A 150- . \\ ‘ c 100 /.d I C 50— L 1 I n I l n 00 I 2 8 I2 I I wears on me AAF om Figure 6: Effect of Duration of AAF Ingestion on RNA Synthesis and Poly A Synthesis by RNA Polymerases I and II in Hepatic Nuclei. RNA and poly A synthesis by polymerases I and II were determined as in Section 2.5 in nuclei isolated from rats fed a 0.05% (w/w) AAF diet. Each point represents the mean 3 range of the data obtained from two animals. Control values and symbols are as listed in the legend to Figure 5. 41 IO- -2) (om x IO INCORPORATION 3H-ATP 0‘ l on TIME (MIN) Figure 7: Effect of Actinomycin-D on RNA Synthesis by RNA Polymerases I and II in Hepatic Nuclei. Assay conditions were optimal for either polymerase I (A) or poly- merase II (B) as described in Section 2.5. The amount of actinomycin-D added to the reaction mixture was either 2 ug (‘); 4 ug (fl); 8 ug (I); no actinomycin-D, 20 ul water (0); no actinomycin-D, 20 pl 95% ethanol (0). Actinomycin-D was suspended in 95% ethanol. Each point represents the mean of duplicate determinations using a single nuclei preparation. 42 95% alcohol alone can markedly inhibit RNA synthesis by polymerase I. Doubling the amount of actinomycin-D in the reaction mixture from 4 to 8 ug produced little effect upon RNA synthesis by polymerase I after 6 minutes of reaction time. As shown in Figure 7B, 95% ethanol also appears to have an inhibitory effect on RNA synthesis by polymerase II. The greatest inhibition of the rate of RNA synthesis by polymerase II was achieved by using 8 ug actinomycin-D in the reaction mixture. Additional experiments on the effects of actinomycin-D and 95% ethanol on the initial rate of RNA synthesis by polymerases I and II produced the data shown in Figures 8A and 8B, respectively. In these experiments the stock solution of actinomycin D was concentrated to allow the addition of 2 ul of solvent (water-95% ethanol, 1:1 v/v) as opposed to the 20 ul of 95% ethanol added in the experiments shown in Figure 7. Also examined in these experiments was the effect of omitting CTP and GTP from the reaction mixture. The data for polymerase I, shown in Figure 8A, suggest 1) addition of 2 ul solvent (1 ul ethanol) can inhibit the rate of RNA synthesis after 6 minutes of reaction time, 2) addition of actinomycin-D can inhibit the rate of RNA synthesis greater than solvent alone and, 3) removal of two of the four nucleotide tri- phosphates can significantly lower RNA synthesis by polymerase I. In contrast to the results with polymerase I, the data for polymerase 11 (Figure 8B) suggest 1) addition of neither solvent alone nor actino- mycin-D markedly alters the rate of RNA synthesis, and 2) removal of CTP and CTP did not decrease incorporation of 3H—ATP. Figure 9 shows the time course of the reaction for RNA synthesis in the presence and absence of poly (dA—dT) for polymerases I and II. All 43 O l .5 CD —I .5 I d N I 3H-ATP INCORPORATION (DPMxIO-z) 6 TIME (MIN) Figure 8: Effects of Ethanol and Actinomycin D on RNA Synthesis by RNA Polymerases I and II in Hepatic Nuclei. Assay conditions were optimal for either RNA polymerase I (A) or polymerase II (B) as described in Section 2.5. Each point represents the mean of duplicate determinations using a single nuclei preparation. Actinomycin-D was suspended in water-95% ethanol (1:1 v/v). (O) 4 ug actinomycin-D; (I) 4 ug Act-D, omitting GTP and CTP from reaction mixture; (I) No actinomycin-D, omitting GTP and CTP from reaction mixture; (A) No actinomycin-D, 2 ul water-95% ethanol (1:1 v/v), omitting GTP and CTP from reaction mixture. 44 N O N N d 0 I4 10 3H-ATP INCORPORATION (mm x 10'2) 2 I 0O 5 'IO 20 30 I TIME (MIN) Figure 9: RNA Synthesis in the Presence and Absence of Poly (dA-dT). Assay conditions for RNA synthesis by RNA polymerase I and polymerase II were as described in Section 2.5. Each point represents the mean of duplicate determinations using a single nuclei preparation. (H) Polymerase I, 8 ug poly (dA-dT); (H) polymerase II, 8 ug poly (dA-dT); (.---«.)polymerase I, no poly (dA-dT); (km-i) polymerase II, no poly (dA-dT). 45 reactions were linear for at least 10 minutes. RNA synthesis by both enzymes was greater in the presence of the synthetic template than in its absence. RNA polymerase I activity (defined in Section 2.11) was greater than polymerase II activity at all time points measured. Table #2 shows the effect of varying the concentration of nuclei in the reaction mixture upon the rate of RNA synthesis in the presence and absence of poly (dA-dT). The data suggest that the optimum amount of nuclei for assay of polymerase I is 25 ul (approximately 35 mg DNA). RNA polymerase II activity could not be detected with poly (dA—dT), despite the addition of up to 75 ul nuclei. The above experiment was repeated for polymerase I only and the results are shown in Table #3. This experiment confirmed 20-25 ul nuclei as the optimum amount of nuclei to be added to the assay mixture. The effect of AAF ingestion on RNA synthesis and RNA polymerase I activity is shown in Figure 10. No statistically significant changes in the capacity for RNA synthesis could be demonstrated in nuclei prepared from rats maintained on the AAF diet for 4 or 7 days. It was elevated approximately 40% in nuclei prepared from animals fed the AAF diet 14 days and returned to control level 5 days after returning to the control diet. At the end of 4, 7 and 14 days of carcinogen feeding, RNA poly— merase I activity was significantly (p<.05) increased to 145, 155 and 230% of control values, respectively. The stability of the carcinogen- induced changes in polymerase activity was variable, depending upon the duration of AAF feeding. Five days after being switched to the control diet, RNA polymerase I activity returned to control levels in those animals fed the AAF diet for 4 or 14 days. However, polymerase activity 46 .mouUGHE OH woumnduca mumB mGOHuommu HH< mafis\zmm m .HH.N soauoom ca mm vmcflwonq .AHwIfluo< mommuwskaom mAHwIfiuo< ommumE%Hom mo uommwm N mqmH=o comm .mumv m>onm mzu aoum monomuunam cow: was o=Hm> mwsu can 2mm com cmnu mmma mmB uamm 30H 0H mHDImm mo coaumuoauoocH .Hmaosz HHQU ANIzv HMShSUGoummIcoz was AHIZV Hmahnuaoumm Baum vmumaomH