Tobacco Documents Online

rcinogens, Cocarcinogiens, and Tumor linhibitors in Ciigiarette Smoke Condensafe' : _• BENJAMIN L VAN'DUUREh! Laboratory of Organic Chemistry andlCarr°nogenesis Institute of Environmental Medicine New York Uttiversity Medical Center New York; New York 101016 -- 'I1te frst report on the csucinogenicity of cigarette smoke condensate (CSC) on mouse skin fWyndec et a1. 1953) prompted chemical studies on thrc:ucino;,_ens responsible for the observed results: In several' reports (Van Duuren 1!958a. 1958b;, Van Duuren et al. 1960). the isolation. characterizauon. and' quamtdta- tion of a series of aromatic hydrocarbons and heterocyclics were described. tituny subsequent reports confirmed' and extended these findings (Wynder and' Hoffmann 1967). Some of these compounds and their coneentnuons in CSC. ane listed in Table L since they are relevant to the discussion to follow. Thee concentrations listed are from a compilation, by Wynder and Hbffmann ( 1'967). From our own studies on the carcinogrniciiy of CSC tOt,ris et al. 1i958x, we concluded at that time that other factors (i.e.. promoters an&or cocarcinogensp must aiso be present in CSC to account for the observed mouse skin1 carcino- genicity: These biologically active cofactors are the subject of this discussion dealing lurgely, with work from our laboratory. INi'f7Y1TORS AND PRC'VfOTEFfS IN TOBACCO CARCINOGIENESIS' An early report by Boutw• Il and Bbsch (1959) ddiscussed the tumor•promoting ability of a series of phenols, In these two-stage carcinogenesis experiments, a single dorsal mouse skin application of a subcsrcinogenic dose of a carcinogen such as 7,12-dimethwibenz(ulanthracene (D11BA) is followed' by repe•rtedd applications of the promoter at the same site. This defimibiom is :iven here to distinguish tumor promoters from cocarcinogens, which will be discussed later. From the work of Boutwell and Bosch. cited above, it was clear that very few phenols were tumor promoters. Phenol itself was a weuk tumor promoter in their work, and we cont0rmed this in subsequent work (Van Duureni et al.. 1968). However., in the same study, we found' that a whole senes of tobacco phenols and related compounds were devoid of tumor-promoting activity. Yevenheiess. CSC whcn tested ss a promoter in the two-stage prutocol de- scribed abovt did show notable activity. lm a group of 60 I1CRJHa Swiss mice 105 a

106 /'B.L Vnn Duuren Tabre1 Concentration~of'Some Carcinogenic Aromatic: Hydrocarbons and Heterocycfias iniCSC . jug in CSC from Compound I 100 cigarettes ~ Carcinogens Betuo(a lpyrene 2:S DibenY[u.1i lattthr.acene 0.4 Benzo(jlfluornnthene 0.61 Dibenzo(a:i IPyrene Dibenz(u,h laeridine ttaces 0.01 Diben2[ur.jhcridine 0.27 7-1$'.Dibenzo(r;X,l0srbazole 0.07 NoncarcinoQens Pyrene ~ 7.S-10.3 Fluoranthene: 7:5-10.3 Benzo(q:h.f lperylenc 0,6 treated once only with 50 gg DMBaA followed by repeated application of CSC. (40 i mg in 0:1 ml acetone rave tiines weekly). 14 animals developed. squamous carcinoma of the skin. The expenment was terminated after 573 ' days. In the same series of expenments: CSC alone, at the same dou, resulted in onlv four animals with skin cancer. The dosage of 50 /xg D'h181A alone did not result in any skin cancers (Van, Duuren et al. 1971). It must then be concluded that the mixture of many chemirals lutown: as CSC is, indt;ed', not: only a carcinotenbut also, a tumor promoter. We next examined the tumor-initiating activity of some noncarcinogenic or weakly carcinogenic aromatic hydrocarbons. which are CSC components. This experiment showed that dibenz(a,c lanthraccne, chrysene, benz(rr lanthracene. etc., are tumor initiators in two•stage carcinogGnesis using a mixture of phorbol' esters prepared from croton oill as the promoter (Van Duuren and Orris t'9b5: Van Duuremt et al l I970). With all these undines and those of other resesrchers in, hand4 we were still not satisfied that we could account for the observed animal cat~cinogenicitv of CSC. In addition. numerous early epidemiologic studies (Public Health Service 1964) llinking cigarette smoking, an& human Iung, c•.utcer sugoested that further animal studies were needed to pinpoint other cotactors.. COCARCINOGENIC AGENTS Alniobvious approach was to test' CSC Components for cocarcinogenesis in the clnssical Bcrcnblum protocol, (13i:rcnbiurn 1941). In Bcrenbllam's experiments erotnn nil -+ ti•-t;• I••••rrne i fqGr IIP1. when annlier,i .,~„...+.•.....~...•i.. .„, _. .. .

0 either agent'applied alone: Thus. this type of test is operationally different from the two-stace carcino-enesis experiments described above and, moreover. it is the way in; whi& humans are exposed to cigarette smoke. We. therefore. undertook a series of experiments on cocarcinogenesis of' CSC components with quite remarkable results. Five major compound types were tested for cocarcino¢enicity on mouse skin. They were phenols. noncarcinoYCnic aromatic hydrocarbons. lonefihain aliphatic hydrocarbons., long-chainm fatty acids. andl alcohols. A summury, of some of these findings are given in Table 3' (Van Duuren et af. l'973: Van Duuren and Goldschmidt 1976). In these experiments. ? I pure chemicai com- ponents of'CSC and related compounds wta tested for cocarcinogenicity. The compounds were applied to: mouse skin. 50 fe-rtala ICR/Ha Swiss mice per group, thrice weekly. together with! 5Ag per application of BI(u )IP. The duration of' these tests was 4-i0 dhys. The compounds listed in Table 2 all enhanced'f sionificandy the carcinogenicity of B(a,]IP but did not notably reduce the days to first tumors. except in the case of fluoranthene. A recent report of the Public Health Service (1979) lists: all the known phenols: in CSC. Of these compounds, catechol is by far the most abundant phenol! in CSC. Its concentration in nontilter cigarettes ranges from 0. 16 to 0.5 mg per cigarette and 0.0I6 to 0.2 mg per filter cigarette. None of the compounds listed in Table 2 are tumor promoters or carcinoaens on mouse skin. We are. therefore, dealin_ with a series of comvounds that appe:us, innocuous but is harmful in the presence of minute amounts of'a carcinogen. In the course of our work on cocarcinogens cited above, we compared the cocarcinogenic activities of a series of compounds with their tumor-promoting activities. Phenol, for example. is a weak promoter (Van Duuren et al. 1968) wwhereas im our cocarcino__enesis experiments referred to above, it' is a~ tumor inhibitor. Only the two most powerful known tumor promoters. phorbol myris• tate acetate and anthralin (1.8-dihydroxy-9-anthrone) ihowed': both types of' activities (Van Duurerr et al. I'958): Neither of these two compoundy are components of'CSC. STRU CTURE-ACTIVITY RELATIONS'HIPS ~ OF COCARCINOGENS It is noteworthy that 1lthough catechol is a potent cocarcinogen. its position isomers, resorcinol and hydroquimmne. are not only inactive, but partially inhibit the tumorieenicity of' B[u]IP. The cocarcinoeenicitii:s of these various phenols are compared in Figure 1. In the aliphatic hydrocarbon series decane undecane (C„H.,,) and tctradecane (C;,H_,,) are potent corarcino- gens. Their lonSer-chain analoes.,hexadccane (C,,,H_„), cicosane (C,,,H,_),, and' octacosane (C.„H:.,,) are inactive as cocsrcinogenx and some of theae longer- .chain hydrocarbons also partially inhibit B(aJP carcinocenicity. We have ob- served in other recent studies ,imilar ,truetucil speciticitirs for tumor pro- moters, both in the anthralin series (Vam Duuren et al'. 1978i) and in the phorbol ester scriex ('Van Duureniet alL 1'979).

1t)S14I.L. Van Duunn 0 Figure I EY E) ® Cocarcinogenicity ofphenoisf (+) Aetive (-)iinutive:. Since the compound types that were cocarcinogenic in our tests vary so Widely, in their chemical structures, it, is unlikely that they aff have the same mode of action. Several possible mechanisms come to mind to account for the cocarcinoaenicity or turnor-inhibitory activity of these: compounds. They in- chtde: I. induction of carcinogen-activating enzymes; 2. other alterations of metabolic pathways, e.g., activation, or deactivatdon: of dctoxifying enzymes: 3. physical'.effect. i.e.. increased'or decreased rate of carcino-en1 absorption: 4. effects on DNA repair mechanisms. These possible modes of' action are in need of further studies. It will also be very important to examine these cocarcinogens using, such: mouse skin carcino- gens as /3-propiolactone, which, is a direct-acting carcinogen and' nitrosamine carcinogens. Inhalation studies in mice and/or rats are also clcsuly indicated. SUMMARY AND CONCLUSIONS In tobacco carcinogenesis, we are dealing with a complex interplay of a variety of biolbgicalJy actiwe agents. These agents are classifed as shown in Tubie 3, The complexity of'the tobacco: carcinogenesis problem. i.e., the dbvelop- tment'of a less hazardous cigarette, is clear from this table. Although seiectivre faltration of various compound types has been attempted! this route towards a less hazardous cigarette is unlikely to be useful. Alterntacive~routes towards the development of a less hazardous cigarette: have recently been discussed! (Gori' and Lynch 1978). These approaches, particularly the increaseriidcvelopment of' tow-tar cig:uattts, are sensiblr and desirable approaches since there are still some 50 million Americans who eontunue to smoke despite wide publicity and warnings about the relation.,,hiro between attnkini+ ;ina1 hrn•• ~-m•...

Table 2 Cocarcinogenesis en Mouse Skin Number of mice with papillomas Numbe(o/ in Dose Days to mice_ with carc oma Compound (mg) first papilloma total papillomas of skin i'alerlud 2 299 36190 31 1'yrugi1Idi1 5 253 4U/95 33 tkcanc 25 230 38/73 34 UnJccanc 25 195 44/105 41 Pyrena 0.012 1 ti6 26/4 2 20 Ifcnvold'IPyrcne 0.015 246 33/79 27 I-luurimhent 0.W0 99 i9/126 37 . tleniululpyrene (control) .(05 251 14/16 10 eztCsosooT

REFERENCES Berenblum• l: 194 1. The cocarcinogenic action of croton resin: Cancer Res. 1r4t., Bourwell, R•K. and D. K. Bnsch. 1959. The tumor-promotine action ot phenol and!related compounds for mouse skin, Cancer Res. 19:413. Gon. G.B. and C.J. Lynch, 1978. Toward' less hazardous cigarettes. J. Amer. Med. Assoc. 230:1255. Orris. L.. B.L. Van Duuren. A.l. Kosak. N : Nelson, and F':L. Schmitt. 1951i. The carcinogenicity for mouse skin and the aromatic hydrocarbon content of cibpuette- smoke condeni J. Nutl. 4'ancar Inst. 21:557. Public, Health S!ers•ice.,196-t: Smokun,ti and Irrultlr. Rtprort oJ 11u, Adrisun Cirmmittte to the Surgeon Genrrul p~ 58. DHEW publication number 1103. Go%ernment Printing OtHice, Wa•RFiington. D.C: • 1979. Smokiue e und'lieultli: A rrAcrrruJ the Surgron Crntrul 1i!- 55 p • , . . . DHE1V' publication number (PHSI 79-50066: Gosernment Printinb; Olfce. Washington. D: C. , Van IDuunenl B.L. 1958a. tdentitication of some polynuclear aromatic hydrocarbons in cigarette-smoke condensate. J. .ti`utf. Cancer Lnst: 21:1. . 195'ltb. The polynuclear aromatic hydrocarbons in cigarette-smoke conden- ^ sate. lill J. Nud„ C.utcrr l ns t: 2'l .623_ VattiDuuren, B.ll- and L Oim. 1965. The tumov=cnh:utcing princi{+lrs ot•Crutun ri,clrrr.n L. Cunccr R~cs. ZSi 137'1'. Van Duuron, B:L. :und BLtitl Galdsichmidt. 1976. Coc:ucinogcnic and tumor,promotiny a8pnts in toburco curcinwgenr,is.J. Nmtl. Cun~ •er /ot, 56:12.17: Vs,n Duuren. B.L.. 1.A. Bilbao. cutd' C-A. 1o,eph The carcinneenic nitrn-»- 1460 . ,. ~ r • . ~. .. . 1, ,' 110 f 6:1.. Van Duuren Table 3 • Somt: Biohogncally Active Agents of CSIC , Carcinogens Aromatic hydrocarbons and heterocyclics Initiating agents lliromatic, heterocyclic carcinoeetis and borderline carcinogens (dibenziu,clntt' thracene, chrysene, etc.) Promoting agents Phenol and oleic acid Cocurcinogens Catechol, pyro2allol, undecatte, tetradecstte. pyrene. and fluoranthene Tumor inhibitors (cocareinogenesis protocol) Phenol. resorcinol, hydroquinone. oleic acid. squalene. eicosane (C,Hand oc- tacosane ( C;,,H:,,,) ACKNOWLEDGMENT This work was supported by a forrrter U.S. Public Health Service contract :• P+OI-CP33241' froni the National Cancer Institute and by U.S. Public Ht:alth Sern•ice center grants ESq026p'and CA13343'. 1 .a. .~ 1005053119

r « CrreiNttyans and Cofaetors 1111 Van Duuren, B.L.. C. Katz: and B.N. Guldschmidt. 1973. Brief communication: Cocarinogenit: agents in tobaueo carcinogenesis. J: Nrrtl. Cancer lnsr, 51:703. Van Duuren. B.L.. A. Sivak. C. Katz. and S. Fielchionne: 197'i'. Cigarette smoke prcinoFenesis: Importance of tumor promotera. J. .Vutl. Cccnrrr Inst. 47:235. Van Duuren. B:L.. A. SiM1ak. B.,*vL Goidschmidt. C. Katz. and S. Melchionne. 1'97tD. Initiating acti.ity of aromatic hydrocarbons in~two-stage carcinogenesis. J. N'url. Cancer lnsr. 13:11,67. Van Duuren. B.L.. A. Sivak. L. Lattgseth• B.M. Goldschmidt. and A. Siegal. 1968. Initiators and promoters in tobacco carcinogenesis. Murl. Cancer Insst. Mvnu,rr• 28: I'73'. %`in Duurenl B.L.. S.-S. T!.en_. A. Segal. A.C. Smith. S. Melchionne: and 1. Seidman. 1979. Eft'ects of structural ch:utges on the tumor•promoting actisiry of phorbol e:yristate dcetate on moue .kin. Cunccr Ri+s. 39:2644. Van Duuren. B,L... A. Segali S.-S, Txng. G.M. Rusch. G. Loewenczrt. U. `tzte: D. Roth., A. Smith. S. Melchionne. and I. Seidtn:ut: 1978. Structure and tumor- promoting actiii.ity of analogues of anthr.Jin ( l.ti'.dihydroxy9-anthroneu J. .1r1'ed. Chern: 21:26. Wynder. E.L. and D. WoFfmann. 1967. Certain constituents ot' tobacco ptroducts. Dni Tobacco and toHuccn smoke. p. 317. Academic Press. New York. Wynder. E.L.. E.A. GrattarnL and A.B. Croninger. 1953. E.xporimental production of carcinoma with cigarette tar. Cancer Res. 13:d55. COMMENTS BoCK: Dr. Van Duuren. in your test of catechol and some of the other relatively prevalent cocarcino¢ens, were the doses you used about the same as those that would be present in CSC applied in the usual manner. VAa DUUttEN: They were-used!ardoses that did not necessarily correspond too their relative quantities in CSC, and they were used at ma.cimurn toter- ated doses where we saw minimal or no skin-damaging effects. They do.. however, occur within an order of magnitude, or closer. to the relative proportions imwhich they occur in ciea:rette tar. B©CK: Well, I was addressing, myself particularly to catechol. It seemed to me that that was almost in the range that would be present in an ordinary condensate. VAN DUUREN: Yes.. WYNDER: This topic is very apropos because todhy everybody talks about • animal tests and' relates this to human decisions. You regard! the mouse skin test as the best thing we have available today twtest this very factor you've tested. Would you make concl(isions from it to the humun prob. Iem?' Drawing upon your experiencr in promoters andi cot:arcinovens and antipromoters and :utticocarcinoeens. what practical suggestion would you make? How does this relate to your idea oP a less h;umful cigarene? VAN DUUREN: My answer to the tdnt question is no. In dealing with coc:u- einobens. l dbn't think we can eztrr,,polate from mouse skin to the human