Tobacco Documents Online

0 .., . . . ~'t. ~ . '. frotn Banbu~ R .A Sc~c CigdrQte , ~ 980 Cbld 3'piing 1Ie[fipCi ~?= 5;..~-..,~e'~!`.e~.-v..;.. ,.. .*..-'r~aY'. `TN~e ~Role of Voiatille and Nonvolatile N-Ni~trosamilnes 1. _ k Tobacco Carcinogeniesis DIETRICH!HOFFMANN. CHI-HONG B. CHEN, and STEPHEN S. HECHT D~~ivision,of~EnviironrnentallCarcinoge~~nesis~~. Naylor Dana Institute for Disease Prevention Arnerican Health Found'ation Valhalla, New Y©~rk~110585' +2 ~ , (ls43~3P4i~i~3~3'f`ti'': ~ri 1 Epidemiologicall studies have correlated smoking of cigarettes with cancer of' - 1, F riSs~siii'23

The Role of Volatile and Nonvolatile N-Nitrosarrnines in Tobacco Carcinogienesis. , `St!ti?S~Skl3"?i°ftjt~ift ytn ~;2ii^? DIETRICH HOFFMANiN, CHI-HONG B. CHEN, and STEPHEN S. HECHT Division of'EnviironmentallCarcinogenesis Naylor Dana Institute for Disease Prevention American Health Foundation Valhalla, New York 10595 Epidemiological studies~ have correlated smoking of cigarettes with cancer of the oral cavity; pharynx, esophagus, pancreas, renal pelivis, andi urinary bladder (Wynder et al. 1!957; Wynder and Bross 1961; Wynder et al. 1973; Royal College of Physicians 1977; Wynder and Goldsmith 1977; Public Health Service 1979). The increased risk of cancer at these sites in cigarette smokers may be due to several factors. One of these is the likelihood that inhalation of' tobacco smoke leads tio, enzyme: inductions that, in turn, aetivate certain en- vironmental agents to their ultimate carcinogenic forms (Jasko 1979). Other risk factors may be due to the presenl organ-specific carcinogens in tobacco smoke or to: smoke compounds that constitute precursors of' organ-specific carcinogens that can be formed in vivo (Hoffmann et al. 1'978). Druckrey and Preussmann (1962) had already suggested' the possibility that tobacco smoke contains organ-specific carcinogens such as N=nitrosamines. However, inten- sive and systematic studies of N-nitrosamines, in tobaccocarc':inogQnesisdid not begin until 11973. . V()LATILE N NITPiOSAN9INES. The volatile amines in tobacco and in, its smoke (Hecht et al. 1977) may undergo partial nitrosation to N-nitrosamines. To assess the nature and quan- tities of such compounds in tobacco smoke, we developedl a rapid analytical method, taking precautions against artifactual nitrosamine formation during, trapping and aging, of the smoke ('Brunnemanni et al. 1'977). The volatile N-nitrosamines (VNAs) were enrichedlby solvent distribution and chromatog- raphy, separated by gas liquid chromatography (G':LC),and detected and quantitated with~~ ai thermal energy analyzer (TEA)'. This detector is highly sensitive and specific for N-nitrosamines(':Fineetal:. 1975). The influence of nitrate fertil'izersin, the soil onVNIA formation inn cigarette smoke was demonstrated by comparing smoke from tobaccos growni under varyi~ng conditions. Figure I represents GLC-TEA data for VNA con-113

114/ D; Hoffmann, C.B. Chen„and S.S. Hecht ` 60 CATTERTON ~' HIGH NITNATE. 0 W N Z N Ui Q W H lOE t y;lI1tP4~Stft~t>21Pitsf y<n lai~?~,~ yi> t;, A 80iCATTERTON LOw NRRATE I I I I I 0 5 10 15 20 MdNUTES Figure 1 Gas ctiromatograms of'mainstneam smoke M-nittosamines as affected by nitrate levels in cigarette -sxy5 ~rr tobacco centrates obtained in this comparison. All four of'the V NAs identified here are carcinogenic in the experimental animal (Magee et al. 1976). Table 1 presents analytical data for the major VNAs in the mainstream smoke of tobacco. Since tobacco itself contains only negligible amounts of VNAs, it is clear that the VNA yield in the smoke depends primarily on the nitrate content of tobacco, the combustibility of the producty and the proteini content of tobacco. Celllalose acetate filter tips can selectively reduce VNAs by at least, 70% (Table 1)„ Thus, optimal means for redttct2on' of carcinogenic. VNAs in mainstream smoke are selection of low-nitrate tobaccos and use of filter tips with potential for selective reduction. Most present+-day commercial filter cigarettes are effectively reducing VNA. Like ammonia (Schmeltz and HIoffirtann 1!977), VNAs are formed in greater proportions during, smoldering of a cigarette or cigar between puffs than dwring,puff drawing, so that VNA levels in sidestream smoke are signifiicantlyy higher~ tha'nin, mainstream smoke. Thi'sis reflected in data that we established for indoor' environments polluted withi cigarette smoke. Measurable quantities

Nitrosamines in Tobacco Products /116 Table 11 Volatile N-Nitrosamines in Cigarette Smoke Cigarette Mainstream smoke Burley, NFa' Briglit, NF Commercial, NF Commerieal, FAb Kentucky 1R1, NF' Catterton, high NO;,, NIF Catterton, low NIO~„ NF' Frenchs NF French, FA French, FP ' Commerciall,,FA Commercial 1, minus FA Commercial 11, FCd Commercial Il, minus FA Sidestream smoke Commercial, NF Commercial, NIF~ Commercial, FA . Commercial, NF aNonfilter. ''Cellulose acetate filter:, °Paper fil ter. °Charcoal~cellulose acetate filter. eNnitrosodi methylamine. 'N-nitrosoethylmethy,lamine: KN-n itrosAd iethylami ne: °N-nitrosopyrroi idine: Volatile nitrosamines (nglt:igt) NDMhA e' NEIU1i4 ` NDEAg NPYRh 75.9 9.1 2.5: 51.7. 13.2 <0! 1 1.8 6:2 13.0 1.8 1.5 11.0 5.7 0.4 1.3 5.1 9.0 1'_5 2.0 6.6 97.0 8.0 4.8 42.0 20.0' 1.2 2.3' 4.1 29ff 2:7 0.6 25!0 4.3 0!481 0.1 10!5 13.5 2.1 0.4 111.0 6.8 0.5 0~8 8.5 27.0 2.2 1.2 33.0 14.0 0.6 7.6 7.6' 1!9:0 1.2' 8.3' 14.0 680.0 9.4 53:0' 300.0 820.0 30.0 8.2 205.0 730.0 10.0 73.0 390.0 1040.0 10.0 63.0 210.0 of dimethylnitrosamine (D11rIh1)were detected in air samples collected under conditions that precluded artifacts (Table 2; Brunnemann and Hoffmann 1978), In highly polluted indoor environments, a nonsmoker may be exposed to~ an equivalent of the mainstream smoke VNAs of 9- 19nonfilter cigarettes,, or 17-35 filter cigarettesduring~ 1 hour of normal respiration. At thisttime, however; there are no epiderniologie studies suggesting a measurable incre'ased risk for tobacco-related diseases among nonsmokers andl for people working in srnoke-polluted environments (Public Health Service 1979).

116 [1). Hoffmannl C.B: Chen, and S.S: Hecht I Kl3t1?3rftSS#itfly;;lf, f<r, ;•ii:f ? ; > 1.; Table 2 Dirmethylnitrosarrline in Polluted Indoor Air Environment Concentration a (rng/l) Expmsurelhr b (ng) 'Ilcaia I (bar car): 0.13 62-110 Ttain1 2(bar car) 0.11i 53-6'2' Bar 0.24 115-200 Sports hall 0.09 43-76 Betting,parlor 0.05 24-42 Betdng; parlor 0:05 ~ 24 -42 Discotheque 0:09 43 -76 Large room of bank 0:01 5-9 Suburban residenceC ' <0;005 - Urban residencec <0.003 - Blank <0.00'1 - e'Isolated amounts. b Respiratory rate 8-14! l/min. e Nonsmoker's residence.. NONVOLATILE N-NITRQSAMINES In general, cominercial tobacco products contain between 1 and 2% of specif'ic', alkalbids. The prevalent compound among, these is nicotine, constituting at least 90% of the alkaloids in mostl tobacco varieties. Nornicot'ine, anatabine„ anabasine, and cotinine' are the' other common tobacco alkaloids (Figure 2; Schmeltz and! Hoffmann 1977). There also are 5' 10 minor alkaloids, as well N N©RNOCOTINE ANATABINE ANABASINE OXYNdCOTTNE COTINI'NE N 3;2' 91PYNIDYL 03, N N Figure 2. Chemical structures of major tobacco alkaloids

Illitrosamines iniTobacco Products /1'17 . , f ;~tlfi?3tStfi~itf.ttsi ~r, . ?;21 as their oxidation products (Enzell et al. 1977; Schmeltz and Hoffrnann 1977): During, curing and fermentation, tobacco-specific N-nirtrosarni'nes aree formed from these alkaloids (Hecht et, al. 1978b). Im vitro nicotine is nitrosated to N"-nitrosonornicotine (NNN)s 4(~N-methyl-N-nitrosarnino)+l-(3'-pyridyl)-1- butanone (NNIC); and 4-(N-methyl-N-nitrosa'mino)-4-(3'-pyridyl)}butanal ('NNA) (Hecht et al. 1978a). As shown in Figure 3, NNN and NNK have been identified initobacco and initobacco smoke, as has N-nitrosoanatabine (NAtB). For the analysis of these tobacco-specific N-nitrosamines, we developed a specific method~ (Hoffmann et al. 1979) that uses high-pressure liquid chrorna} tography (HPI1,C) with a TEA detector in the final step, resulting, in a clear separation of'the tobacco-specific N-nitrosamines (see Fig. 4). Table 3 sum- marizes analytical findings that demonstrate the relatively high concentration of these N-nitrosamines in tobacco products, ranging from 0i2-90 ppm in to- bacco, 0! 1-4.6 Ftg, in the mainstream smoke, and 0.1-6' F,ug, in the sidestrearn smoke of a cigarette (Hmflfmann et al. 1'979; J.Ji. Piade ehal., in prep.). Studies withi [14C]NNN have shown that 40-50% of the NNN in the smoke is produced by transfer from the tobacco and the remainder is synthe- sized during smoking (Hbffmann et' al. 1977; J.J!. Piade et al., in prep.). Thus, a significant reduction in the formation of these tobacco-specific nitrosamines diaring, tobacco processing shou'ld also be reflected in the rediictioni of these compounds in the smoke. Reducing these cyclic nitrosatnines in tfJbacco is especialQyianportant because they may not beamenabletb, selective filtration at a pH below 6.5. Nornicotine. Nicotine Anatabine NNN NNtf . NAtB Figure 3 Formation ot major tobacco specific N nitrosamines from nicotine. nornicotine, and anatabine during tobaceo process.ing and smoking

~S?3l3tS#f~?i?f 3'n . #it~ii ,r > r t~.,' 1181 D. Hoffmann, C.B. Chen, and S.S. Hacht NYttB NNN N N` CH3 NO NNK T -i I 0 10 24 30, M I NUif E5 Figure 4 High-performance liquid chromatogram ofl nomolatile nitrosamines from tobacco CARCINOGENICITY OF THE'TOBACCO-SPECIFIC'N 11rIITROSAMINES The detection of alkaloid-derived nitrosamines in tobacco products raised thee question of their earcinogeniaity. Results from several bioassays are presented in Table 4. In mice, NNN and NNK induced lung adenomas andl adenocar- cinoma and, in some instances, tumors of the salivary glands (Boyland et all. 1964; Hbffmann et! al. 1976; Hecht et al. 197~9')t When given subcutaneously in rats, NNN inducedl priinarily carcinoma of the nasaU cavirty,, whereas NNK elicited tumprsoflthe nasal cavity„as,well as of the lung and liNer(Hdcht et al. 1980). When administered in, the drinking water, NNN induced primariPycancer of theesophagutsand also some nasal tu~mors, (Hoffmann et al. 1975; Singer and Taylor19716). These results indicated that tobacco-specific nitros- amines may be contact carcinogens as well as organ-specific carcinogens. This observation may offeroneexplanatyorn fbrtheincreased risk for cancer of the oral cavi'tyand esophagus, not, only to cigarette, cigar, and pipe smokers, but also to tbbaceochewers (PublicHealft Service197'9). In Syrian golden hamsters, NNN indiDCestracheal tumors(H'ilifrich et al. 1977). N11dIKis currently under study, and so far it has been shown to,indwce ttacheall tumors. Currently we are painting theora'l eavitiesof aleohpl-treated

z Table 3 Tobacco-Specific N-Nitrosamines in Tobacco Products - - Tobacco (ppm) Mainstream (µg/cigt) Sidestream (µg/cigt) Tobacco Producte NAtB NNN NNK NAtB NNN NNK NAtB NNN NNK Burley cigarette, NF 3.2 7.0 n.dt. b 4.6 3.7 0.32 1.5 6.1 0.66 Bright cigarette, NF 0.44 0.22 0.37 0.41 0.62 0.42 0.39 1.7 0.50 Commercial cigarette, NF 1.6 1.7 0.74 0.33 0.24 0.11 0.27 1.7 0.41 Commercial cigarette, FA 1.3 1.4 0.70 0.37 0.31 0.15 0.15 0.15 0.19 Kentucky IRI, NF 0.62 0.63 0.13 0.53 0.39 0.16 0.19 0.21 0.24 Commercial French cigarette NF, 70 mm 1.8 2.9 0.526 0.18 0.48 0.44 - - -- Commercial French cigarette FA, 70 mm 1.5 2.7 0.37 0.18 0.49 0.36 - French cigarette, NF 2.0 11.9 1.1 0.68 3.2 0.43 - - - French cigarette, FA 2.0 11.9 1.1 0.19 .0 0.19 - - - French cigarette, FP 2.0 11.9 1.1 0.16 0.73 0.12 - - - Little cigar, FA 13.0 45.0 35.0 1.7 5.5 4.2 0.57 0.88 0.81 Cigar (Colombia tobacco) (5.7 g) 3.3 10.7 1.1 1.9 3.2. 1.9 n.d. `' 16.6 15.7 Fine-cut chewing tobacco 44.0 39.0 2.4 aAil cigarettes ana the little cigar were 85 mm long, _ng, except otherwise stated. ''Not detected. 'Not deterrnined (interference by unknowns). tE6fi09Tzaz ~. .x. ~

J o TabIe 4 Carcinogenic Activity of Tobacco-Specific N-Nitrosamines Compounds Species Application Principal organs affected NNN mouse i.p.'' lung (adenoma, adenocar- cinoma) salivary glands (?) NNN NNN NNN NNK rat hamster mouse rat s.c. b p.o.C (water), nasal cavity (carcinoma) liver esophagus (papilloma, carcinoma) pharynx (papilloma) nasal cavity (carcinoma) trachea (papilloma) nasal cavity (carcinoma) %ntraperitoneal. °Subcutaneous. °Per os. SCsVo9zzoz lung (adenoma, adenocar- cinoma) nasal cavity (carcinoma) liver lung (adenoma, carcinoma) ,i;5~ ~. References Boyland et al. (1964) Hoffmann et al. (1976) Hecht et al. (1978b) Hecht et al. (197tfb) Hecht et al. (1980) Hoffmann et al. (1975) Singer and Taylor(1976) HilPrich et al. (1977) Hecht et al. (1978b) Hecht et al. (1980)

Nitrosamines in Tobacco ProductsJ72'1 Syrian hamsters and hamsters on a normal dia wsthi olive oil, solutions of NNN and NNK. Sasedl on macroscopic examination, these applications induce tumors of the:mouth. METABOLIC ACTIUATION OF'TOBACCO-SPECIFIC N-NITROSAMINE& ILike most~ N~nitrosamines, the tobacco-specific nitrosamines are procarcino- gens, which requim in vivo metabollc activation to their ultimate carcinogenic forms. Figure 5 summarizes our present knowledge about the metabolic activa- tion of NNN and NNK (Cheniet a1_ 1978'y 1979). In both cases, the initial step appears to be a-hydroxylatiom The:resulting,a-nitrosaminoalcohols are unsta- blb and decompose with formation of diazohydroxides and, subsequently, carboniumi ions. The latter are possibNy the ultimate carcinogenic forms of these t;obacco-speci'fic carcinogens.. In the case of NNN, the a-hydroxylation occurs in both a-positions leading to two different carbonium ions. These carbonium ions react primarily with intracellular water, forming, a keto alcohol or a hydboxyaldehyde. The latter are oxidieed in vivo to the corresponding~ ketio~acidk. NNK is a,hydroxylated au the methyl group and at the a-methylbne group, respectively. These compounds decompose with formationi of a metliylcar- bonium ion and a 4-(/3-pyridyi)-4-ketobutylcarbonium ion. The butylcarbonium ion rmay, be correlated in rats with tumor formation in the nasal cavity, whereas ~M•0 0 M~ 0 / M ~N3 Figure 5 !vletabolism oE NNN and'NNK LM I ON

122! D. Hoffmannl C.B. Chen, and,S:S. Heoht the methylaarboniumi ion may be responsible for the occurrence of liver, tumors. Currently; we are studying this concept with, ["4C]NNQC Iabeled, at the methyl group or at the carbonyl gnoup.. In addition to «-hydroxylation, we detected other metabolites of NNN that are )3-hydroxylated. Pyridine-N-oxides were also identified as metabolites of both NNN' and NNK. At present, we are detenmining, the biological signifi- cance of these N-oxides in bioassays. Furthermore, we are exploring methods of chemoprevention by inhibiting, metabolic activation of tobacco carcinogens and'by increasing the :detoxificatiom TOBACCO-SPECIFI6!N-MITROSA'MINES IN THE SAL'IVA OF'TO'BACCIO CHEWERS The study of the metabolic activation of the tobacco~specific carcinogens is na merely of acadenua interest to us. It gives us the opportunity to examine fluids of smokers and chewers for metabolites of NNN and NNK and, thus, too evaluate whether the metabolism of these carcinogens in man is similar to that in theexperi'mentale animal. ft is also important to, explore why the degree ofmetabol'ic activation of these tobacco carcinogens differs widely in individual chewers and smokers. In vitro experiments have shown tha N-nitrosamines can be formed from the : tobacco alkaloids by incubation of snuff wit'h~ saliva (Hecht et al'. 19°75). We therefore investigated'tihis phenomenon in tobacco chewers and& smokers. In a preliminary study, we were also able to show that concentrations of nonvolatile N-nitrosamines in the saliiva of individual tobacco chewers vary signifiaantNy. This supports the concept that in vivo nitrosation of the tobacco: alkaloids is involved in the formation of these nitrosamines; We have initiated a large-scale controtledl study on the formation ofINNN, NNK, and NAtB in the oral cavityofl chewers and smokers, including analysis of saliva for metabolites of NNN and NNK., SUMMARY Chemical-analyt'ical studies have shown that tobacco smoke contains small amounts of volatile N-nitrosamines. In the mainstream smoke of' ai cigarette, theseearcinogenscan be reduced significantly by utilization of tobacco lonvin nitrate content and especially by filtration through cellulose acetate filter tipsthat~ selectively , retain volatile: N-nitrosamines. Cured and fermented tobacco and tobacebsmoke contain three majQr nonvolatile N-nitrosamines. These:are formed from nicotine, nornicotiine, and anatabine. In mice, rats, and Syrian golden hamsters, NNN' and NNIfK are proven carcinogens. They are organ-specific carcinogens, burtmay also be contact carcinogens. NIAtB', the third identified totiacconitrosamine, is cuo- rentlybe:ing bi~~oassaye~d, for its carcinogenic activity. :<x>SSss ~ is.{it~ti9a~"'

Nitrosamines in Tobacco Products /123 ~•;~t!$?dtSt#~3y3Y?iif f~3'r?t h'~ii' ~3?i NNN and NNK are metabolically activated by a-hydroxylation. The a-hydroxynitrosamines are unstable and decompose with formation of car- bonium ions, the probable ultimate carcinogenic form of these tiobacco,speci'fic compounds. N-nitrosation of the alkaloids occurs during the processing of tobacco and ditring, smoking, but may also take place in vivo. This has been demonstrated by evidence for the formation of NNN, NNK and NAtB' during tobacco chewing,. Detailed studies on the in, vivo formation of these tobacco carcinogens in chewers and smokers has begun. These studies will be folltnwed by the development of methods that inhibit the formation of these compounds. ACKNOWLEDGMENTS Our studies are supported, in part, by Public Health Service contract NO 1-CP' 55666 and by grant CA-21393 from the Divisioni of Cancer Cause and Prevention of the National Cancer Institute andl by American! Cancer Society research grant BC-56. S. S. H. is a recipient of National Cancer Institute research career development awardl 5K04CA00124. C. B. C. is recipient of National Institute of Environmental Health Sciences development award no: ESO12236. REFERENCES Boyland, E., F.J.C. Roe, and J.W. Gortod. 1964. [nductioni of pulmonary tumors in mice by nitrosonornicotine, a possible constituent! of tobacco smoke. Nature 2a2:111z6_ Brunnemann, K.D. and D. Hoffmann. 197$' Analysis of volatile nitrosamines in tobacco smoke and polluted'indoor environments. /nt: Agency Sci. Publ. 9:3431 Brunnemann„ K.D., L. Yu, and D. Hoffmann. 11977„ Assessment of carcinogenic volatdlo N=nitrosamines in tobacco and in mainstreamiand sidestream smoke from cigarettes. CancerRes., 37:3218. Chen, B.C., SS. Hecht, and D! Hoffmann. 1978. Metabolic a-hydroxylatiorr of the tobacco-specific N-nitrosonornicotine. Cancer Res. 36:3639. Chen„B.C., S.S. Hecht, R. Young, T. Ohmori, and D: Hoffmann 1979. Comparative carcinogenicity and metabolism of the tobacco-specific caricinogens, NNN and NNK. Pkoc. Am. Assoc. Cancer Res. 20:81. Dtuckrey, H. and R. Preussmann. 1'962. Zur Entstehung karzinogener N8trosamine am Beispielldes Tabakrauches. Naturwissenschaften 49:498. Enzell, C.R., I. Wahlberg, and AS. Aasen. 1977. lsoprenoids and alkaloids of tobacco. Fortschr. Chem, Org. Naturst: 34:1. Fine, D.H., F. Rufeh, D. Lieb, and D.P. Rbunbehlcr. 1975. Description of the thermad energy analyzer (TEA)fartrace determination of volatile and narnvolatileN-nitroso compounds. AhaP. Chem. 47c l L88. Hecht, S.S., R.M. Ornaf, and D. Hoflfmann. 1975. N-nitrosonornicotine in tobacco; analysis of possible contributing factors andlbioloQic implications.1. Natl. Cancer Liant. 54i 12-37: Hechti, S.S., 1. Schmeltz, andl D; Hoffmann. 1977. Ni'trogenouscompounds in cigarette smoke and their precursors. Recent Adt'. Tobacco Sd,. 3:59! vrr ~ fa:t.i~ii?! "i i~,i::}t;•

1'24l D. Hoffmann, C.B. Chen, and S.S. Hecht li.'tisfirt~t?~slt3}i;?i~ s~n, rY;~ Hecht, S.S., C:B. Chen, and D! Hoffmann. 1979. Tobacco-specific nitrosamines: Occurrence, formation, carcinogenicity and metabolism. Acc. Chem, Res. 12:92'. Hecht, S.S., C.B: Chen,, T. Ohmori, andl D. Hoffmann. 1!980. Comparative carcino- genicity in F~34k1 rats of the tobacco-specific nittnsamines N'-nitrosonomicotine and 4-(iN-methyl-N-nitrosamino)-1-(3-pyridyl)-1-butanone. Cancer Res, (in press): Hecht, S,S:,, C.B:. Chen, R.M. Ornaf, E. Jacobs, J.D. Adams, and' D: Hbffmanm 1978a. Reaction of nicotine and sodium initrite: Formation,of' nitrosamines and fragmentation of the pyrrplidine ring. J. Org. Chem. 43:72. Hecht, S'.S., C.B. Chen, N. Hirota, R.M. Onnaf; T.C. Tso, and!D, Hoffmann. 1978b Tobacoo>specific nitrosamines: Formation from nicotine in vitro and during to- bacco curing and1 carcinogenicity in1 strain A mice. J. Natl: Cancer Inst. 60:819. Hilfiich„ J., S.S, Hecht,, and D. Hbffimann. 1977. Effects of N'-nittosonornicotine and N'-nitrosoanabasine in Syrian golden hamsters. Cancer Lett,. 2:1169. Hoffrnann, D., 0. Dong, andl S:S'. Hecht. 1977. Origin in tobacco smoke of N'- nitrosonornicotine„a tobacco-specific carcinogen. J. Nat1: Cancer Ivrsr. 58i l'84L. Hoffmann, D., S.S. Hecht, R.M. Ornaf, and E:L. Wynderi. 1'976; N'-nitrosonor- nicotine in tobacco: Formation and careinogenicity. Intern. Agency Sci: Publ.. 14:307. Hoffmann, D:,, 1 Schmeltz, S.S~: Hecht, and E.L. Wynder. 1978. Tobacco car- cinogenesis. In Polycyclic hydrocarbons and cancer (ed. H. V. Gelboin and P. O. P. Ts'o),, vol. 1, p:85 L Academic Pness, New York. Hoffmann, D., R. Raineri, S.SI Hecht, R. Maronpot, and E.L. Wynder: 1975. Effects of N'-nitrosonomicotine and N'-nitrosoanabasine in rats. J. Natl. Cancer Inst: 55:977: Hoffmann, D., J.C. Adams, K.D. Brunnemann, and S.S. Hecht. 1979. Assessment of tobacco-specific N-nirtrosamines in tobacco products. Cancer Res. 39:2505. Jasko, W.J. 1'979! Influence of' cigarette smoking, on drug metabolism in man. Drug Metab. Rev: 9:22. Magee, P.N., R'. Mbntesano, and R. Preussmann. 1'976: N-Nitroso compounds and related carcinogens. Avn: Chem. Soc. Monogr. 173:491. Royal College of: Physicians. 1977. Smoking and h'ealth. Pitman Medical Publishing Co., Ltd., London. Public Health Service 1979. , SSmoking and, health: A report to the Surgeon GeneraL DHEW publication number 79-50066. Government Printing Office; Washington, ID.C: Schmeltz, I: and D. Hoffmann. 1977. Nitrogen-containing compounds in,tobacco and tobacco smoke. Chem: Rev. 71:295. Singer, G,.[vt. and H.W. Taylor. 1976. Carcinogenicity of N'-nitrosonomicotine in Snrasue-Dawley rats: J. NatL Cancer Inst. 57:1275. Wynder; E. L. and L.J: Bross. 19611. A study of etiological factors in cancer of the esophagus. Cancer 14:385: Wynder, E:,L. and R. Goldsmith. 1977. The epidemiology of bladder cancer. A second'' look. Cancer 40:1246. Wynderi, E.L., I.J. Bross, and K.ML Feldman., 1957. A: study, of etiological factors in cancer ofl the mouth. Cancer 10:13I00, Wynderi, E. L.:, K. Mabuchi, N. MaruQhi, and J. G. Fortner. 19731 Epidemiology of cancer of the pancreas: J: Nat. Canc•er Inst. 50:50.

Nitrosamines in Tobacco Productsl125 CO'MMEN1rS' GORi: Have you foundi any compounds that may be inhibitory of the car- cinogenic activity of nitrosamines? HOFFMANN: We have no data on the in vivo inhibition of the formation of tobacco-specific N-nitrosamines. We have found the opposite. This aspect was already suggested by Boyland et al. (19Cn4) many years ago. Thiocyan+ ate catalytically increases,tihe in vivo nitrosamine formationiduring chewing andl smoking. We found the thiocyanate concentration in the saliva, to be correlated witbi the q}iantity of tobacco smoked. This was expected, since thiocyanate is the major detoxification product of' HCN, a major smoke constituentL As discussed before, N-nitrosamines are procarcinogens and, thus, they have to be activatedl metabolically. Thus far„ our bioassay data are not completed, however, in vitro studies by Chen and Hiecht'. (1978) from our group have shown t'hatone can inhibit the metabolic activation of N-nitrosonornicotine. The best approach towards reduction of the nitrosamines in smoke wouldd obviously be to use low-nitrate tobacco. However, when, we reduce the nitrate content of tobacco, we have increased formation of carcinogenic polycyclic hydrocarbons. At this moment we regard selective filtration as the best approach towards reducing volatile and tobacco-specific N- nitrosarnines in the smoke. WYNDER 11 wonder whether you could comment on those experiments which again support our interplay between epidemiological observation and ex- perimental stWdies. Hurnandatalhave shown that high alcohol consumption increases the risk of cancer of the esophagus, and the experimental work done at our institute by McCoy et al. (1i979) showed that the metabolicc activation of nit'rosamines via a-hydroxylation is increasedL HOFFMANN: Yes, one of our studies has shown that Syrian golden hamsters maintainedion anialcoholic diet had increased a-hydroxylatiionin the liver and cheek pouch of N-nirtrosonornicotine, compared to the a-hydroxylation occurring,inthe same organs of hamsters oni a standard diet. Human studies, especially in France and the U.S., have documented that drinking, in additioni to smoking, increases the risk for cancer of the esophagus andloral cavity over that for nondrinking,smokers (Tuyns 1970; Kissin et al, 1973; Wynder et all 1977). Therefore, model studies with the Syrianigolden hamster were: initiated limourinstitute. As mentfioned before, Dr. McCoyhasshown~that whenihamsters, wereonanalcoholicdiet,,themetabolic activation of the carcinogenic nitrosamines was significantly increased in both the liver and the hamster pouch. We have an ongoing experiment in which: weapplytobacco-s~pecificn'itrosamines by swab totheoratcavity;of hamsters. Onegpoup of hamsters i,

1261 @. FHoffmann, C.B. Chen, and S:S, Mecht 1 i~ll~t~4iitn~s?fti?!?£ it~'r=S 3'~?1i1 ° > ~~.~;' is on ainormal dietand the second group on an alcoholic diet. The technique of painting the hamster pouch is relatively simple and by simple manipuba- tionione can readily observe the treated tissue. There is no questioni that! alcohol andl nitrosamine together induce more tumors ini the hamster pouch than the nitrosamine in conjunction with the standkrd alcohol-free diet. At present, however, these results arebasedlon macroscopic observations only. Upon histopathological confirmationof the macroscopic observation, we expect this study to fully support the epidemiological data, namely that smoking andldrinking together represent a high risk for cancer of'the oral cavity and esophagus. BATTISTA: Over the years we have been talking about nitrates; an increase in, nitrates results in the decrease of the carcinogenicity of tobacco smoke (when applied to the skin): You were finding that increasing the level of nitrate increases the nitrosamine levels, which one wouldI expect, to give a higher incidence ofl cancer. Now, in none of the data presented here have we looked directly at effects of nitrosamines in smoke on the lung. We apparently have two diametrically opposite effects to nitrate. Would you like to clarify this point?' Hio,FFMAIwN: Nios we: have done it. I'have not ment'ionedI it because of our time limits today. We have painted NNN together withitobacco tar on the skin of mice. This resulted in an increase of lung adenomas„but not of skin tumors,, again suggesting that N-nitrosamines are organ-specific carcinogens. Obvi'- ously, one goal in tobaccocarcinogenesis is to develop low-tar, low niicotine cigarettes. Tobacco which is high in nittatehas increased combustibility andl the nitrogen oxides generated from, the nitrate dL+ring smoking serve as scavengers of C,H-radicals and„ thus, inhibit the pyrosynthesis of the carcinogenic pollycyclic hydrocarbons. This is, of course, desirable since reductionof polycyclic aromatic hydrocarbons in the smoke leads to reduced tumorigenic activity as measured on mouseskim Inhalation experiments with hamsters by Dontenwill et al. (1973) have shown; thatsrnoke fromnitrate-rich cigarette tobaccos is also less, tumorigeniic i~nthe larynx. Thus, we believe that the use :of nitrate-rich tobacco has been and is a step forward inithe directioniof the less harmful cigarette. What we have to try now is to reduce the nitrosamines in the smoke of these cigarettes. In the case of the volatile nitrosamines, this has already been accomplished in the smoke of U.S. cigarettes. Ninety percent of all U.S. cigarettes have celQuloseaaetate filter tips. These are capable ofseiecti'velyrernovingt'~he volatile nitrosamines in the smoke up to: 85%. In that respect there wouldl be two ways to reducetobaeco-specific carcinoge~ns~ First! is selective fiiltratiion, and preliminary data in this area appear quite promising. A second approach would be to keep:the alkaloids low in the tobacco, because without the alkaloids, these nitrosamines cannot be formedl. That is why we are very much opposed to hssk`s:

Mitrosamines in Tobacco Products / 127 the introduction of low-tar„high-niicotine cigarettes. One thing we are sure of is when you have a high alkaloid content in tobacco, youi have : a high yield ofl tobacco-specific carcinogens in the tobacco and in the smoke. Thus, the suggcstion;of low-tar, high-nicotine cigarettes as discussed inithe United Kingdom,, is counterproductive because the higher levels of nicotine in the smoke will continue to support habituation of smokers and will also result in higher nitrosamine content ofthesrnoke, unless methods for their selective filtration are found. References Bioyland'4 E., F.J.C. Roe, and J.W. Gorrod. 1'964. Inductionof pulmonary tumors in mice by nitrosonornicotine, a possible constituent of tobacco smoke. Nature 202:1126. Chen, B.C:, S.S. Hecht, and D. Hoflfrttann. 1978. Metabolic a-hydroxylationi of the tobacco-specific N'-nitrosonomicotine. Cancer Res. 36:3639: Dontenwill, W., H. Chevalier, H. Harke, U. Lafrenz, G'. Reckzeh, and B. Schreider. 1973. Investigatinnion the effects of chronic cigarette-smoke inhalation in Syrian golden hamsters. J. Natl: Cancer Inst. 51i:17$'1. Kissins B., M.M. Kaley, W.H. Su, and R. Lermer: 1!973. Head and neck cancer ini alcoholics. The relationshipof drinking, smoking and dietary patterns. J. Am. Med. Assoc. 224:1174, McCoy, G.D., C-h. B. Chens S:S: Hecht, and E.C. McCoy. 1979. Enhancement of metabolismiand mutagenesis of nitrosopyrrolidine in liver fractions isolated from chronic ethanol-consuming hamsters. Cancer Res. 39:793. Tuyns, A.J. 1'970. Cancer of the esophagus. Further evidence of the relation of drinking habits inFrance. Intl. J. Cancer 5: 1I52'. Wynder, E. L. ,, IVI_ Mushinski, and J,C. Spivak. 1977. Tobacco and alcohol consumption in relationito the development of multiple primary cancers. Cancer 40:1872. ki<.uxN>oooSka ~~;p~,ysyM.+