Tobacco Documents Online

I ( ( V ( ( , 7 ( ( '1'1n: COUxc11. T'c,:e 'I'e11sAcco I2>tal:Auci1-U.ti..1.. Irc. 11U F:Ai:T 51si1t s9TRF'.M.T \F:N' YORK. \. 1'. 10022 (1., 121 4 t1.rx%. M:. Applicotion for Research Grant (Use extra pages as needed) 1. Prinrpal Investiqotor (give title and degrees): Carol J. Henry, Ph.D. and Richard E. Kour:, Ph.D. Director, Department of Director, Department of Experimental Oncology ' Biochemical Oncology 2. Insriwtion & address: Microbiological Associates 5221 River Road Bethesda, Maryland 20016 3. Oeportmenr(s) where reseerch will be done or collaboration provided: Department of Experimental Oncology Department of Biochemical Oncology SSP i g 1978 G- ~ - 6'.,1 4. Short ude o( s+udy: Evaluation and Characterization of an Alkaline Elution Assay as a Measure of Pulmonary DNA Damage Induced by Chemical Carcinogens or the Chemicals in Cigarette Smoke. S. Proposed storrinp dote: November 1, 1978 6. Estimored time ro complete: Three years 7. 8rief description of specific research oims: Exposure to many carcinogens in vivo has been shown to result in damage to DNA which can be detecte3 as single strand breaks using an alkaline elution (AE) assay. We propose to characterize the AE assay simultaneously using two specific approaches: 1) as a tool in predicting lung tumorigenic capacity of chemical carcinogens, and 2) as a tool in predicting the capacity of the chemicals In cigarette smoke to interact with lung DNA in vivo. Using our model systems for lung carcinogenesis and cigarette smoke inhalation, we intend to address the foliowing questions: 1. Can damage to lung DNA be detected after intratracheal (IT) treatment with chemicals under conditions where these chemicals inducellung carcinomas? 2. For a particular-class of chemical carcinogens, Is there a correlation between conditions that yield high inci- dences of lung carcinomas and amount of DNA damage? 3. Is the amount of lung DNA damage related to the capacity of lung tissue to metabolize these chemicals? -1- ( 50137808 Dote: 9/ 11/78 EXHIBR NO.-I-L S. NELSON I C`'~' ~`~~~ ~~4;~ ;~"`~~

A Continuation of Brief Description of Specific Research Atms: 4. Does exposure to whole cigarette smoke cause damage to DNA as detected by the AE assay? 5. Does exposure to the chemicals in smoke as repre- sented by the cigarette smoke condensate and con- densate fractions cause damage to DNA as detected by the AE assay? . 6. Does exposure to whole cigarette smoke alter the DNA damage lnduced by other chemicals? ABBREVIATIONS ) • AE alkaline elution AHH aryl hydrocarbon hydroxylase ASG alkaline=sucrose gradients BaP benzo(a)pyrene C3 C3H/Anf mouse strain CSC cigarette smoke condensate CSCF cigarette smoke condensate fractions D2 DBA/2 mouse strain DEN diethylnitrosamine DMH 1,2-dimethylhydrazine DMN dimethylnitrosamine DMNA N-nitrosodimethylamtne EMS ethyl methylsulfonate ENU N-ethyl-N-nitrosourea MFO mixed function oxidase MMS methyl methanesulfonate MNNG N-methyl-N'-nitro-N-nitrosograndine MNU N-methyl-N-nitrosourea 4-NQO 4-nitroquinoline oxide PAH polycyclic aromatic h ydrocarbons RCM reconstituted material RNA ribonucleic acid TCDD 2,3,7,8-tetrachlordibenzo(p)dioxln TPM total particulate matter -2- 50137809 ~~~ ~ ~~ CTR HN

L. ( _ 8. Brief statement of 'working hypolhcsis: Our working hypothesis Is that chemical carcinogens or the metabo- lites of the carcinogens have the potential to Interact with DNA and that this interaction or DNA damage is one of the early steps In the carcinogenic process. This DNA damage can be quantitated in vivo by a simple, sensitive, short-term, and inexpenstve techni uqe w~ ch measures the rate at which this damaged DNA elutes from a filter under alkaline conditions. In this way, a method that has the capacity to quantitate one of the earliest presumed steps In the initiatlon of carcinogenesis can be s ecificall used to address the question of defining those classes o c em ca s and those in vivo conditions responsible for determining the susceptibility _oT 'iung tissue to chemical carcinogenesis. . ( ( 9. Deio11s of experimenrol design and procedures (append extra pages as necessary) I. Background Many chemicals tho.ght to be toxins, mutagens, or carcinogens are believed to exert their effects through interactions with DNA. This interaction can be a covalent attachment of the chemital to a s peci- fic nucleotide, an Intercalation between bases which resultsiin helix distortion, a cross-linking between nucleotides on opposing DNA strands - or phosphotrlester formation (see reviews 1, 2, 3). Such damage can ' cause localized perturbations In the DNA structure and either directly lead to single-stranded breaks, or indirectly result In single stranded breaks formed because of the action of endonucleases (part of the normal repair processes). The amount of damage can be quan- titated by measuring the size of the DNA strands after exposure to alkaline conditions (>pH 12.0). Exposure to alkali results in the unwinding of the double strands of DNA and the size of these strands can be directly quantitated by either alkaline-sucrose gradient (ASG) centrifugation (4) or alkaline elutlon (AE) (5, 6, 7). The ASG method is tedious, time-consuming, limited in the number of individual assays that can be performed, and sometimes suffers from poor recovery 8f labeled DNA. Moreover, single strand sizes of up to only 2-5 x 10 daltons can be measured (8). Above this size, no generally applicable method has been available. The AE assay,on the other hand, provides size measurements of extremely long DNA single strands (7). This assay has been used to study DNA replica- tion (9), DNA repair followin X-ray exposure (10), DNA scission In cells treated with bleomycin ?il), and DNA cross-linking in cells treated with nitrogen mustard or with chloroethylnitrosoureas (12). The AE method is based on the very simp le Idea of re~ainin cells in which the DNA has been prelabeled with 3H-Thymidine ( H-TdR~ on poly (vinylchloride) filters, lysing the cells on the filters, and eluting the DNA from the filters in an alkaline buffer (pH 12.2). • The elution rate seems to be governed by the properties of the DNA itself because:(7) -31 50137810 ~' ~~' ~'"I~>~ 04,23724

it ge l, Continutation of Details of Experimental Design and Procedures: a) almost all non-DNA material 1s removed by the lysis solution Including protein, lipids and memb ranes, b) removal of protein with proteinase-K and RNA by ribonuclease does not alter the elutlon kinetics, therefore,proteln and RNA play no role In the DNA elution, c) the rate of DNA elution is independent of the number of cell~ on the filter in the range of 0.1 - 2.0 x 10 cells; therefore, retention on the filter does not depend on the aggregation or trapping of DNA strands in a large-scale d) conditions that cause DNA shearing result In an immediate and striking Increase in DNA elution, 0 e) retention of DNA Is independent of the filter material, also essentlally independent of pore size from 1-5 µm, f) DNA under alkaline conditions is susceptible to' strand scission by visible light (13), and the elution of DNA is markedly sensitive to exposure to room light. The limit of sensitivity of the AE assay has been estimated following exposure to low-level X-rays. Th int ajeil lar NA siieet~)are at least 1.3 x 107 nucieotides ~4 x~0 da~tonsg in Current speculation is that the rate-limiting process Involves the search for a molecular conformation that will allow a long DNA single strand to be pulled through a single filter pore. Since the strand dimensions are much larger than the filter pores, portions of a single strand may be pulled through many pores at the same time. Elution must require that the strand segment In one pore acquire enough dominance to pull the entlre strand th rough . The AE assay can be streamlined Into a rap id and inexpensive method to test the capacity of unknown chemicals to Induce damage to DNA (5,6). Moreover, the AE assay can be used In the Presence of exogenous metabolic activation systems (6). Assay of 46 differ- ent chemicals(with and without metabolic activation) Indicate that every known carcinogen causes DNA damage (ncluding such weakly active chemicals as safrole, DDT, dieldrin and cyclophosphamide, while all non-carcinogens yield no DNA damage (5). Recent studies In our • laboratory suggest that the amount of damage to DNA Imparted by benzo(a)pyrene (BaP) is directly related to the aryl hydrocarbon hydroxylase (AHH) levels in tissue hemogenates from lung or liver, -4- 50137811 CTR MN 043`2`2215

c r r r Continuation of Details of Experimental Design and Procedures: ~ and not the level of epoxide hydrase (14). Thus, at least In 1n vitro systems, the amount of BaP metabolized to forms that cause YNA aamage, parallels quite nicely the amount of BaP metabolized to 3-OH BaP (i.e., AHH activity). The AE assay can also be used to quantitate the level of DNA damage following In vivo exposure of chemical carcinogens (5,15). This DNA ls prelabel'ey 3H-TdR and•the DNA detected by radio- metric means, or directly measured by a fluorometric method (15). DNA damage can be assessed in liver, lung, thymus, brain, kidney, stomach, duodenum, colon, bone marrow and mammary gland (5,15). Most direct acting carcinogens (e.g. N-methyl-N'-nitro-N-nitroso- guanidine (MNNG), N-methyl-N-nitrosourea, N-ethyl-N-nitrosourea, methyl methanesulfonate, and ethyl methanesulfonate) caused sig- nificant DNA damage in target and non-target organs, but ten other known chemical carcinogens induced the greatest DNA damage In the target organs for carcinogeniclty (5). In fact, dimethylhydrazine, a potent colon carcinogen, causes dose-dependent DNA damage In colonic tissue, and strains of mice that are susceptible to di- methylhydrazine-induced colonic cancers are more sensitive to DNA damage in this tissue, than are strains that are resistant to the tumorigenic effect of this chemical (15)'. Since there Is adequate evidence that methods for evaluating ' the capacity of a compound to damage DNA and to Initiate DNA repair may be utilized as a rapid prescreening test for assessing potential carcinogenic activity (5,15), this evidence coupled with the relative ease and sensitivity of the AE assay suggests a rather straight forward, sensitive, short-term bioassay that should be able to quantitatively evaluate the kind of chemicals and the In vivo conditions necessary for optimal induction of lung cancer.-T~F e- animal model developed In our laboratories for the defined chemical induction of lung cancer ts described below. The use of this model in cigarette smoke lnhalation studies is also described. Both the chemical induction of lung tumors and bioevaluation of cigarette smoke require long-term studies (one year or longer) before an effect can be observed or the lack of an effect can be determined to besignificant. Hence, the desirability of the AE assay as a sensl- tive, short-term test which would show direction In the chronic, long-term In vivo studies. Doses of 250 µg MCA given intratracheally one, two, four and six times to C3H/Anf Cum mice yielded approximately 35, 45, 76 and 77% lung carcinomas, respectively, within one year after treatment (Table i). Doses of 1.2 mg BaP given Intratracheally five, ten and fifteen times to C3H/Anf mice yielded approximately 19, 15 and 24% tumor Incidences within one year after treatment. The derivative, 7,8-dlhydro dth ydroxybenzo(a)pyrene (BaP-7,8 diol), is much more tumort genic to lung tissue than is the parent chemical, BaP. Data in Table 1 show that IT administration of 250 µg BaP-7,8-dlol given • five times at biweekly intervals results In approximately 40-50% lung carcinomas after about one year after treatment. The tumors -5- I 50137812 ~ r TCTR ' MN 0437,26-)

I Continuatlon of Details of Experimental Design and Procedures: observed In these experiments Include bronchogenic squamous cell c arcinoma, squamous neoplasm, alveolar adenocarcinoma and alveolar adeno-squamous carcinoma. These malignant lesions have been defined and characterized In the mouse and are representative of the major classes of the lung cancers found In man. The characterization of these lesions In our model animal systems has included a descrip- tion of the pathogenesis of these lesions; that is, a documentation of the steps Involved In the progress'ton or formation of the tumors from specific preneopl-asttc lesions (Table 2). This model system has also been characterized for levels of aryl hydrocarbon hydrox- ylase (AHH) In the pulmonary tissues of inbred strains of mice after IT inoculation of chemical carcinogens (16), and the susceptibil- ity of lung tissue to cancers induced by these chemicals is geneti- c ally linked to the capacity of lung tissue to metabolize these chemicals (see Appendlx A). Thus, we have an In vivo model system for lung cancer that is sensitive, well-characterTzed, analogous to the human situation, and specifically dependent on the capacity of the lung tissue itself for the bioactivation of these chemicals to forms carcinogenic to•the lung. It is interesttng to note that an Inhibitor of MFO activlty dl- sulfiram, has been shown to alter the amount of hepatic DNA damage induced In rats by diethylnitrosamine (DEN) (23). In addition, multi- ple feeding of disulfiram has also been shown to alter and delay the onset of hepatocarclnogenesls induced by DEN (24). Thus a known ln- • hlbttor of MFO activity Is capable of altering the amount of DNA damage Induced by a chemical requiring these enzymes for bioactlvations. This murine lung model s ystem has been utilized In our labora- tories for evaluating the biological activity of whole cigarette smoke In vivo. Cigarette smoke has been characterized as a weak carcinogen orto-carclnogen and hypothetically can exert Its effects at various stages during the process of tumor formation and ex- pression (see discussion, 25). Cigarette smoke contains between 5,000 and 10,000 chemicals (17), however, only a few have been characterized with regard to concentration In smoke and potential biological activity. Experiments evaluating the carcinogenic potential of such a complex chemical mixture as cl garette smoke In animal models must satisfy several specific criteria matnly because cigarette smoke possesses, at best, very weak biological activity. Thus, conditions must be found that mimic those that exist for man; thatis, exposure for long periods of time (in humans 30-40 years) at fairiy hl gh levels (3-4 packs/day), and then oniy 0.5-2% of the subjects will be observed to have smoke-associated lung cancers. The smoke exposure conditions presently in use In our labora- tory for the inbred strains of mice have been developed to assure maximal de posltion of total particulate matter (TPM) In the lung. Fresh, whole cigarette smoke ls generated using the Smoke Exposure Machine (SEM 11, Process and Instruments, Brooklyn, N1(.), delivered • as a smoke aerosol (0.3-0.5 µm mass median diameter) to the animal containment unit (Figure I). The mice are restrained In stock- type head restraint trays (5 mlce/tray) which are placed on the -6- 50137813 I I C T R 111N 0 4 3 1-"2 -7

: 1 > Continuation of Details of Experimental Design and Procedures: ~ supports of the polycarbonate module In the animal containment unit for smoke exposure. Smoke Is delivered through a channel In the polycarbonate module,into which the noses of the mice protrude. A soft rubber dam material provides a seal around the nose to prevent smoke leakage. Mice are obligate nose breathers and this exposure system, therefore, forces the mice to breathe the air or smoke provided In the channel (see Figure 1). The deposition and distribution of TPM In SC3F1lCum mice after exposure to cigarettes In the SEM 11 system has been characterized and representative data are shown In Table 3 and Figure 2. Such dosimetry experiments utilize a radioactive tracer 1n the cigarette smoke to quantitate the TPM distributed through- out the body of the mouse after smoke exposure. The SEM 11 allows the smoke concentration and exposure time to be varied and as the results Indicate, a dose response is obtained for TPM In the lung with Increasing smoke concentration or Increasing smoke exposure time. Elghty-90% of the TPM was found In the total respiratory tract of the mouse (head, larynx and lung) and 60-80% of the TPM was found 1n the lung itself. The deposition In the lung represents 80-90 µg TPM per cigarette at the lon ger exposure times (200-300 seconds total). Mice can tolerate this level of exposure fairly well and doses of 10 cigarettes per day (Kentucky reference 2A1 cigarettes) have routinely been used for periods of longer than ~ one year. This dose of 10 ctgarettes per day results In almost I mg TPM/day mouse lung or 50 mg TPM/kg body weight, which Is equivaient to a human smoker exposed to the smoke from 3-4 packs of cigarettes per day (18). The capacity of clgarette smoke to induce pulmonary DNA d amage requires several approaches In order to be bhoroughly evaluated. The question addressed Is whether there are chemicals in cigarette smoke which Interact with DNA and whether this lnter- action can be detected as damage by the AE assay. The most direct approach Is to expose mice to known quantlties of smoke and monitor the DNA damage at various times after the exposure. The use of whole smoke alone, however, precludes ascribing ~ the effect (i.e., DNA damage) to specific chemicals or even a class of chemicals. Condensates of whole cigarette smoke (20) and frac- tionation of those condensates (21) represent one of the only logical methods for breaking down the complex chemical mixtures found in smoke into some workable number of subcomplexes. The procedures for making clgare.tte smoke condensates (CSC) and the resulting fractions (CSCFs) have been previously discussed (20, ~ 21,22) and are only briefly described here. Cigarettes are smoked i on an analytical smoking machine using a two-second puff of 35 ml volume, once a minute to a butt iength of not less than 20 mm. The smoke is condensed In traps cooled in a dry-ice acetone mixture and the condensate removed from the traps with acetone. The • solvent is removed In vacuo at a low temperature and the conden- sate stored under nlcrogen at -20°C or lower until fractionation. - 7- ( 50137814 I CTR HN 043 '.1,26

~ Continuation of Details of Experimental Design and Procedures: The fractionation scheme is given In Table 6. Twelve fractions are obtained. A reconstituted sample was prepared by pooling one-third portions of each fraction. This reconstituted material (RCM) is needed to control for possiblE. alterations in chemical composition during the fractionation procedure. The need for standardization of procedures and establishment of quality stand- ards in this area has been recognized (20,22). The twelve con- densate fractions and reconstituted material from several research cigarettes are available to us for evaluation as potential Inducers of pulmonary DNA damage (Table 8). As discussed previously, compounds which can alter MF0 activ- ity can alter lung cancer susceptibility and DNA damage caused by other chemicals as measured by the AE assay (23). Cigarette smoke and CSC contain chemicals which lnduce pulmonary AHH and theoret- ically could alter the lung cancer susceptibility and amount of DNA damage induced by other chemical carcinogens. The effect of whole cigarette smoke In altering the amount of DNA damage caused by other chemicals may require long-term studies in ll yht of recent information from our laboratories. In collaboration with Dr. R. Rasmussen, we have observed that whole cigarette smoke is capable of lnhibiting DNA repair capacity in an in vitro assay (Table 7), however, the effect Is not observed unttl-after 2-3 months of con- tinuous exposure (Table 7). Even though DNA repair is inhlbited , in this assay, the role of this lnhibition on the DNA damage or ' persistence of this damage Induced by other chemicals is diffi- culttio predict. There could be an enhancement of the effect, tnhtbition or no effect on DNA damage.lnduced ]M vivo. Special Considerations The Departments of Experimental and Biochemical Oncology possess special expertise required for: 1. Short-term and long-term animal handling under conditions where such factors as adventitious agents, parasites, and housing conditions that may alter the carcinogen metabolizing capacity of the Inbred strains of mice are well controlled. 2. On-hand and in-depth experience in the model systems for inducing lung cancers in the Inbred strains of mice. 3. On-hand and In-depth experience In exposing the in- bred strains of mice to varying doses of whole cigarette smoke under a variety of exposure conditions. 4. More than a year.of experience in analyzing the amount of DNA damage produced In mammalian cells using the alkaline elution assay. This experience includes on-slte visits by the departmental staff to the laboratories of Drs. J. Swenberg and G. Petzold of the Upjohn Corporation to gain first-hand experience at the nuances of this assay • precedure. -8- 50137815 CT1 ! I .• I 04' 1.W I~ E-1^

Continuation of Details of Experimental Design and Procedures: II. Experimental Approach ( ( i 0 A. Evaluation of AE assay as a tool In predicting lung tumorigenic capaclty of chemical carcinogens. 1. Can damage to lung DNA be detected after intratracheal (IT) treatment with chemicals under conditions where these chemicals lnduce carcinomas? Intraperitoneal trea'tment with such chemicals as dimethylnitrosamine, 4-nltroqulnoline-N-oxide (4-NQO), 1,12-dimethylbenz(a)anthracene, ethylnitrosourea, methyl methanesulfonate (5) and subcutaneous or oral treatment with dimethylhydrazine (15) all cause damage to lung tissue DNA as measured by the AE assay. The major question 1s, can the AE assay detect DNA damage In lung tissue under conditions where a significant incidence-of malignant lung carcinomas will eventuate? The question will be addressed by measuring the amount of DNA damage in lung tissue using the AE assay at 0.5, 1.0, 2.0, 4.0, 8.0, 24 and 48 hours after IT instillat-ton of MCA, BaP, and BaP-7,8 diol to female SC3F1 mice. The procedure for the AE assay ls given in Table 4. The negative control will be the 0.2% gelatin- saline vehicle and the positive control will be 2.0 mg 4-NQO given IT to these mice. Recent studies In our labo- ratory have indicated that an In vitro positive control _ should also be Incorporated Into tFiTs test to insure that ' the buffers, filters, lysing solution etc. are functional. This control will be MNNG-treated V-79 cells (treated In culture) that are stored frozen at -120°C until assayed. The total DNA eluted from the filter will be determtned simultaneously method (the animals will be IP treated as neonates with a total of 150 pCI 3H-TdR over a 3 week period of time and a fluorometric method (15,19). This latter technique, although a little less sensitive, has the advantage of not requiring prelabelling of the DNA. This is especially Important when analyzing for DNA damage after multiple treatments with these carcinogens, because the In vivo labeling Index of lung tissue has a halflife of abou`=-3 weeks and thus by 6-9 weeks the ability to detect DNA damage may be severely impaired because of an Inability to detect the DNA which may elute f rom the filters. 2. For a particular class of chemical carcinogens, is there a correlation between conditions that yield high Incidence of lung carcinomas and amount of DNA damage? The major factors that Influence lung cancer susceptibility In BC3F1 mice are type of chemical carcinogen and dose of carcinogen. The data from section 1 should determine those conditions whereby MCA, BaP, or BaP-7,8 dlol _ Induce lung DNA damage. The second approach will be to use _s various doses of MCA (10-500 jig) and BaP (600-1800 µg) and various times of treatment (1-6 times) in order to observe -9- 50137816 ~ CTR PIN 043730

, 0 L• Continuation of Details of Experimental Design and Procedures: if the extent of DNA damage parallels the ultimate suscepti- bility of that organ to cFiemical carcinogenesis after a specific dose and treatment. Thus, we propose to evaluate the extent of DNA damage at only 2-3 times after multiple IT treatments with these chemlcals (e.g. 2, 4 and 6 hours, the exact time will be determined by the results of thE study in section 1). 3. Is the amount of lung DNA damage related to the capacity of lung tissue to metabolize these chemicals? The answer to this question also provides support for the question addressed in section ii. That is, strains of mice are variable in their susceptibility to lung cancer induced by MCA and this variability at least partly results from the relative levels of those enzymes capable of metabo- lizing this chemical carcinogen (see Appendex A). Therefore, by measuring the amount of DNA damage in various strains of mice with known AHH levels following IT treatment with MCA, we can determine not only if the levels of AHH are playing an'important role, but also if this damage parallels the ultimate susceptibility of these strains to MCA-induced lung cancers. We propose to answer this question by measuring the amount of DNA damage in lung tissue of B6D2F1 x D2 mice following IT treatment with MCA. These progeny animals represent an ideal model system, because the capacity to recognize MCA and metabolize this chemical to a variety of biologically active forms segregates as a sin e autos.omal dominant gene in this cross (see Appendix B for discus3~ on). Thus 50% of the progeny animals from this cross will be capable of responding to the IT-instilled MCA by Increased )evels of pulmonary AHH, and 50% will not. The genetic linkage that may exist In lung tissue between AHH responsiveness, cancer susceptibility, and degree of DNA damage can, therefore, be unequivocably obtained. Approximately only 20 B6D2F1 x D2 animals would have to be analyzed in order to arrive at this conclusion. A simple method of confirming the relationship between AHH responsiveness and amount of pulmonary DNA damage ls based on the observation that there exists a chemical that can specifically alter the AHH capacity of the genetically non-responsive strains of mice. That is, 2,3,7,8-tetrachlorodibenzo(p)dioxin (TCDD) treatment of DBA/2 mice will result in increased AHH levsls. Thus, TCDD-treated D2 mice have AHH capacities very similar to that of the responsive B6 strain. TCDD, under conditions where it induces AHH in D2 mice, selectively enhances the tumor susceptibil ity of this strain (see Appendix ). The most direct test for establishing a relationship between AHH levels and amount of DNA damage in pulmonary tissue is to compare the level and kinetics of DNA damage in control MCA-pretreated, and TCDD-treated D2 strains of mice. -10- 50137817 G `"~~' HN 4~.0 4 3 ,~~'_'4p .11, i