variety of individual types of cancer (including laryngeal cancer) with the history of such use in persons with the remaining cancers thought not to be related to tobacco use (ZEi,,. Prior experience with smokeless tobacco was divided into two levels of exposure. The estimates of the relative risks were controlled for age, race, and smoking. Relative risks of laryngeal cancer in men of 2.0 and 1.7 were found among individuals with low and high levels, respectively, of exposure to chewing tobacco or snuff. These estimates were not significantly different from 1.0. They are based on 106 cases, 11 with relatively low exposure and 5 with higher exposure, and 2,102 controls of which 98 had low exposure and 7 1 had high exposure. Only 13 female laryngeal cases were available for analysis in this study, which was insufficient to provide any meaningful results. A case-control study by Wynder and Stelhnan included 387 male cases of laryngeal cancer and 2,560 hospital controls (13). The percent- ages that had previously used chewing tobacco and snuff were 11.9 and 3.9, respectively, for the cases, and 9.0 and 2.7, respectively, for the con- trols. Based on these findings, crude relative risks of 1.4 for chewing tobacco and 1.5 for snuff were obtained. Neither estimate differs signifi- cantly from 1.0. No control for smoking or alcohol was done, although the authors state that cigarette smoking in users and nonusers of chew- ing tobacco was simiIar. Interviews with 560 laryngeal cancer patients and 2,000 controls from the general population of Bombay revealed significantly increased risks, compared to nonchewers, among chewers of betel without tobacco (relative risk 2.5) than with tobacco (relative risk 2.6) (21). I.ayngd cancer was noted to comprise an unusually high proportion of all cancer diagnoses in a hospital series in eastern India where pan chewing is com- mon, but no assessment of the role of tobacco was made (26) Stomach Cancer Zacho et al. noted that, in Denmark, both gastric cancer and use of chewing tobacco and snuff are directly related to age, more common in men than women, more prevalent in rural than urban areas, and in- versely related to socioeconomic status (27). On the basis of these obser- vations, they hypothesized that use of smokeless tobacco increases the risk of stomach cancer. Obviously, other differences among individuals within Denmark could also explain these findings. Weinberg et al. conducted a casecontrol study of stomach cancer in a coal mining region of Pennsylvania (28). Cases who had died of stomach cancer from 1978 through 1980 were compared with three control groups: persons who died of other cancers of the digestive system, per- sons who died of arterial sclerotic heart disease, and persons who lived in the same neighborhood as the case. All controls were matched to indi- vidual cases on age, sex, race, and location of residence. Data on the use 51 of various forms of tobacco were obtained by interviewing next&kin or (for neighborhood controls) the subjects themselves. About 16 percent of all men in the study had used chewing tobacco. This percentage did not differ significantly among the cases and the three control groups. No women in this study had chewed tobacco. This study provides some evidence to suggest that chewing tobacco does not increase the risk of gastric cancer, although a small increase in risk could have been missed due to lack of statistical power. The case-control analysis of the interview data from the TNCS found a relative risk of stomach cancer of 1.7 in men in the highest level of use of chewing tobacco and snuff, no increase in men in the lower use category, and no increase in women (W). These results are based on 120 male cases, 12 of which were users, and 82 female cases, 2 of which were users. The power of this analysis to detect a true increase in risk is ob viously low. The relative risk of 1.7 was not significantly greater than 1.0. In an abstract describing a cohort mortality study of U.S. veterans, the standardized mortality ratio for stomach cancer among non- smoking users of smokeless tobacco was 151, but no study details were provided (16). Urinary Tract Cancer Constituents of smokeless tobacco can enter the blood stream, and some are excreted in the urine. The kidney and bladder are thus poten- tially exposed to these agents but presuma bly in lower concentrations than are tissues of the upper aerodigestive tract. In a hospital-based case-control study in Seattle, Washington, patients who chewed to bacco were reported to be at nearly a fivefold increased risk of renal cancer compared to nontobacco users (29). Only 6 percent of the 88 male cases were chewers. No association between the use of smokeless to bacco products and either renal cell or renal pelvis cancer was reported in a case-control study of these tumors in England (39). Among 106 renal cell cancer case-control pairs in this study, 10 cases versus 11 con- trols had at some time used smokeless tobacco. Among 33 renal pelvis cancer-control pairs, 2 cases and 3 controls reported ever using smoke less tobacco products. In a large population-based study in Minnesota involving 495 cases and 697 controls, a nonsignificantly increased rela- tive risk of renal cell cancer of 1.7 (95~percent confidence interval 0.5-6.0) was found among snuff users after adjusting for smoking (31). There was a deficit in risk, however, associated with ever using chewing to bacco (relative risk 0.4, 95percent confidence interval 0.1-2.6). A review of eight epidemiologic investigations revealed no consistent evidence that the risk of bladder cancer is altered in users of smokeless to bacco products (table 2) (13,25,32&?). The National Bladder Cancer Study is the largest of the investigations of bladder cancer considered in this review (37). Cases for this study were selected through 10 population- 52 TABLE I.-Estimates of Relative Risks of Bladder Cancer in Persons Who Have Ever Used Chewing Tobacco and Snuff Relative Risks Years Fist Author Case Chewing (ref.) Diagnosed Sex Tobacco Both Snuff Wynder (32) 195783 Male 1.4* 0.7* Dunham et al. (33) 1958-64 Male 5.3*t 0.9*t - Female 1.1*t - 0.3*t Cole et al. (34) 196688 Both 1.1* 1.0* Williams and 1969-71 Malelevel 1 1.61 Horm (25) level 2 1.15 Female-level 1 0 level 2 1.78 Wynder and 1974-75 Males 0.9 0.7 stellman (13) Howe et al. (36) 1974-76 Males 0.9 Hartge et al. 137) 1977-78 Males 1.02 0.77t o Estimated from published report. t Based on analysis of nonsmokers only. based cancer registries in the United States. Controls were a random sample of the same population from which the cases came. Information was obtained from interviews of 2,982 cases and 5,782 controls. Analy- ses of smokeless tobacco use were restricted to the 340 cases and 1,227 controls who claimed never to have smoked cigarettes. Of these, 11 per- cent of the cases and 10 percent of the controls had ever used chewing tobacco, and 3 percent of the cases and 4 percent of the controls had ever used snuff. The relative risks of bladder cancer in users of chewing tobacco and snuff were estimated to be 1.0 (0.7-1.5) and 0.8 (0.4-1.6). respectively. Wynder et al. conducted a hospital-based study of 300 male bladder cancer cases (j72). Eleven percent of the 300 cases and 8 percent of the 300 hospital controls had ever used chewing tobacco; 2 percent of the cases and 3 percent of the controls had used snuff. The percentage of users was not significantly different in cases and controls, and no attempt was made to analyze the data further. Dunham et al. interviewed 493 bladder cancer patients and 527 hospi- talized controls in New Orleans (33). Among nonsmokers, there was an increased relative risk associated with chewing tobacco use among males but a deficit in risk associated with snuff use among females, but the numbers of cases involved were small (four males and three females). Cole et al. interviewed 470 cases from the Boston area and 500 population-based controls (34). Forty-six of the cases had used chewing 53 tobacco and three had used snuff. Based on the prior experience with smokeless tobacco in the controls (controlling for age and sex), 42.3 and 7.9 cases would have been expected to have used chewing tobacco and snuff, respectively. Some increase in the risk of bladder cancer was found in the TNCS survey, but none of the risks from this study are sig- nificantly different from 1.0 (table 1) (2.5). In addition, no evidence of a dose response is seen. In a second hospital-based casecontrol study (13) of similar design to the first (32), Wynder and St&man found that 8 percent and 1.9 percent of 586 cases had used chewing tobacco and snuff, respectively, com- pared to 9 percent and 2.7 percent of 2,560 controls who had used these two products. When analyses were restricted to nonsmokers in a con- tinuation of this study, a significant excess risk of bladder cancer was associated with snuff use among women, but only 3 of 76 cases were users (35). A population-based casecontrol study was conducted in three Cana- dian provinces by Howe et al, (3@. Controls were matched to individual cases on neighborhood, age, and sex. The ratio of male pairs discordant for the use of chewing tobacco was 29134, giving a relative risk of 0.9 (95~percent confidence interval, 0.5-1.6). This estimate was not altered by controlling for smoking. No female cases or controls gave a prior history of use of smokeless tobacco. In Denmark, 165 male and 47 female patients with cancer of the uri- nary bladder from a hospital serving a specific geographic area were interviewed, as were geographically-matched controls (3&Z?). The esti- mated relative risk associated with tobacco chewing was 2.0 (1.2-3.4) based on 39 exposed cases. In a logistic model containing variables for tobacco chewing, smoking, and other major correlates of bladder can- cer, the relative risk associated with chewing was 1.7 and statistically significantly higher than 1.0. The authors estimated that tobacco chew- ing might account for 9 percent of the bladder cancer diagnoses in the area. Although two studies did report elevated relative risks associated with smokeless tobacco use, on balance these studies provide little evi- dence to suggest that smokeless tobacco alters the risk of bladder cancer. It is possible that a small increase in risk has not been detected by the studies not reporting increases due to lack of statistical power. Other Cancers All other organs of the body are likely exposed to even lower concen- trations of products of smokeless tobacco via the blood. In a large prospective study in Norway, 16,7 13 individuals were inter- viewed to obtain information on the use of tobacco and alcohol and were followed up for development of pancreatic cancer (40). Sixty-three per- sons in the cohort developed this neoplasm during a lo-year followup. 54 After controlling for cigarette smoking and alcohol consumption, a rela- tive risk of 2.9 was observed in regular users of chewing tobacco or snuff (compared to nonusers). The 95percent confidence limits of this value include 1.0. Risk was greater in regular users than former or occasional current users, and a trend of increasing risk with amount used was of borderline statistical significance (P=.O6). The case-control analysis of the interview data from the TNCS (24) with respect to pancreas cancer is based on only 91 male cases (3 exposed to smokeless tobacco) and 85 female cases (none exposed); and although no increase in relative risk of pancreatic cancer in relation to smokeless tobacco was observed, the power of this study to detect such an increase is low. Other cancer sites were found to be related to the use of smokeless tobacco in the casecontrol analysis of the interview data from the TNCS (24). Relative risks for colon cancer at low and high levels of expo- sure were found to be 0.9 and 1.5 for men and 0.4 and 2.0 for women, respectively. Relative risks of cervical cancer in users of these two levels of exposure were 3.1 and 2.3. No studies have been conducted to con- firm or refute these findings. In view of the large numbers of possible associations investigated, these results should be considered of value only in generating hypotheses for further investigation. Summary The epidemiologic studies showing an association between the use of snuff and oral cancers indicate that topical exposure of tissues to smokeless tobacco can cause cancers at the site of the exposure. Case reports of neoplasms developing in the ear and nose of individuals who used snuff at these sites raise the possibility that direct exposure may increase the risk in locations besides the oral cavity. Other tissues that come in contact with constituents of smokeless tobacco in more dilute concentrations include the linings of the esophagus, larynx (supraglotic portion), and stomach. Results of studies of cancers of these three sites in relation to smokeless tobacco are inconclusive; many are of limited power to detect small increases in risk and did not control for relevant, potentially confounding variables. However, some studies of these three cancers do show an increase in risk in relation to the use of smoke less tobacco. Constituents of smokeless tobacco can enter the blood- stream, and some are excreted in the urine. The kidney and bladder are thus potentially exposed to these products and their metabolites but presumably in lower concentrations than are tissues of the upper aero digestive tract. Evidence suggests that the risk of bladder cancer is not altered to any large extent in users of smokeless tobacco products, but results from studies of kidney cancer are inconsistent. Information regarding the risks of other cancers in relation to smokeless tobacco use is sparse. 55 References (1) Root, H.D., Aust, J.B., and Sullivan, A. Snuff and cancer of the ear. N. Engl. J. Med. 262: 819-820, 1960. (2) Redmond, D.E. Tobacco and cancer: The first clinical report, 1761. N. Engl. J. Med. 2%: 18-23, 1970. (3) Higginson, J., and Oettle, A.G. Cancer incidence in the Bantu and Cape colored races of South Africa: Report of a cancer survey in the lYans- vaal. J. Natl. Cancer Inst. 24: 589-671. 1960. (4) Shapiro, M.P., Keen, P., Cohen, L., and de Moor, N.G. Malignant dis- ease in the lYansvaal, III. Cancer of the respiratory tract. S. Afr. Med. J. 29: 95-101, 1955. (5) Keen, P., de Moor, N.G., Shapiro, M.P., and Cohen L. The aetiology of respiratory tract cancer in the South African Bantu. Br. J. Cancer 9: 528-538, 1955. (6) Baumslag, N. Carcinoma of the maxi&q antrum and its relationship to trace metal content of snuff. Arch Environ. Health 23: l-5,1971. (7) Acheson, E.D., Hadfield E.H., and Macbeth, R.G. Carcinoma of the nasal cavity and accessory sinuses in woodworkers. Lancet 1: 311-312, 1967. (8) Anderson, H.C., Anderson, I., and Solgaard, J. Nasal cancers, symp- toms, and upper airway function in woodworkers. Br. J. Int. Med. 3p: 201-207, 1977. (9) Brinton, L.A., Blot, W.J., Beck er, J.A., et al. A case-control study of cancers of the nasal cavity and paranasal sinuses. Am. J. EpidemioL 119: 896-905, 1984. (10) HouJensen, K. On the occurrence of post nasal space tumors in Kenya. Br. J. Cancer 18: 58-68,1964. (11) Shanmugaratnam, K., and Higginson, J. Etiology of nasopharyngeal carcinoma: Origin and structure. In: C. Muir, K Shanmugaratnam (eds.). Cancer of the Nasopharynx. UICC Monogr. 1: 153-162,1967. (12) Wynder, E.L., and Bross, I.J. A study of etiological factors in cancer of the esophagus. Cancer 14: 389-413,1961. (13) Wynder, E.L., and St&mm, S.D. Comparative epidemiology of tobacco related cancers. Cancer Res. 37: 4608-4622,1977. (14) Martinez, I. Factors associated with cancer of theesophagus, mouth, and pharynx in Puerto Rico. J. Natl. Cancer Inst. 42: 1069-1094,1969. 05) Pottern, L.M., Morris, L.E., Blot, W.J., et al. Esophageal cancer among black men in Washington, D.C., 1. Alcohol, tobacco, and other risk fac- tors. JNCI 67: 777-783,198l. (lf$ Winn. D., Walrath, J.. Blot, W., and Rogot, E. Chewing tobaccoand snuff in relation to cause of death in a large prospective cohort (Abstract). Am. J. EpidemioL 116: 567, 1982. (17) Bjelke, E., and Schuman, L.M. Chewing tobacco and use of snuff: ReIa- tionships to cancer of the pancreas and other sites in two prospective studies. F'rowdngs of the 13th International Congress on Cancer, 1982, p. 207. 56 (18) International Agency for Research on Cancer. Tobacco habits other than smoking: Betel-quid and areca-nut chewing; and some related nitrosamines. IARC Monogr. EvaI. Carcinog. Risk Chem. Hum. 37: 103-104, 1985. (19) Stephen, S.J., and Uragoda, C.G. Some observations on oesophageaI carcinoma in Ceylon, including its relationship to betel chewing. Br. J. Cancer 24: 11-15, 1970. (20) Waterhouse, J., Muir, C., Shanmugaratnam, K., and Powell, J. Cancer Incidence in Five Continents, Vol. IV. Lyon, France, International Agency for Research on Cancer, 1982. (21) JussawaIIa, D.J., and Deshpande, V.A. Evaluation of cancer risk in tobacco chewers and smokers: An epidemiologic assessment. Cancer 28: 244-252, 1971. (22) JussawaIIa, D.J. OesophageaI cancer in India. J. Cancer Res. Chn. Oncol. 99: 29-33, 1981. (23) Shanta, V., and Krishnamurthi, S. Further study in aetiology of carci- nomas of the upper alimentary tract. Br. J. Cancer 17: B-23, 1963. (24) Cook-Mozaffari, P.J., Azordkegan, F., Day, N.E., Ressicaud, A., Sabai C., and Aramesh, B. OesophageaI cancer studies in the Caspian littoral of Iran: Results of a casecontrol study. Br. J. Cancer 39 293-309,1979. (25) WiIliams. R.R., and Horm, J.W. Association of cancer sites with to bacco and alcohol consumption and socioeconomic status of patients. Interview study from the Third National Cancer Survey. J. Natl. Cancer Inst. 58: 525-547, 1977. (26) Sarma, S.N. A study into the incidence and etiology of cancer of the Iarynx and adjacent parts in Assam. Indian J. Med. Res. 46 525-533, 1958. (27) Zacho, A., Nielsen, J., and Larsen, V. On the consumption of unburned tobacco in patients with cancer of the stomach. Acta Chir. Stand. 134: 272-274, 1968. (28) Weinberg, G.B., KuIIer, L.H., and St&r, P.A. A case control study of stomach cancer in a coal mining region of Pennsylvania. Cancer 56: 703-713, 1985. f29) Bennington, J.L., CampbeII, P.B., and Ferguson, B.R. Epidemiologic studies of carcinoma of the kidney, II. Association of renal adenocar- cinema with smoking. Cancer 22 821-823.1968. (30) Armstrong, B., Garrod A., and Doll R. A retrospective study of renal cancer with special reference to coffee and animal protein consumption. Br. J. Cancer 33: 127-136, 1976. (31) McLaughlin J.K., Mandel J.S., Blot, W.J., Schuman, L.M., MehI, E.S., and Fraumeni, J.F., Population-based casecontrol study of renal ceII car- cinoma. JNCI 72: 275-284, 1984. (32,) Wynder, E.L., Onderdonk, J., and Man@ N. An epidemiological inves- tigation of cancer of the bladder. Cancer 11: 1388-1406, 1963. &J Dunham, L.J., Rabson, A.S.. Stewart, H.L., Frank, A.S., and Young, J.L. Rates, interview and pathology study of cancer of the urinary bladder in New Orleans Louisiana. J. Natl. Cancer Inst. 41: 683-709, 1968. 57 (34) Cole, P., Monson, R.R., Haning, H., and Friedell, G.H. Smoking and cancer of lower urinary tract. N. Engl. J. Med. 284: 129-134, 1971. (35) Kabat, G.C.. Dieck, G.S., and Wynder, E.L. Bladder cancer in non- smokers. Cancer 57: 362-367, 1986. (36) Howe, G.R., Butch, J.D., Miller, A.B., Cook, G.M., E&eve, J., Morri- son, B., Gordon, P., Chambers, L.W., Fodor, G., and Winsor, G.M. Tobacco use, occupation, coffee, various nutrients, and bladder cancer. JNCI 64: 701-713. 1980. (37) Hartge, P., Hoover, R., and Kantor, A. Bladder cancer risk and pipes, cigars, and smokeless tobacco. Cancer 55: 901-906, 1985. (38) Mommsen, S., Aagaard, J., and Sell, A. An epidemiologic study of blad- der cancer in a predominantly rural district. Stand. J. Urol. Nepbrol. 17: 307-312, 1983. (39) Mommsen, S., and Aagaard, J. Tobacco as a risk factor for bladder cancer. Carcinogenesis 4: 335-338, 1983. (40) Heuch, I., Kvale, G., Jacobsen, B.K., and Bjelke, E. Use of alcohol, tobacco and coffee, and risk of pancreatic cancer. Br. J. Cancer 48: 6374x3,1983. CHEMICAL CONSTITUENTS, INCLUDING CARCINOGENS, OF SMOKELESS TOBACCO Chemical Composition of Smokeless Tobacco `RI date, at least 2,500 known compounds have been identified in pro cessed tobacco (1). Besides polysaccharides and protein tobacco con- tams Nicotzizna alkaloids (0.5-5.0 percent), alkanes (0.1-0.4 percent), &penes (0.1-3.0 percent), polyphenols (0.5-4.5 percent), phytosterols (0.1-2.5 percent), carboxylic acids (0.1-0.7 percent), aromatic hydra carbons, aldehydes, ketones, amines, amides, nitriles, N- and 0-hetero cyclic compounds, chlorinated organic compounds, alkali nitrates (0.2-5.0 percent), and at least 30 metal compounds (83). The most important habituating agent in tobacco is nicotine, the ma- jor representative of the alkaloids that constitute 0.5-5 percent of the leaf depending on the strain, variety, and agricultural practices that are employed during the tobacco cultivation. In total, the alkaloids are composed of 85 to 95 percent nicotine (4) and of other major alkaloids such as the secondary amines nornicotine, anatabine, and anabasme with lesser amounts of cotinine, myosmine, nicotyrine, 2,3'-dipyridyl, and N `-oxynicotine (5). Carcinogens in Smokeless Tobacco At present, three classes of carcinogzIls are known to occur in smoke less tobacco products: N-nitrosamines, polynuclear aromatic hydrocar- bons (PAH), and polonium-210 ( 210Po). Although chemical-analytical 58 FIGURE l.-N-Nitroeamhes in Smokeless `Ibbacco 1. Volatile Nitrosamines ' ' N-NO R----- R = -a-l, NDMA R = -C,H, NDEA 2. Nonvolatile Nitrosamines HO-CH,-Cl-l,, HO-CH+H, , N-NO NDEIA n N A0 NPYR 0 N JJO NPIP A0 NMOR H&-N-R R= -COOH NPRO R= -COOH NPIC LO R= -CH&OOH NPYRAC R= -CH&OOH NPIPAC R = -CH,-CH,-COOH NMPA R = -CH,CH,-CH,-CCOH NMBA 3. Tobacco-Specific Nitrosamhee NNN NAT NAB QLEc:3 @L":c.3 NNK NNAL data are lacking, some smokeless tobacco mixtures contain or are sus- pected to contain traces of cadmium and nickel compounds (6). formal- dehyde, and coumarin, all of which are known animal carcinogens (78). N-Nitroeamhea Tobacco leaves contain an abundance of amines in the form of pre teins and alkaloids. `lbba~ also contains up to 5 percent nitrates and traces of nitrite. Thus there is the potential for the formation of N-nitrosamines from the nitrate, nitrite, and amines during the process- ing of smokeless tobacco products. In tobacco, we distinguish between volatile nitrosamines, nonvolatile nitrosamines, and tobacuxqAfic nitrosamines (figure 1). With the exception of some N-nitrosamino 59 FIGURE 2.-Formation of `Ibbacco-Specific Nitrosamhes NICOTINE NORNICOTINE ANARASINE ANATABINE NITROSATION NNAL NNK NNN NAB NAT acids, the nitrosamin es in tobacco are animal carcinogens that are formed after harvesting of the tobacco during curing, fermentation, and/or aging. The N-nitrosamin o acid, N-nitrosoproline, occurs in pro- cessedfoodandcanalsobeformedinhumans by endogenous nitrosation of proline. This nitrosamino acid is not carcinogenic on the basis of pres- ently available data (912). Table 1 s ummarizes the available data for the volatile nitrosamin es in smokeless tobacco. Only one of the volatile nitrosamines, NDMA, has been found in U.S. looseleaf tobacco, but four nitrosamines have been found in American snuff. N-Nitrosomor- pholine is formed during tobacco processing or aging from morpholine, a cyclic amine that is not known to occur in uncontaminated tobacco (13,14) but originates from packing materials and/or flavor additives. Table 2 lists the presently known nonvolatile nitrosamines in smokeless tobacco. N-Nitrosod.iethanolamine (NDELA) in U.S. tobacco originates primarily from residues on tobacco leaves of the sucker-growth inhibi- tor maleic hydrazidediethanolamine (MH-30). Use of this formulation of the agricultural spray was banned in the United States in 1981, and the concentration of NDELA in smokeless tobaccos has markedly de creased since then (14,15). Figure 2 presents the formation of the tobaccospecific N-nitrosamines (TSNA) from the alkaloids. There is progressive nitrosation of the alka- loids during curing and processing and even during the shelf life of the commercial products (16). Table 3 summar&es the presently available quantitative data for four out of five TSNA's that are present in smoke less tobacco. The nitrosamines are detectable in snuff and tobacco prod- ucts from various parts of the world. Analyses of Swedish snuff brands manufactured between 1980 and 1985 have revealed a significant decrease of the levels of TSNA; such a trend has not been observed for U.S. snuff brands (14,16,17). It has been suggested that the lowering of TSNA levels in Swedish snuff brands is due to better control of the bac- terial content of the tobacco products. Reduced bacterial activity will probably reduce nitrite levels and, consequently, inhibit nitrosamine 60 TABLE l.-Volatile Nitroaminea in Smokeless `Ibbam (ppb)* Pruduct NDMA NPYR NPIP NMOR Fleference U.S. Looseleaf t Snuff ND - 380 (4) ND - 215 (26) ND - 1.2 (4) ND - 291 (16) ND (4) ND - 2.5 (4) ND - 107 (16) ND-696 (26) 1q14,1794 13y4,17,20, 29,%37 Sweden Chewing Tobacco Snuff ND - 0.6 (4) ND-60 (53) 0.9 - 3.7 (4) ND - 210 (27) ND (2) ND - 0.8 (2) 1736 ND - 0.5 (37) ND - 1.2 (53) 14,17,36 Canada Snuff 23 - 72.8 (2) 321 - 337 (2) 14 Denmark Chewing Tobacco Norway Chewing Tobacco India Chewing Tobacco U.S.S.R. Nass$ ND - 8.6 (6) 7.0 - 25.5 (6) ND (4) ND - 32.8 (6) 17.36 84.0 - 280 (2) 2.8 - 15 (2) 28-37 (2) 37 -220 (2) 17 ND - 0.56 (4) ND (4) 14 ND (4) 14 1.55 - 4.48 (4) ND (4) 1.74 - 8.82 (4) * Number m parentheses. number of samples analyzed t One sample also contained 8.6 ppb NDEA. e $ Also contained ND 69.6 NDEA /Ml. i? TABLE 2.-Nonvolatile Nitr osamines in Smokeless Tobacco (ppb)* Tobacco product NDELA NMPA NMBA NPRO NPYRAC NPIC NPIPAC Reference U.S. Looseleaf 224 - 680 (3) Snuff 160 - 6,800 1,250 - 7.420 (13) (51 Sweden Snuff 230 - 390 510 _ 4,400 (8) (12) Canada Plug Tobacco 110 (1) Snuff 1,180 - 2,720 (3) Germany Plug Tobacco 50 (2) Belgium Chewing 1,600 (1) Tobacco U.S.S.R. Nass India Chewing Tobacco 40 (4) 30 - 110 (4) 450 - 463 (2) 120 - 2.240 500 _ 50.900 ND - 2,000 (5) (13) (5) ND _ 260 890 - 29,500 100-300 (12) (12) (5) 100 (1) 8,800 - 16,600 (2) 500 - 700 (2) 100 (1) 3,300 (1) 200 (1) ND - 180 (4) 190 - 410 (4) ND - 6.100 (5) ND - 5,560 (12) 100 (1) 13J4p.M ND - 1,500 1315,34, 3Jm 100 - 200 14,15,*40 (5) 14 14 14 200 (1) 40 14 14 o Number in parentheses. number of samples analyzed TABLE 3.-`Ibbacco-Specific N-Nitrosamhes in Smokeless Tobacco (ppb)' Product NNN NNK NAT NAB Reference U.S. Looseleaf Plug lbbacco Snuff Sweden Snuff Plug Tobacco Canada Snuff Norway Snuff Denmark Snuff Chewing Tobacco Germany Plug lbbacco Snuff U.S.S.R. Nass Indii Chewing Tobacco Belgium Chewing Tobacco 620-8.200 (9) 3,400-4,300 D) 1,600-135,000 (21) 3.050-154.000 (34) 350-2,090 (3) 50,420-79,100 (2) 13,000-29,000 (2) 4.460-8,000 (3) 210-1.400 (4) 1,420-2,130 (2) 6,080-6,700 (2) 120-520 (4) 470-2,400 (5) 7,380 (1) ND-380 (4) 130-2,300 (5) loo-13,600 (21) 1,560-338,000 (21) 510-2,950 (34) ND-240 (3) 3,200-5,800 (2) 1,600-21,400 (34) 6%1,580 (3) 152.000-170,000 (2) 2,700-3,900 (2) 9,100-16,000 12) 1,350-7,030 (3) ND-210 (4) 30-40 (2) 1,500-1,540 (2) 20-130 (4) 130-230 (4) 970 (1) 2,680-6.170 300-2.80-O 1:; 330-500 (2) 3,920-4.370 (2) 32-300 (4) 300-450 (4) 130 (1) ND-140 (5) M-6,700 (12) 110-150 (19) ND-100 (3) 4,000-4,800 (2) l,OOO-2.400 (2) ND-60 (4) 30-50 (2) 8-30 (4) 30-70 (4) 14,17,41,42 43 614,16;17,384243 14.1~1738 l/17 14 17 16 17 14 16 14 14,41 38 E * Number in parentheses. number of samples analyzed. TABLE I.-JMimated Exposure of U.S. Residents to Nitmsamines* source of Exposure Nitrosamines Primary Exposlue Daily Intake Route ug/Person Beer Cosmetics Cured Meat; Cooked Bacon Scotch Whiskey Cigarette Smoking NDMA NDELA NPYR NDMA VNAt NDELA NNN NNK NAT+NAB Ingestion Dermd Absorption Ingestion Ingestion Inhalation Inhalation Inhalation Inhalation Inhalation 0.34 0.41 0.17 0.03 0.3 ii:: 2.9 > 16.2 1.2 Snuff DippingS VNA Ingestion 3.1 NDELA Ingestion 6.6 NNN Ingestion 75.0 NNK Ingestion 16.1 NAT+NAB Ingestion 73.4 > 164.5 o From the National Research Council /lgl. amended by data for snuff dipping 1131. In addition. it has been est.& lished that u NDMA and r n inhalation of the air in cars with new leather upholstery daily exposure amounts to 0.50 yg of .20 ug of NDEA 1181. t VNA. NDMA + NEMA + NDEA + NPYR 1371. 1 Bnmnemann et al. 113/; average values from the leading five U.S. finecut tobaccos used for snuff dipping in 1981; assumed daily consumption 10 g/day of snuff: VNA = NDMA + NPYR + NMOR. formation (17). NNK and NNN are powerful carcinogens in mice, rats, and hamsters, NAB is moderately carcinogenic, and NAT is inactive in rats in doses up to 9 mmolkg (table 3, page 82) 13). The daily exposure of an "average" snuff dipper to carcinogenic N-nitrosamin es exceeds by at least two orders of magnitude the esti- mated exposure of U.S. residents to nitrosamines in products other than tobacco products (table 4) (18,19). Furthermore, the concentrations of carcinogenic nitrosamines in snuff exceed very signifi~tly the per- missible limits for individual nitrosamines in consumer products (table 5). During snuff dipping or chewing of tobacco, the TSNA's are ex- tracted by the saliva. Consequently, the saliva of snuff dippers is reported to contain 5.0-420 ppb of NNN, up to 96 ppb of NNK, and 6.6-555 ppb of NAT (16). The saliva analyses of Indian tobacco chewers showed the presence of 1.2-220 ppb of NNN, 3.2-51.7 ppb of NAT, and up to 2.3 ppb of NNK (20,21). Recently, three additional TSNA's have been isolated from U.S. commercial snuff: 4-(methylnitrosamino)-l- (3-pyridyhbutanol-1 (NNAL), 4-(methylnitrosam.ino)-1-(3-pyridyl) butenel (NNO), and 4-(methytitrosamino)-4-(3-pyridyl)butanol-1 (Red NNA) (figure 3) (22). Additional amounts of TSNA's are most likely also formed by nitrosation processes that occur in the oral cavity during chewing (1422,23). 64 TABLE 5.-Permissible Limits for Individual N-Nitrosamines in Consumer Products Product Permissible Limit wb Whz) Agency Bacon (Meat) Beer Rubber Nipples of Baby Bottles 5 USDA* 5 FDA? 10 FDA$ Range of Individual Nitrosamines Present in Snuff Tobaccos ppb bdcd NNN 5,800 - 64.000 NNK 100 - 3,100 NAT 3,300 - 215,000 NAB 200 - 6,700 NDELA 160 - 6,800 Range in the leading 5 US. brands (1984-85) Range in 13 U.S. brands (1980-1985) o No "confirmable levels of mtrosammes' (441. t Regulation set for N-nitroscdimethylsminine 14.9 $ Regulation set for any mdividual volatde N-nitrosamine (461 Polynuclear Aromatic Hydrocarbons A number of naphthalenes have been identified in processed tobacco and especially in Latakia, which is flavor enriched by treatment with wood smoke (24,,25). While smoking tobaccos were found to contain 300-5,000 ppb of phenanthrene, 1 N-4.200 ppb of anthracene, 76-1,800 ppb of pyrene, 15-14,000 ppb of fluoranthene, and 8.5 ppb of benzo(a)pyrene (BaP) (26,27), analyses of British snuff in 1957 showed levels of 260 ppb of pyrene, 335 ppb of fluoranthene, and 72 ppb of BaP (28). In the five most popular snuff brands in the United States that were analyzed in 1985, BaP ranged from C 0.1 to 63 ppb (29). Polonium-210 This alphaemitting element has long been incriminated as a human carcinogen (30). The levels of 21oPo in dozens of U.S. and foreign cigar- ette tobaccos were between 0.1 and 1.0 pCi/g (31). In recent samples of the five leading U.S. snuff brands, 21oPo ranged from 0.16 to 1.22 pCi/g (29). It appears that 21oPo in tobacco leaves stems partially from certain types of fertilizers and airborne particles that are taken up by the tri- chomes (glandular hair) of the tobacco leaf (3133). Summary In processed tobacco, more than 2,550 chemical compounds have been identified. Among these are traces of known carcinogens such as 65 FIGURE 3.-`Ibbaax Specific N-Nitrwamines in Snuff U.S. Brands, 1985 Nitmmamines Relative Carchogenicity in Rats* Concentration ia Snuff We) (Dry Weight) A B NNN CP Q Co NAB cl" Q io NAT OH CH3 h-NO +++ 3.3 64 + -t 44 215 ybi 3 NO +++ + ? ? 1.1 6.7 1.8 0.3 trace$ 1.3 3.1 0.14 trace* 1.8 o + + + `Xmors with 1 mm&kg: + tumors with 9 mmollkg; (for type of tumors induced se table 4. page 381; + insignificant number of tumors with 9 mmollkg: ? not tested. t IsoM.ed amount.3 only. $ < 0.01 upig. PAH, 21OP0, and N-nitrosamines. The most prevalent organic carcine gas are the t.obawespecific N-nitrosamines that are formed from the Nicotiunu alkaloids during the processing of tobacco leaves. Their con- centrations in snuff exceed the levels of nitrosamines in other consumer products by over one hundredfold. During snuff dipping or chewing of tobacco, the nitrosation process continues within the mouth stimulated by oral bacteria. 66 References (1) Dube, M.F., and Green, CR. Methods of collection of smoke for analyti- cal purposes. Recent Advan. Tobacco Sci. 8: 42-102, 1982. (2) Wynder, E.L., and Hoffmann D. Tobacco and tobacco smoke. Studies in Experimental Carcinogenesis. New York, Academic Press, 1967, p. 730. (3) International Agency for Research on Cancer. Tobacco habits other than smoking Betel&d and areca-nut chewing and some related nitro samines. IARC Monogr. EvaI. Carcinog. Risk Chem. Hum. 37: 291,1985. (4) Sisson, V.A., and Serverson, R.F. Alkaloid composition of the Nicotianu species, 24. Tobacco Chemists Res. Co& 1984, p. 10. (5) Piade, J.C., and Hoffmann, D. Quantitative determination of aIkaIoids in tobacco by liquid chromatography. J. Liquid Chromatogr. 3: 1505-1515, 1980. (6) Baumslag, N., Keen, P., and Petering, H.G. Carcinoma of the maxillary antrum and its reiationship to trace and metal content in snuff. Arch. En- viron. Health 23: l-5, 1971. (7) Vainio, H., Hemminke, K., and Wilbur-n, M. Data on the carcinogenicity of chemicals in the IARC Monographs Programme. Carcinogenesis 6: 1663-1665, 1985. (8) Grigg, G.W. Genetic effects of coumarins. Mut. Res. 47: 161-181, 1977178. (9) Ohshima H., and Bartsch, H. Quantitative estimation of endogenous nitrosation in humans by monitoring N-nitroso proline excreted in the urine. Cancer Res. 41: 3658-3662, 1981. (10) Hoffmann D., and Brunnemann K.D. Endogenous formation of N-nitroso proline in cigarette smokers. Cancer Res. 43: 5570-5574,1983. (11) International Agency for Research on Cancer. Monograph on the evaluation of the carcinogenic risk of chemicals to humans, Vol. 17. Some N-nitroso compounds. Lyon, France, International Agency for Research on Cancer, 1978, p. 365. (12) Preussn-mm R., and Stewart B.W. N-nitroso carcinogens. In: C.E. Searle ted.). Chemical Carcinogens, Second Edition. Am. Chem. Sot. Monogr. 182: 643-828,1984. (13) Brunnemann, K.D., Scott, J.C., and Hoffmann, D. N-NitrosomorphoIine and other volatile N-nitrosamines in snuff tobacco. Carcinogenesis 3: 693496,1982. (14) Brunnemann, K.D., Genoble, L., and Hoffmann, D. N-nitrosamines in chewing tobacco: An international comparison. J. Agr. Food Chem. 33: 1178-1181, 1985. (25) Brunnemann, K.D., and Hoffmann, D. Assessment of the carcinogenic N-nitrosodiethanolamine in tobacco products and tobacco smoke. Car- cinogenesis 2: 1123-l 127, 1981. (16) Hoffmann, D., and Adams, J.D. Carcinogenic tobacco-specific N-nitrosamines in snuff and in the saliva of snuff dippers. Cancer Res. 41: 4305-4308, 1981. 67 (17) ijsterdahl B.G., and Slorach S. N-nitrosamines in snuff and chewing to bacco on the Swedish market in 1983. Food Additiv. Contamin. 1: 299-305, 1984. (18) National Research Council. In: The health effects of nitrate, nitrite and N-nitroso compounds (Ch. 7, Pt. 1). Washington, D.C., National Academic Press, 1981, p. 51. (19) Hoffmann, D., and Hecht, S.S. Nicotinederived N-nitrosamines and tobacco-related cancer. Current status and future directions. Cancer Res. 45: 935-944, 1985. (20) Nair, J., Ohshima, H., Friesen, M., Croisy, A., Bhide, S.V., and Bartsch, H. `Ibbaccospecific and betel nut-specific N-nitroso compounds. Occur- rence in saliva and urine of betel quid chewers and formation in vitro by nitrosation of betel quid. Carcinogenesis 6: 295-303, 1985. (21) Wenke, G., Rivenson, A., Brunnemann, K.D., and Hoffmann, D. For- mation of N-nitrosamines during betel quid chewing. IARC Sci. Publ. 57: 859-866, 1984. (22) Brunnemann, K.D., Chou, D.. Adams, J.D., and Hoffmann, D. On the isolation and identification of new tobaccospecific N-nitrosamines (Abstract). 39th Tobacco Chemists Res. Conf., Montreal, October 2-5, 1985. (a) Sipahimalani, A.T., Chada, M.S., Bhide, S.V., Pratap, A.I., and Nair, Y. Detection of N-nitrosamines in the saliva of habitual chewers of to bacco. Food Chem. Toxicol. 22: 261-264, 1984. (24) Schmeltz, I., Tosk, J., and Hoffmann, D. Formation and determination of naphthalenes in cigarette smoke. Anal. Chem. 48: 645-650,1976. (2.5) NicoIaus, G., and Ehnenhorst, H. Nachweis and quantitative Bestim- mung von AIkyInaphthaIinen in Latakia-?gbak. Beitr. lbbakforsch. 11: 133-140, 1982. (26J Campbell, J.M., and Lindsey, A.J. PolycycIic hydrocarbons extracted from tobacco: The effect upon total quantities found in smoke. Br. J. Cancer 10: 649-652, 1956. (27) Onishi, I., Nagasawa, M., Tomita, H., and Fukusumi, T. Studies on the essential oil of tobacco leaves, Part XVI. Neutral fraction (3). PolycycIic aromatic hydrocarbons of Burley tobacco leaf. Bull. Agr. Chem. Sot. Japan 22: 17-20.1958. (28) Campbell, J.M., and Lindsey, A.J. PolycycIic aromatic hydrocarbons in snuff. Chem. Ind. London, 951,1957. (29) Hoffmann, D., Harley, N.H., Fisenne, I., Adams, J.D., and Brunnemann, K.D. Carcinogenic agents in snuff. JNCI (in press). (30) Lundin, F.E., Jr., Wagoner, J.K., and Archer, V.E. Radon daughter exposure and respiratory cancer. Quantitative and temporal aspects. U.S. Dept. of Health, Education, and Welfare; Joint NIOSH/NIEHS Monogr. 1, 1971. (31) Harley, N.H., Cohen, B.S., and `Iso, T.C. Polonium-210: A questionable risk factor in smoking-related carcinogenesis. Banbury Report 3: 93-104. 1980. 68 (32) Martell, E.A. Radioactivity of tobacco trichromes and insoluble cigar- ette smoke particles. Nature 249: 215-217, 1974. (33) `l%o, T.C., Harley, N.H., and Alexander, L.T. Source of lead-t&in and polonium-t&in. Science 153: BBO-882,1966. (34) Brunnemann. K.D., Scott, J.C., and Hoffmann, D. N-Nitrosoproline, an indicator for N-nitrosation of amines in processed tobacco. J. Agr. Food Chem. 31: 905-909, 1983. (35) Palladino, G., Adams, J.D., B runnemann, K.D., Haley, N.J., and Hoffmann, D. Snuff-dipping in college students: A clinical profile. Mili- tary Med. (in press). (36) &terdahl, B.G., and Slorach, S.A. Volatile N-nitrosamines in snuff and chewing tobacco on the Swedish market, Food Chem. lbxicol. 21: 759-762, 1983. (37) Brunnermum, K.D., Yu, L., and Hoffmann, D. Assessment of carcino genie volatile N-nitrosamines in tobacco and in mainstream and sidestream smoke from cigarettes. Cancer Res. 37: 3218-3222, 1977. 08) Nair, J., Ohshima, H., Malaveille, C., F riesen, M., Bhide, S.V.. and Bartsch, H. N-Nitroso compounds (NOC) in saliva and urine of betel quid chewers: Studies on occurrence and formation. Carcinogenesis 6: 295-303, 1985. (39) Ohshima, H. Identification and occurrence of new N-nitrosamino acids in human urine and environmental samples. Presented at the Con- ference on Organic and Biological Chemistry of Carcinogenic and Car- cinostatic Agents Containing Nitrogen-Nitrogen Bonds, Harper's Ferry, West Virginia, May 17-21, 1985. (40) Oh&ma, H., Nair, J., Bourgade, M.C., Friesen, M., Ganeen, L, and Bar&h, H. Identification and oczaurence of two new N-nitrosammo acids in tobacco products: 3jN-nitrosoN-methyiaminoylamino)propionic acid and 4-(NnitxosoNmethylaminolbutyric acid Cancer L&t. s 153-162,1985. (41) Hoffmann, D., Hecht, S.S., Omaf, R.M., Wynder, E.L., and `lbo, T.C. Nitrosonomiwtine: Presence in tobacco, formation and carcinogenicity. IARC Sci. Pubi. 14: 307-320,1976. (42) Munson, J.W., and Abdine, H. Determination of N-nitrosonomiwtine in t.obacco by gas chromatography/mass spectroscopy. Anal. Letters 1Q 777-786, 1977. (43) Adams, J.D., Bnmnemann, K.D., and Hoffmann, D. Rapid method for the analysis of tobaccospecific N-nitrosammes by gas-liquid chroma- tography with a thermal energy analyzer. J. Chromatogr. 256: 347-351, 1983. (44) U.S. Department of Agriculture, Food Safety and Quality Service. Nitrates, nitrites and ascorbates (or isoascorbah) in bawn. Fed. Reg. & 20992-20995, May 16,1978. 1&J U.S. Food and Drug Administration. Dimethyl nitrosamine in malt beverages; availability of guide. Fed. Reg. 45: 39341-39342,198O. (4s) U.S. Food and Drug Administr h a `on. Action levels of total volatile N-nitrosammes in rubber baby bottle nipples; availability of revised wm- phance policy guide. Fed. Reg. 49 5078450790,1984. 69 Abbreviations BaP NAB NAT ND NDEA NDELA NDMA NMBA NMOR NMPA NNAL NNK NNN NNO NPIC NPIP NPIPAC NPRO NPYR NPYRAC PAH 2lOPo Red NNA TSNA Benzo(a)pyrene N `-Nitrosoanabasine N `-Nitrosoanatabine Not detected Nitrosodiethylamine Nitrosodiethanolamine Nitrosodimethylamine Nitrosomethylbutyric acid Nitrosomorpholine Nitrosomethylpropionic acid 4-(Methyhr.itrosamino)-1-(&pyridyl)-1-butanol 4-(Methyhritrosamino)-l-(3-pyridyh-1-butanone N `Nitrosonomicotine 4-(Methylnitrosamino)-l-(3-pyridyl)butenel Nitrosopipecolic acid Nitrosopiperidine Nitrosopiperidine-acetic acid Nitrosoproline Nitrosopyrrolidine Nitrosopyrrolidineacetic acid Polynuclear aromatic hydrocarbons Polonium-2 10 4-(Methyhritrosamino)-4-(3-pyridyh-1-butanol Tobaccospecific nitrosamines METABOLISM OF CONSTITUENTS OF SMOKELESS TOBACCO The tobaccospecific nitrosamines 4-(methylnitrosamino)-l-(3-pyridyl)- 1 -butanone (NNK) and N `-nitrosonornicotine (NNN) are quantitatively the major known carcinogens that are present in snuff and other types of smokeless tobacco. Molecular changes that are induced in the genetic material of tobacco chewers are most likely to arise from the metabo lism of these two nitrosamin es. Although present in similar quantities, N'nitrosoanabasine (NAB) and N'-nitrosoanatabine (NAT) are less car- cinogenic than NNK and NNN and are less likely to play an important role in the induction of oral cancer in man. Some snuff products contain considerable amounts of N-nitrosomorpholine (NMOR) and N-nitro 70 FIGURE l.-Metabolic Pathways of NNK sodiethanolamine (NDELA); the former is a potent carcinogen. The levels of benzo(ab>yrene (BaP) and 2loPo in snuff tobacco are low com- pared to those of the nitro samines (see previous section). This section will focus on the routes of metabolic activation of the compounds that are most likely to be involved in the induction of tumors that are related to snuff use-NNK, NNN, and NMOR. Metabolism of NNK The overall metabolic scheme for NNK, as determined by in uivo and in vitro studies in F-344 rats, Syrian golden hamsters, and A/J mice, is illustrated in figure 1 (l-4). A key feature of this metabolic scheme is the conversion of NNK to the alpha-hydroxy intermediate 4, which is un- stable and undergoes spontaneous conversion to the keto aldehyde 8 and, most likely, methyl diazohydroxide 9. The latter is a methylating agent that is well known for its ability to methylate DNA forming 7-methylguanine, 06-methylguanine, 4-methylthymidine, and a spec- trum of other products (5). Among these, O&methylguanine, which is generated from precursors such as N-methyhritrosourea (NMU) or N-nitrosodimethylamine, has been unequivocally shown to be able to in- duce miscoding during DNA replication, and the resulting point muta- tion is sufficient to activate proto-oncogenes (6,7). Many studies have demonstrated a correlation between 06-methylguanine persistence in rep licating tissues and the initiation of the carcinogenic process, although it is clear in other cases that additional factors are also involved (89). 71 FIGURE 2.-Scheme Linking Nicotine to Formation of the Promutagenic DNA Adduct, 06-Methylguanine TOBACCO PROCESSING METABOLIC OR PCTIVbTIOI(- [CHSN=NOtl] - 7-YETWLGUANINE CIGARETTE YETHYLOIAZO- 06-YETHYLGUANINE SYOKING HYDROXlDE IN DNA NICOTINE NNK - Recent studies have demonstrated that NNK can methylate target tissue DNA of rats; 7-methylguanine and 05methylguanine have been detected in the DNA of rat lung, nasal mucosa, and liver but not in the nontarget tissues, kidney, and esophagus (1014). These studies have also shown that, in the case of NNK, 06-methylguanine formation alone is not sufficient for tumor induction since persistent levels of 06-methyl- guanine in the lung were less than those observed upon treatment with equivalent quantities of N-nitrosodimethylamine, but the latter did not induce lung tumors (13). It is clear from these, and related studies with NNN, that DNA adducts are also formed via pyridyloxobutylation or related processes. Regardless of the mechanism, it is significant that NNK causes DNA methylation; this creates a mechanistic link between nicotine, the habituating factor in tobacco, and 08methylguanine for- mation in DNA, as illustrated in figure 2. Immunoassay methods are currently being developed to detect 06-methylguanine in the exfoliated oral cells of snuff dippers. Its presence can be inferred from the animal studies that are discussed above and by the demonstration that human tissues, including buccal mucosa, can metabolize NNK by alpha- hydroxylation (15). In this respect, it is significant that injection of Syrian golden hamsters with the methylating agent MNU, combined with irritation of the buccal mucosa, resulted in the induction of oral cavity tumors (16). The pathway of NNK metabolism leading to the alpha-hydroxy inter- mediate 3 is also considered to be important in NNK carcinogenesis. This pathway gives rise to the electrophilic diazohydroxide 7. The prop erties of this intermediate have been investigated by using a model compound, 4-(carbethoxynitrosamino)-l-(3-pyridyl)-l-butanone (CNPB). Generation of 7 from CNPB is strictly analogous to the well- known ability of NMU to generate methyl diazohydroxide. Mutagen- icity assays in S. typhimurium of CNPB have shown that it is more mutagenic than NMU (17). Chemical model studies have demonstrated that it modifies the NZ-position of deoxyguanosine (18). This adduct and other adducts that may be formed from the diazohydroxide 7 and related intermediates are likely to play an important role in tumor in- duction by NNK. Autoradiographic studies have demonstrated that radioactivity from [carbonyl-14C]NNK is firmly bound to target tissues of rats and hamsters (4,19) and to tissues of the marmoset monkey (20). 72 FIGURE 3.-Metabolic Pathways of NNN A third key feature of NNK metabolism is its rapid conversion in vivo and in cultured tissues from experimental animals and humans to its reduced form, NNAl, which has similar tumorigenic activity to that of NNK (1,34,15,21). NNAl is slowly metabolized as indicated in figure 1 and also by reconversion to NNK. Like NNK, it methyl&es DNA in vitro and in rho. While the full details of the NNK-NNAl equilibrium have not yet been elucidated, it is clear that NNAl can act as a cir- culating source of NNK metabolites. It may play an important role in tissuespecific carcinogenesis by NNK. Metabolism of NNN Metabolic pathways of NNN are illustrated in figure 3. These path- ways have been elucidated by in uiuo and in vitro studies in rats, hamsters, and mice @,3,22-29). The stable met&o& NNN-l-N-oxide (1) has tumorigenic activity somewhat less than that of NNN but is still an effective carcinogen in F-344 rats (30). Metabolism of NNN to the 2'- and 5 `-hydroxy intermediates 2 and 5 constitutes a major pathway in uiuo and in vitro in experimental animals, human liver microsomes (31), and cultured human tissues, including buccal mucosa (15). Of particular interest is the ability of two NNN target tissues, lingual mucosa and esophageal mucosa, to carry out preferential 2 `-hydroxylation of NNN @7,32). The intermediate that is formed by 2'-hydroxylation of NNN is diazohydroxide 8, which is identical to that formed by methyl hydroxy- lation of NNN (7, figure 1). As described above, this intermediate is 73 FIGURE I.-Metabolic Pathways of NMOR IO II highly mutagenic, and this or related intermediates appear to play an important role in carcinogenesis by both NNN and NNK. The inter-me diate 9 is significantly less mutagenic than 8 in S. typhimutium (33). and various lines of evidence indicate that it is less important in NNN tumorigenesis than is 8 (33,3#. Autoradiographic studies have demon- strated that radioactivity from [2'-14C]NNN is bound to tissues of mice, rats, and marmoset monkeys (ZYJ35-37). Immunoassays are currently being developed for the putative DNA adducts that are produced by 2 `hydroxylation of NNN and methyl hydroxylation of NNK; it will be important to assess the levels of these adducts in the exfoliated oral cells of snuff dippers. Their levels may relate to the susceptibility of in- dividuals to the effects of smokeless tobacco. The metabolic pathways that lead to these intermediates can be affected by alcohol consumption and dietary components &?2,38#3). Metabolism of NMOR The metabolic pathways of NMOR are illustrated in figure 4. These have been elucidated by in vitro and in uiuo studies in rats (44-47). Struc- ture activity studies bad shown that 3-hydroxylation of NMOR, leading 74 to intermediate 4, was likely to he important in NMOR carcinogenesis (48). This pathway could result in the formation of glyoxaldeoxyguano sine adducts (49); 2-hydroxylation of NMOR also occurs, giving the mutagenic product 2. The latter also forms glyoxaldeoxyguanosine adducts (50). These adducts, which are likely to have miscocling proper- ties, also should he present in the DNA of snuff dippers since human tissues are capable of metabolizing NMOR 151). Summary Persuasive evidence exists that the carcinogenic nitrosamines that are present in high quantities in snuff and other forms of smokeless to bacco are metabolized by target tissues of experimental animals and by human tissues to intermediates that can modify the genetic material of the cell. References (1) Hecht, S.S., Young, R., and Chen, C.B. Metabolism in the F-344 rat of 4-(N-methyl-N-nitrosamino)-I-(3-pyridyl)-l-butanone. a tobacco specific carcinogen. Cancer Res. 40: 4144-4150.1980. (2) Hoffmann, D.. Castonguay, A., Rivenson, A., and Hecht, S.S. Compar- ative carcinogenicity and metabolism of 4-(methyl-nitrosamino)-l- (3-pyridyl)-l-butanone and N'nitrosonornicotine in Syrian golden hamsters. Cancer Res. 41: 2386-2393, 1981. (3) Castonguay, A., Lin, D., Stoner, G.D., Radok, P., Furuya, K., Hecht, S.S., S&t, H.A.J., and Klaunig, J.E. Comparative carcinogenicity in A/J mice and metabolism by cultured mouse peripheral lung of N'-nitroeonornicotine, 4-(methyhritrosamino)-l-(3-pyridyl)-1-butanone and their analogues. Cancer Res. 43: 1223-1229,1983. (4) Castonguay, A., Tjjilve, H., and Hecht, S.S. Tissue distribution of the tobacco-specific carcinogen 4-(methylnitrosamino)-l-(3-pyridyl)-l- butanone, and its metabolism in F-344 rats. Cancer Ftes. 43: 630638, 1983. (5) Singer, B., and Grunberger, D. Molecular biology of mutagens and car- cinogens. New York, Plenum Publishing Corp., 1983, pp. 45-96. (s) Loechler, E.L.. Green, C.L., and Essigmann, J.M. In viva mutagens by O~methylguanine built into a unique site in a viral genome. Proc. Natl. Acad. Sci. USA 81: 6271-6275,1984. (7) Sukumar, S., Nofario, V., Martin-Zanca, D., and Barbacid, M. Induc- tion of mammary carcinomas in rats by nitrosomethylurea involves malignant activation of H-ras-1 locus by single point mutations. Na- ture 3iX? 658-662,1983. (8) Pegg, A.E. Methylation of the 06-position of guanine in DNA is the most likely initiating event in carcinogenesis by methylating agents. Cancer Invest. 2: 223-231. 1984. 75