Tobacco Documents Online

retention time of NAtB and for NNN and NNK. After the first enrichment of the nonvolatile nitrosamines, as described above, the concentrate was evaporated to dryness, dissolved in dichloro- methane and chromatographed on 500 g of basic alumina (Woelm, activitiy II-III). Those fractions which gave positive responses with the HFLC-TEA at the retention times for NAtB, NNN and NNK, were combined and concentrated to =0.3 ml. For the identification of NAtB, NNN and NNK in the tobacco extract, about 10 u1 of this concentrate was injected into the GLC-MS system. For the unambiguous identi- fication of NAtB, NNN and NNK in smoke, it was necessary to rechromatograph the final concentrate (=0.3 ml) on thin layer chromatography plates (silica gel, 50 p; 10% methanol in chloro- form) and to extract the respective bands and analyze the concentrated extracts by GLC-MS. The GLC separation of NNN and NNK was achieved at 220° on a 1.8 m glass column filled with 10% Carbowax 20M-TPA on Gas-Chrom Q (injection port 250°), and that of NAtB was achieved at 200° on a 3.5 m glass column filled with 10% UCW-98 on Gas-Chrom 0 (injection port 250°). RESULTS AND DISCUSSION, Chart 3 represents a high performance liquid chromatogram of concentrates with tobacco specific nitrosamines detected.by W - 10 -

/ the TEA. The peaks were identified by comparison with the retention times of reference compounds and by mass spectral analyses of concentrates from 100 g of chewing tobacco or from the mainstream smoke of 200 cigarettes with- out filter tips. As in earlier studies (1L), we did not detect NAB, likely because the precursor anabasine is only a minor alkaloid, whereas, anatabine is the most abundant secondary alkaloid in the major tobacco varieties (7,19). ' 2.7% by anatabine, 0.96% by nornicotine and 0.35% by Recently it was reported that about.94.5% of the alkaloid fraction of cigarette tobacco was represented by nicotine, anabasine (7). Chart 4 compares mass spectra of synthetic NAtB and NAtB isolated from chewing tobacco and from cigarette smoke. The presence of Niz1tB was suggested earlier by Klus and Kuhn who found in a thin layer chromatogram a compound with a- retention time of this nitrosamine (25). Evaluation of the carcino- genic activity of NAtB is currently in progress in bioassays with mice and rats. The quantitative values obtained for nonvolatile nitros- amines with a HPLC-thermal energy analyzer system could possibly be inflated, or reduced by the presence of unknowns.in the con- centrates. One must consider such interferences since tobacco smoke contains more than 2,000 known compounds and since merely a short enrichment procedure was used for the nonvolatile nitros- amines. The presence of unknowns in the concentrate may affect the signal originating from the excited 110 2 which derived from nitrosamimes by cleavage and ozoni~zation. In order to clarify this point, we analyzed NNN fron the mainstream smoke of one 14 type of cigarette four times with [ C]\".V as a; standaxl. After (v V 1 1 '' 4 - 11 - ~'

...QS. the short standard enrichment procedure, we recorded the average value from`four runs at 0.21 + 0.025 ug NNN/cigarette. After this analysis, which required less than 38 of the sample we enriched the NNN from the concentrate by 2 successive coluan chromatographies on basic alumina. The average NNN value from 4 runs, after one column (enrichment step 1) was 0.21 + 0.020 pg/cigarette and after two columns (enrichment step 2) 0.20 + 0.024 pg/cigarette (Table 1).Thus we concluded that the quantita- tive values for NNN and, most likely also,for the two other tobacco specific nitrosamines obtained after the short enrich- ment step and by the HPLC-thermal energy analyzer, do in fact represent the actual am'ount present in the smoke concentrate. In Table 11, we have summarized the findings for tobacco specific N-nitrosamines in tobacco, mainstream smoke and side- stream smoke. These data indicate that tobacco and smoke of little cigars and cigars, as well as of cigarettes made entirely of Burley tobacco are richer in tobacco-specific nitrosamines than in tobacco and in smoke from cigarettes made of Bright and blended tobaccos. As was shown earlier, Burley and cigar tobaccos are rich in nitrate (23') and thus generate during smok- ing relatively high amounts of nitrogen oxides. The smoke of 1000121525 - 12 -

these tobacco products also has pH values of 6.5 and above, especially for the last puffs and thus contains •unprotonated . nicotine and other tobacco alkaloids (4). The unprotonated form of tobacco alkaloids is at least partially present in the gas phase and not just in the particulate phase as in the smoke of Bright or blended tobaccos(pH'<61. High levels of. nitrate in tobacco and nitrogen oxides in the smoke and elevated pH appear to favor the formation of nonvolatile N-nitros- amines in the smoke as was previously observed for volatile nitros- amines (6). presently,we are exploring this ooncept experimentally. Whereas most of the volatile nitrosamines are formed during smoking by pyrosynthesis*(6), some tobacco-specific nitrosamines transfer directly from tobacco into smoke (13). The influence of smoke pH, and the concentrations of nitrogen oxides and tobacco alkaloids on the formation of these tobacco specific carcinogens must be fully studied in order to permit approaches towards the reduction of these carcinogenic nitrosamines in the smoke. It is important to note that in some cases the sidestream smoke contains higher concentrations of nonvolatile nitrosamines,than mainstream smoke. Similar to volatile

nitrosamines,nonvolatile nitrosamines appeared to be formed in higher amounts during smouldering (sidestream smoke forma- tion) than during puffing (mainstream smoke formation). The sidestream smoke of one popular nonfilter cigarette ~ contains 1,7_00 nq NNN, 410 ng NNK and 270 ng NAtB compared ~ to 680 ng NDMA and 300 ng NPYR. For the determination of the a volatile and nonvolatile nitrosamines in sidestream smoke an air flow of 25 m1/sec was drawn through the apparatus. This flow rate was chosen in order to reproduce values for the mainstream smoke corresponding to open air smoking. It is known from our previous study however, that the levels of nitrosamines in sidestream smoke are influenced by the air flow rate thYough-the smoke apparatus (6). With the aid of the TEA, we detected volatile 2iDMA in smoke polluted rooms (10-130 ng/m3). In highly polluted rooms the amount of NDMA in the air inhaled during one hour is equivalent to the amount of NDMA present in the mainstream smoke of 1-35 cigarettes (5). However, origin of NDMA in room air is not necessarily limited to tobacco products. Based on the data for sidestream smoke from this study, we may be able to detect tobacco-specific nitrosamines in polluted in- door environments and thus define more clearly the contribution of tobacco smoke to the carcinogenic potential of a given polluted indoor environment.

w rv 4 xiFEREr:CLs 2. Bethmann, M. von, Lipp, G. and Nooy', H. van, Feuchtigkeits- Bates, Griffith, R.B., Harlo:v, H.S., Senkus, W. and tsakehamy H. Determination and Reporting of Total, Particulate Matter, Water in Total. Par ti culate Matter ... and Nicotine. Tobacco Sci.,12: 192-196, 1968- bestim,-nung im Tabak. Beitr. iabakforsch.,1: 19-29, 1961.- 3. Boyland, E., Roe, F.J. and Gorrod, J. W. Induction of Pulmonary Tumors in Mice by Nitrosonornicotine, a p©ssible Constituent of Tobacco Smoke. Nature 202: 1126, 1964. 4. Brunnemann, K.D. and Hoffmann, D. Chemical Studies on Tobacco Smoke XXV. The pH of Tobacco Smoke.Food Cosmet. Toxicol., 12: 115-124, 1974'. 5. Brunnemann, K.D. and Hoffmann, D. Analysis'of Volatile Nitrosam_ines.in Tobacco Smoke and Polluted Indoor Environ-~ ments. Intern. Agency Ies. Cancer Sci..Publ: 19: 343-356, 1978. 0 C- 6. Brunnemann, K.D., Yu, L., and Hoffmann, D. Assessm°nt of ~ Carcinogenic Volatile N-Nitrosaaiines in Tobacco and in OA Mainstream and Sidestream Smoke from Cigarettes. ~ •Cancer Res., 37: 3218-3222, 1977 Q~ 7. Elmenhorst,• H. Gas Chromatographic Analysis of Nicotine and Minor Alkaloids. Coresta znform. Bull. 1978 Special,120.

+ YECS~iC .. 1 ,.r0 5 .,iC`SS Fine, D.H-, Rufeh, F., Lieb,"D_ and Rour.behler, D.. Description of the Thermal Energy Analyzer (TEA) for Trace Determination of Volatile and i.'on-volatiLe NLNitroso Compounds. Anal. Chem., 47: 1188-1190, 1975 9. Hecht, S.S., Chen, C.B., Hirota, N. Ornaf, R.M., Tso, T.C. .and Hoffmann, D. Tobacco Specfie•Nitrosamines: Formation from Nicotine in Vitro and During Tobacco 10. Curring and Carcinogenicity in Strain A Mice. J. • Natl. Cancer. Inst., 60 : 819-824, 1978. . Hecht, S.S., Chen, C.B., Dong, tL,, Ornaf, Hoffmann, D. and Tso, T.C. Chemical Studies on Tobacco Smoke LI Studies on Non-Volatile Nitrosa.mines in Tobacco. Beitr. Tabakforsch.. 4: 1-6, 1977. 11. . Hecht, S.S., Schmeltz, I. and Hoffmann, D. Nitrogenous Compounds in Cigarette Smoke and their Possible Precursors. Recent Adv. Tobacco Sci., 3: 59-93, 1977. 12. Hilfrich, J., Hecht, S.S. and Hoffmann, D. Effects of N'-i>sitrosonornicotine and'N' -'Mitrosoanabasine in Syrian Golden Hamsters. Cancer t,etters, 2: 169-176, 1977• 13. . Floffmann, D., Dong, M. and Hecht, S.S. Origin in Tobacco Smoke of N'-Nitrosonornicotine, a Tobacco-Specific Carcinogen: Brief Communication. J. Natl. Cancer Inst.,58: 1841-1844, 1977. 14.' • HoffrtZnn, D., Flecht, S.S., Ornaf, R.M. and Wynder, E.L. N'-N.itrosonornicotine in Tobacco•. Science,186: 265-267, 1974. Hoffmann, D., Hecht, S.S., Schneltz, I,,Brunnemann, R.D. and Wynder, E.L. New Separation Techniques for Classes of Smoke Compounds. Recent. Adv. Tobacco Sci., 1: 97-122, 1975-

a Hoffmann, D., Raineri, P.., Hecht, S.S., Maronpot, R_ and Wynder, E.L. A Study of Tobacco Carcinogenesis XIV Effects of N'-Nitrosonorr.icotine and N'-Nitroso- anabasine in Rats. J. Natl. Cancer Inst..,55: 977-981, 1975. 17. Hu, M.W. , Bondinell, I•7.E. and Eo:fmann, D. Chemical Studies on Tobacco Smoke XXIII. Synthesis of Carbon-14 • Labelled Myosmine, Nornico tine and N'-Nitrosnornicotine J. Labelled Comp.,10: 79-88, 1974. 13. . International Committee for Cigar Smoke Study. Machine Smoking of Cigars. Coresta Inf. Bulletin 1974-1, 31-34. 19. Kuhn, H, Die Tabakalkaloide und ihre Pyrolyseprodukte im Tabakra•uch. Mitt, Oesterr. Tabakregie,5: 73-82, 1964. Mirvish, S.S., WalLcave, L., Eagen, Ei_ and Shubik, P. Ascorbate-Nitrate Reaction: Possible 2leans of Blocking the Formation of Carcinogenic %I-Nitroso Compounds. Science,177: 65-68, 1972. 21. Quan, P.M., Karns, T.K.B. and Quin, L.D. The Synthesis of 20. 22. 23. Anatabine and Related Compounds. J. Org_ Chem.,30: 2769-2772, 1')65 Schmeltz, I., Brunnemann, K.D., Hoffmann, D. and Cornell, A. Chenical Studies on Tobacco Smcke XLV. On the Chemistry of Cigar Smoke: Comparison beto-:een Experimental Little andl Large Cigars. Beitr. Tabakforsch.;8: 367-377, 1976. Singer, G.M. and Taylor, H.W. Carcinogenicity of N'-Nitroso- nornicotine in Sprague-Dat:ley Rats. J. Natl. Cancer Inst., C 57: 1275-1276, 1976. ~ t•7yn.der, E.L. and Hoffmann, D. Tobacco and To3acco Smoke. ~ Studies in Experimental Carcincgenesis. Academic Press, ~ New York, New York, 1967, pp 453-458. ~ ~ . O

25. Klus, H. and Kuhn, H. Untersuchungen t7ber die nichtflilchtigen N-Nitrosamine der Tabakalkaloide. Fach. Mitt. Austria Tabakwerke A.G. 307-317, 1975.

Table 1. Quantitative Analysis for NNN in Cigarette Nainstream Smokea,b ug/cig. Analysis 1 2. 3 4 Average Standard Method 0.21 0.18 0.24 0.20, 0.21 Enrichment Step 1 .0.21 0.19 0.23 0.19 0.21 Enrichment Step 2 0.21 0.18 0.23 0.18 0.20 a85 mm cigarette without filter tip. bDetails see "Results and Discussion"