Tobacco Documents Online

Nl'1'~iVSE1I~T1Nt:~ : ~ WiulJ• LilUl. ~~ N11~.U.N11'U1V U~ N1l.Ul1N,. LUi.1tVG CURING OF TOBACCO AND CARGINOGENICITY IN STRAIN A MICE1'2 Stephen S. Hecht3, Chi-hong B. Chen3, Norio Hirota3, Raphael M. Ornaf3, T.C. Tso4, And Dietrich Hoffmann3 1. Received ; Accepted 2. Supported by Public Health Service'Contract NDl CP 55666 from the Division of Cancer Cause and Prevention, National• Contract 12-14-100-11031(34). Stephen S. Hecht is recipient of National Cancer Institute Research Career Development Award No. 5KO4CA 00124. 3. Division of Environmental Carcinogenesis, Naylor Dana Institute for Disease Prevention, American Health Cancer Institute, and by Agricultural Research Service Agricultural Research Service, USDA, Beltsville, Maryland 20705 4. Tobacco Laboratory, Beltsville Agricultural Research Center, Foundation, Va1ha11a,-N.Y. 10595, r . ,.. spectrometry; CI, chemical ionization mass spectrometry; Abbreviations used: NNN, N''-nitrosonornicotine; NNK, 4-(N- methyl-N-nitrosamino)-1-(3-pyridyl)-l-butanone; NNA, 4-(N- methyl-N-nitrosamino)-4-(3-pyridyl)-butanal; GLC, gas-liquid phase chromatography; GLC-MS, combined gas-liquid phase .chromatography-mass spectrometry; EI, electron impact mass HPLC, high performance liquid chromatography.

nicotine, was examined. Detached leaf tobacco was fed either n N e ana yze or -nitrosonornicotine-2 - nicotine-2'-14C or nornicotine-2'-14C, and air cured. The 1' , 14 cured leaf was th 1 d f (NNN-2'- C). The yield'of NNN-2'- C was .007% from nornico- nicotine is considered to be the major precursor for the carci- to nornicotine in conventional nicotine-type tobacco is-2U1-100 tine and .009% from nicotine. Since the ratio of nicotine sis of tobacco revealed the presence of NNK (0.6-2.4 Vg/g) 0 (N-methyl-N-nitrosamino)-1- (3-pyridyl)1-1-butanone (NNK), and 4-(N-methyl-N-nitrosamino)-4-(3-pyridyl)butanal (NNA). Analy- NaNO2, gave'rise to NNN, as well as two other nitrosamines, 4 nicotine in vitro was then studied. Reaction of nicotine with nogen NNN in tobacco. The formation of other nitrosamines from O in chewing tobacco and snuff. The tumorigenic activity of NNN, FFO NNK, and NNA in strain A mice was studied. Both NNN and NNK iAb C!i induced significantly more l~u:.gg adenomas; .~ouse than observed ~ in controls, while NNA was inactive. In addition, there occurred two cases of undifferentiated glan'ds in the NNN experimental groups. study clearly indicate that in tobacco carcino:na of the salivary The results of this precursor for specific carcinogenic nitrosamines.

Cancer of certain sites is associated with tobacco usage. Epidemiologic studies have indicated that smoking is related to an increased risk of not only lung cancer, but also cancer of the oral cavity, esophagus, pancreas and bladder (1-12). The chewing of tobacco is also associated with higher incidence of oral cavity and esophageal cancer (5,7,13-14). While certain of these diseases are most likely caused by contact carcinogens, carcinogens, when tested in experimental animals. Depending site specific carcinogen. Nitrosamines are organ.specific others, such as pancreatic cancer, are more likely caused by a on structure and species, tumors of various sites are obs,erved; among these are the respiratory tract; esophagus, bladder (15-19) .

The most commonly reported environmental nitrosamines are derived from simple amines such as dimethylamine and pyrrolidine (15,20). These substances, andother volatile nitrosamines, also occur in tobacco smoke (21). Of special interest, however, are tobacco specific nitrosamines, which are derived from the tobacco alkaloids. Such compounds could be related to the incidence of -tobacco related cancers. The prototype of tobacco specificc nitrosamines is NNN, which could be derived either from nicotine (see text-fig.l) or nornicotine. NNN,which induces nasal cavity and esophageal tumors in rats (22,23), respiratory tract tumors.in hamsters (24), and lung adenomas in mice (25,26), has been detected in both unburned tobacco (0.3-9.0 ppm) in cigarette tobacco and 2.4-88.6 ppm in chewing tobacco, 27-30) and in tobacco smoke (40-250 ng/cig, 31-33). NNN was not present in fresh tobacco, but rather formed during curing (26). Various lines of evidence suggested that the major precursor for NNN in tobacco was nicotine rather than nornicotine (26,27,34). This question has now been examined in more detail. In this report, we present evidence that nicotine is the major pre- ~- cursor for the tobacco specific nitrosamine NNN. As previously ~ suggested (30,35), at least two other nitrosamines, NNK and NNA ~ (text-fig. 1) can also be for.::ed from nicotine and one of these, Q1 Ca NNK, has now been detected in tobacco. The tumorigenic activity ~' of these three nitrosamines in strain A mice was also examined.

instrument equipped with a flame ionization detector and column I, 6-foot x 1/8-inch 10% Carbowax 20.M-TPA on Gas-Chrom Q; column II, 6-foot x 1/8-inch 10% Carbowax 20M-2% KOH on Chromo- sorb W with helium (60 ml/'min) as carrier gas. GLC-MS (EI and CI) was done with a Hewlett Packard Model 5982A dual source instrument using a membrane separator (EI) and helium (EI) or methane (tI) as carrier gas. For direct effluent counting, a Hewlett-Packard Model 5711 GLC was coupled via an effluent splitter to a Packard Model 894 Gas Proportional Counter. HPLC was performed with a Waters Associates Model ALC/GPC-202 high speed liquid chromatograph equipped with a Model 6000A'solvent delivery system, a Model 660 solvent programmer, a Model U6K septumless injector, and a Model 440 UVlvisible detector. HPLC columns were as follows: column A; 6 mm x 30 cm microporasil (Waters Associates) ; column B; 6 r.tm x 30 cm microbondapak/C18 (Waters Associates); column C; 6 ft x 1/8 in basic activity II-III ti;oelm Alumina (38-75 u). Liquid scintillation counting was done with a Nuclear Chicago Isocap 300: scintillation exc].usion of other Aroducts that mav a 7czn, h4- s» i t ah1 P_ CD Dept. of Agriculture and does not imply its approval to the CA stitute a guarantee or warranty of the product by the U.S. 'jp1 Menti.on of a trademark or proprietary product does not con- system. Evaporations were done under reduced pressure with'water tempera- tures tures below 60°C. Q O 5 O

-Reagents - All organic solvents were spectroquality. Nico- tine-2'-14C, nornicotine-2'-14C, and NNN-2'-14C were obtained from New England-Nuclear, Boston, Mass. NNN-3 H was obtained by nitrosation of nornicotine-3H, as previously described (32). The purity of the nornicotine-2'-14C, nicotine-2'-14C, and NNN- 3H was reconfirmed by TLC on silica gel (see ref. 36) and by 14 C was repurified before use by GLC (column I, 220°). NNN-2'- HPLC (col. A, elution by cyclohexane/chloroform/methanol/30/ 68.6/1.4). Nicotine was obtained from Aldrich Chemical Company, Milwaukee, Wisc. and was redistilled before use; it contained <.01g nornicotine according to glc analysis (co1. 1I,.165°). Nornicotine was synthesized by literature methods (36), as were NNN., NNK, and NNA (36,37). The latter was stored in CH2C at 0° to prevent decomposition. Trioctanoin (Eastman Kodak Company, Rochester, N.Y.) was redistilled before use. Column chromatography was done with activity II-III basic alumina -.- • , (WoeTm).

C 14 14 14 C or norr.icotine-2'- C fed tobacco NNN-2'- C in nicotine-2`- a. Growing and selection of experimental tobacco- Five experimental groups were used: (1) negative control, .in which tartaric acid buffered to pH 6 was fed to 8 leaves, b. Feeding of nicotine or nornicotine to tobacco leaves Leaves of Nicotiana tabacum cv. Burley 21 were used for this study. The tobacco plants were grown in the field at Beltsville Research Farm employing regular Burley production practice. When they reached harvesting stage, two well developed middle leaves from each plant were selected. These experimental leaves were primed (detached) immediately before feeding. (2) cold nicotine control, in which nicotine buffered to pH 6 with tartaric acid was fed to 8' leaves, (3) cold nornicotine control in which nornicotine buffered to pH 6 with tartaric (4) nicotir.e-2- C, in which nico- acid was fed to 8 leaves, '1Q tine-2'-14C buffered to: pH 6 with tartaric acid was fed to-two groups of 4 leaves each and (5) nornicotine-2"-14C, in which nornicotine-2'-14C buffered to pH 6 with tartaric acid was fed to two groups of 4 leaves each. All feeding was carried out in laboratory hoods. The appro- priate solution (2 ml) was placed in a small vial and the base of a detac:ed leaf was inserted' into the solution. When the solution had been completely absorbed, a 2 ml portion of dis- tilled water was added to the vial. This operation was repeated twice.* The total feeding time for each leaf was approximately 24 hours.

Experimental leaf tobacco was hung in the hood under condi- tions similar to air curing. Environments were adjusted to prevent quick drying and thus providing conditions essential for biochemical changes normally occurring during air curing. This curing process was completed in three weeks. d. Analysis for NNN-2'-14C All analyses for NNN were done by a previously published method (33) using NNN-3H as internal standard. In a typica.l analysis (20-30 g tobacco), 1.1 x 106 dpm (0.08 ug) NNN-3H was added. An aliquot of the final HPLC eluant containing NNN was counted by dual channel counting to determine recovery. Another aliquot was injected into the combined GLC-effluent counter system (column I, temperature, 210°) which was cali- brated with known amounts of NNN-2'-14C. NNN-3H g ave no response in this system. This allowed determination of NNN- 2,-14C content and, by splitting to the flame ionization detector, total NNNcontent. An aliquot of each 1~C-treated tobacco sample was submitted for combustion analysis (New Eng,land Nuclear, Boston, Mass. )~ to determine the actual amount of nicotine-2'-14C or nornico- tinQ-2'-14C in the cured' tobacco sam.ples. Alkaloid analyses were done by published procedures (33).

Reaction of nicotine and NaNO2 Nicotine (1.0 g, 6.17 mmol) was dissolved in 1 ml H2O and the pH adjusted to the desired value by addition of iN HC1. This was then added to the appropriate buffer to a total volume 'was also added and the total volume was 50 ml. NaNO2 (0.6 g, 8.7 mmol) was added in one portion and the resulting solution hydroxide.- In one experiment, ascorbic acid (10.0 g, 56.8 mmol) phosphate; pH 4.5, citrate-phosphate; pH 7, phosphate-sodium of 20 ml. Buffer solutions were: pH 2, HC1-KC1; pH 3.4, citrate- was stirred for 17 hrs at 20°. The reaction mixture was brought to pH 10 by addition of aq NaOH, saturated with NaCl, and ex- a mixture which was dissolved in 1.0 ml MeOH for quantitation of nitrosamines by GLC. Aliquots were injected on column I at a column temperature of 220°. The relative retention times of nicotine, NNN, NNA, and NNr under these conditions were 0.18, 1.00, 1.69, and 2.38, respectively (absolute re- tention time of NNN '= 13 min. ). The mass spectrLm of each peak was determined by GLC-MS. The area under each peak was de- termined by integration. Yield's were calculated by comparison O tracted 6 times with equal volumes of CHC13. The combined CHC13 extracts were dried (Na2SO4) and concentrated to give to response curves constructed from 8 standards. E_

Tobacco Analysis for NNK Tobacco products were obtained on the open market (1975-1976) in the New York metropolitan area. Snuff and fine cut chewing tobacco were analyzed directly, without grinding. Other pro- ducts were ground in a blender before extraction. An aliquot was taken for water determination (39). Ground tobacco (80- 100 g) was stirred at 20° overnight with 700 ml aqueous ascor- bic acid (5 mM) at pH 4.5 (citrate-phosphate buffer). The weak- i (28) except that ethyl acetate was substituted for chloroform. ly basic (pyridine fraction) was obtained as previously described This material was passed through a plug of activity II-III basic hexane to remove highly polar or insoluble materials. The alumina (2.5 g) with elution by-10 cc 30% isopropanol in cyclo- eluant was concentrated and chromatographed by HPLC (column C) HPLC (column A) using solvent program6 in 10 min: initial con- ditions solvent B (30%) ; final conditions, solvent B (1000) ; volume (20-31 ml) as standard NNK was collected'. Several in- jections were necessary for a sample of this size. After con- centration the combined fractions were rechromatographed by hexane, solvent B was 3:0% isopropanol in cyclohexane; flow rate. 1.5 cclmin. That part of the mixture having the same retention using solvent program 8 in 15 min: initial conditions, solvent B (25%);final conditions, solvent B(.100%);, solvent A was cyclo- solvent A was 3% cyclohexane in CHC13 and solvent B was 3~% iso- propanol in chloroform; flow rate, 1.5 cc/min. The peaks Isolated amounts of NNK were determined by comparison of peak correr.ponding to E- and Z- NNK were collected and concentrated for determination of mass spectra by direct inlet (EI and CI). ` 0