Tobacco Documents Online

FlDENTIA ASSESSMENT OF THE CAECINCXGENIC N-NITROSODIETHATTOI.AMINE American Health Foundation Submitted' for publication. i :}•~~ : IN TOBACCO PRODUCTS ANIDiTOBACC4'J SMO1GEI .fx Klaus D. Erunnemann and Dietrich Hoffmann Division ofEnvironmental Carcinogenesis Naylor Dana Institute for Disease Prevention ,vi 3t" ' -M ~ hU''ec4, ----~~-~~ .^.~5..._.~.

5 Running Titles Nitrosodiiethanolamine imTobacco Psoduicts. ' ~^i ' , . ~ ' .. . ~ . . . . . .. . . .. ~ Sj, , ' e

f. k.~free of NDELA. The tobacco of 0'.Sg smoking products contained, M~ 'ng/cigarette, whereas hanid-suckered tobacco and its smoke were 'tained 1i 1i5-42'0 ppb of NDELA and their smoke containedi 20'-290 1115-42'01 ppb of NDELA and the mainstream smoke from such pro- 4 d~ucts yielded 10~-618 ng of NDELA per cigar or cigarette. NDELA ~ levels in, chewing tobacco, ranged from 2'20 to 280 ppb and , in 2 NDELA in tobacco is formed from the DELA residue during the to- bacco processing. Based on bioassay data from various laborar 0.6-1!.9 ppm of NDELA it is evident that the major portion of the five analyzed MH-DELA pre arations.contained between commercia~l snuff pr~o~~ducts~, were~ 3',2~~0~~0! and 6,800 ppb- w,Althiough iories which have showrn, that NDELA is a relatively strong car- cinogen and based on, the results of this study the use of' ~. _,.._ .... ' ME-DELA for the cultivation olf tobacco is questioned. .. 0 97 f f Abstract w 4 simple~,-"~repro~~du~cible~ GG'-TEA method has_been d'ev~e~loped m>~.~,F :~t~l,and tobacco smoke. .The extract of tobacco or the trapped par- for~'~th~e~~ analysis o~f~ N~-nitrosodi~ethanolamine ~ (NDELA)i i'n to~b~acco~ . ., . ''~"ticulates of' tobacco smoke are chromatog,raphed on silica gel. '='i•~`The NDELA containing fractions are concentrated,_silylated and 1v` analyzed with a mod~ified' GC-TEA system,. ; NDELA-14~ serves as ` internal standard'for the quantitative analysis. :,Experimental cigarettes made from tobacco which was treated with the sucker "ry ' g,ro th inhibitor maleic hyd!raaide-diethanolamine (MEi-DELA) _con- T(.r { ,,1 I )kr Yr WL -t r, ay.

INTRODUCTION N-Nitros©diethanolamine (NDELA) appears to be one of the 0-~%-~ "'rmost'widespread N-nitroso eompoundis in the environment of deve- loped countries (1i).'--`NDELA is found in a variety of products ' , .. . - .. . ' .. „ . . ti . . . .:,Y` ... . . . . . . .. ~ incl.udingcutting fluift'and cosmetics and is formed by reac.- `_tion of dii' . .... , . . ..:,:. ., . _ . ethanolamine and triethanolamine with N.-nitrosating agents (1-3).0 'r'Formation of NDELA is 'largely due to the exten- sive use of these two amino alcohols. =..ND'ELA induces carcinoma y liver,and kidney of ihe ,ry in rats (4-b)anal carcinoma or ttrie 11 ~nas~al,cavity~ and papillomaiof the t~rachea in S~yr~ian gol!den h~~am- toxicity" was induced in Syrian golden ham- sters compared to those receiving the smoke of'control cigaret- . ,:. es or of' cigarettes containingi 1A of the sodium salt of' MEi n the United States about 3'.2 million pounds of'maleic hydraaid'e are used annually as systemic plant growth inhibitor (12), about 40i$' as the M!H-DELA formulation (13, 14). In a preliminary study, we reported that processed OS to- bacco treated with MH-DELA contains 100-170, ppb of NDELA (15'). For the analysis, NDELA was enrichedl from tob!acco extract by sters ('7 )i .' ,`,Thie'carcinogen penetrates rat ,. skin ('9) and 10) ~~. . :.r . skin (8), and human is primarily excreted via the urinary tract.(8', n a 2'6 week inhalation study with smoke of'cigarettes -''7:co1ntaining~' 1$~~ male~~ic~~ hyd~razide~-die~th~ano~~lamine~~ (~,MF3+-DELA)~ a~ "Mild '. ? NY 1. 9W th 4: ,,, V

/ k: 0 Abbrev i~ations: So .4-a i Y~O 11 15 T. '1i •.. P.Q9w' u"t ~ A1DET,A, ' IQ-nitrosodiethanolamine; MH', maleic hydtazide; ;DELA, a3i- ,;17"ethanolamine; TEA, Thermal Energy' Analyzer; :_GC~-MS, Gas Chroma- ! ... tography-Mass-Spectrometry; BSTFA, N,O~-bis(Ttimethylsilyl)tri- _ .. . 1 ,. . . ,. ., .. , . . ; , ~ ' fluoroacetamide; TMCS, Trimethylchlorosilane; NNN, 2Ni', n~troiso- , nornic~ot~ine~~;~~ NNK, 4'~-m~,ethyl-4~-nitosamino-1i-~~(3':-pyridyl~)~- butanone;, NAT, N''-nitrosoanatabine. ,, , a, i r: L y 4 ~4W791 t;

repeated chromatographic clean-up steps fallowed by ..;gas chro- neither 1arg,e'-scale studies nor the trace analysis of NDELA in cigarette` smoke. `'".:It was the purpose of this investigation to ~~'`develop"a reprod~uciblle and simple method for the determination ~"of NDELA in processed tobacco, and in tobacco, This method, however, was cumbersome and permitted

ly1ATERTAhS' ANdD METHODs 'X L:i.'t :; i~-1rJ ... ,.,, i.. ~. .k ~Js'~" s.3f1 'C~3t'li.f 3!~ ~ For the~ analiysis of the mainstream smoke we used a 2'0i-port xc~ .,.I .: 1 .., Th Eli t o~ Cbr tw,Taltham t~ass )'tth'is was c r n ~ . J Apparatus <.i.:.;",.,? Hamburg', Germany)i with' rotating ,,l~~ead of which,, ever~r 2nd port was connected wrilth, a nitrogen ,,. , , . . ,. .- sec«~(16)."~For the sidestream smoke analysis we utilized a. s~o~iaro~,`_'thuis~~`replacing~~~ the'~~a~ir in 'the~ traps with n~itr~ogien, every ermany an ' single-channei sm~alcer (H. Borgwaldt, -Ham urg,,, .. _ b, ~ )' di sidestreamismo}~e apparatus with an' air 'f]~ow rate of 25, ml~'sec, f~.ed by removing, the TC detector and acco~o~dating the GC inter- . ~.~.r.. ..~ 1i7). '-A Hewlett-Packard Model 700 g'as chromatograp~'~~~' was modi- face face and the ceramic pyrolyaen of' the Model 543 Thermal._ Energy Ana y~er (, ermo e accomplished by drilling a'hole through the oven housing thus allowing, a direct' interface' to, the chromatographic column. The 'TEA pyrolyzer interface was further modified by addition of a :.,, ~ , reducing union (Swagelok SS'-8''10-6'-41) into which a raph'ite fer- . rule (Applied Science, State'College, PA., part #15457) was placed (Figure 1) This allows the connection' of tt'fe gas chro- mato!graphic column directly to the interface rather than to the Y` brittle ceramic tube itself. S'ilylations were carried out in, 1 ml hiypovialis ( P'ierce 12901). Mass spectral analyses were per- formed on a Hewlett-Packard Model 5710!-59'8'0 instrument. }6

Cigarettes N a.,. ,. . . , .. , ~ . The commercial tobaccolproduicts were purchased!"in 1i980 on the open market in Westchester County, N.Y'. , and one fine cut :~tj.~ ~-Otobacco was obtained in Tennessee. Experimental cigarettes made from ~~ hand-suckered tobacco, ~from, MH-DELA treat~ed~ to~lbacco~ i: andifrom hand-suckered tobacco g,rownion pesticide-free soil on ~vq Prince Edward Island, .Canada,' were kindly supplied by Dr. T.C'« Tso,' USDA, Beltsville, MD1. ;,.-.1iR1 cigarettes from the University fa.of Kentucky served as reference standards. All cigarettes and 'cigars were stored in a humidity chamber at 60i (+ 3$ ) R. fl. Throughout this study, we applied the standard smoking °s conditions for cig,arettes (18), and' cigars (19). ~..;.L;~r;:Reagents Y .z .Y.79,.t•~~'....T .. ~ NDELA was synthesized iniour laboratories by nitrosation, of d~iethanolamine and subsequent purification on basic alumina ~~(~Woe~lm~~),,` activity III;~ its p~~urity~ was a~s!sured~ by~ G~,C~-MS.~ NDELA- 14C was obtained by nitrosation of dliethanol[i2-1'4'C]I-amine HCl (spec. act. 116 mC~i/mM;~~, Dhiom, Prodlucts~,, R. Bollywoo~d,, CA) and~ purification by column chromatography using 250 g silica gel (1-34041, -Baker Chemical Co. ) and ethyl acetate. M'ethanol, ace- tone, tone, ascorbic acid andlCelite 545 were obtained from Fisher ~r = Scientific Co. Acetonitrile and BSTFA with 1% TMCS' were pur- r M chiased -at Pierce Chemical Co. The chromatographic packing OV-225 was obtained from Analabs, North Haven, CT.

,LTobacco Anialysls ~ Forty~ g~ fine~ cut~ tobacco were~~ combined~with~ ,I.O.g g.ascorbic~ 120 ng NDEII.A-14C' ( 2301, 0100 . dpm)i and 400 ml ethyl acetate. { This mixture was stirred overnight and then filterEd thxough ~~,. ..: .. . _. . ~ . . *wR..-.....,.......:....... , ...... .~....-~w-..Y~~.. ' . . ` 't ~ . . , . . ~ : . P,4,..,Celite 5451..w The filtrate was thenconcentrated~ by lrotary eva, ~.~ :a ... ;.i.;poration at 35°C to approx._ 3 ml and, chiromatographed on A small ;~,"rcoSuim~~n ('5~10~ g! silica ge~l) ,'. with ethyl acetate-methanol ('9~~9:1)! .~ , . ,, ,. . . . ~ : r . . .. ., ..- ;\Anialiquot of each 1i010 ml fraction was checked' for A-activity. added (the silylation reaction is very~rapid and does was then placed in a 1~ ml hypo~ vial and: 0'~ . 21 ml B'S~TPA~ with~ 1i $~ not r:equire~ additional heatin~g). ~~.~Anl.aliqwot olf~a,5~_w1 was~z then injected into, the GC-TEA. : The, conditions were: 6 ft x 1/4" _:V!j,~Jnterface T60°C, TEA pyrolyzer 5'20°C. ;;The carrier flow was 40, so:rb~~ W' HP,~ 8'~0~-~10~~01 mesh:;~ injection ~ port 17 01° C , o~ven~ 14 ©~° C-, TEA~ ;;L~ml Ar/min, the TEA cold trap was kept at ,12'5°C'. ~~ ,.. , x .~- . Tobacco Smoke Analysis radioactive fractions were pooled, evaporated redissolved in 2.0;~._ml acetonitrile. ~-,.An aliquot~ of' 0~.5~ ml .,, . .. . . (2 mm i.di. ) glass column packedi with,3$ OV-225 on Chromo- : . _. _ „ ...: , ~ ~ ~ ~~ ~~ .One hundred cigarettes or 40'cigars were smoked on an al MAin~stream~ Smoke smoker with~ rotating,~ he~adl,~ and~ the~~ smo~~ke~ was~ passed through two gas wash bottles containing 1.00 0 ml ethyl acetate and 1 g ascorbic acid each, one cold trap immersed' in dry ice and a Cambridge filter, ((69 mml). The ethyl acetate fractions LA ' ~ C ~.

were pooled, filtered and' evaporated to approximately 3' ml, ~. ~,y,.~ ;. and then chromatographed~,'sily]iated and analyzed by G'C'-TEA as , s-1-'jdescribed :yunder' "Tabacco' Analysis" ,.. - A ' s ~~~ ~ :}~ Sidestream Smoke F~:: ;- za b Cigarettes were smoked individually by `piston-type machine.` The sidestream smoke was lcd into 2 gas wash bottles at1 an air flow rate of 2'5 m1/'sec. late matter and nicotine in the mai.nstream smoke are the same ~ s~jr • ~„ : . .- Forty cigarettes were smoked in'a sidestream apparatus (~Thisair~~-~flosa through the~ sidestr~e~am, smoke apparatus with a chamber v~o~~lume~~ oIf 20I0I ml assures th~at y~ie~ldsof total Sf.r:,?: . i_. as those obtained from smoking of this cigarette in the open, r . ' air~),.°~ Th~e~ wo~~rk-ug~~ o~~f~ th~e~~ tr~apped'~, srnoke~~ wa~sidentical to th~at. 'of the mainstream smoke described above. Analysis of MHrDELA Preparations - 0 X xr . ~"The'commercial MH-DELA solutions contain 58% maleic hydra- zide diethanolamine salt and 42% "inert' ingredients" (;water ~ s~ etc. ); this~i equivalent to 3'7'$maleichydraaid'e. Five hundred mil acetone and NDELA-14C as internal standard were added to 1!0.1 gi MH-DELA solution and the mixture was subjected to rotary evaporation for removal of water. The concentrate was dissolv- ed in methanol andi chromiatographed on 501 gi silica gel with ethyl acetate as eluant. The fractions containing !3-activity were combined, concentrated Z~ 1..--, „)- to 3 ml and then again chromato- t1t 6 - ~' M.

graphed and analyzed, as described . under '"Tobacco An,a~lys'is" . "Y41-~colunnns 'of' 250 g silica gel each using ethyl Xace~ate, followed' ~ .. .. _ . , .. . .~ .. ._ . _J:fi~ ~,'s ~ .. M~i~.Y`zr-by an additional column of 50 g alumina, with 1'0!%methanol .- The fractions cantaining the tate as eluant ~~'~}~r t~i l . e y ace . ~i.-: . `Mass' Spectral ldentification, ~'~O1 R1 reference ' cig,arette, then enriched by two chromatographic ;.. . «-1,,,~~Js~-activity~ we~~re~ combined, =carefhlly~ ~ evaporated to" dryness and th~n redissolved in 100 p1l silylating agent. b) Mainstream Smoke Fraction t:' Fivehundred, 1iR1referencecigarettes were smoked. as~ ~ _ . . .,. , ``'`-~ "~,GC-MS using a capillary column,,30' m x 0i.25^mm i.d. SP -210'0, 1, Icribed!earlier. The enrichment, Ehromatographic~clean-up and silylation steps were the same,as those for.the ~tobacco~frac- )i Tobacco Fraction An extract was prepared from 2010 ,g of ~he tobacco of the "` ,, .. ;.., ..z tion. An aliqu~ot of the silylated concentrates ,was analyzed by . mil He/inin, splitless mode; the , GC program was 4't~min at 8'0`C, then 4°/min to 200'°C. -•Under these conditions, the silylated ~ derivative o1f' NDSLA had a retention time of 251.51 min. RESULTS ~ F~igure~~ 2, depicts typical chrom~atogramsof~ s~i~lylai~ed! NDS~LA~ from the concentrates of tobacco, cigarette smoke and MH-DELA

'~obtained ~uinder the `eonditions described~ under "Materials an . ., Occasionally, we observed_several larger peakss following the elution of' NDELA (TMS)., These were found to be ; ¢7.ue tci }tobaeco spebific nitrasamines NNN NNK and NAT (21For .,j« t1ass hie m spectral confirmation, we enriched NDELA from 200 g of •. . ' tobacco or from the mainstream smoke of 500 cigarettes (Figure -.- _ .~~ The precision'af the methodiwas' established by 5 analyses M , _ _ .. of 401 g Foff' ~tobacco of' the 1 R1 Kentucky cig,arette, and 5, analyses f tbesm~oke~ o1f~ 100 Kentucky~ IR1 cigarette~seach~:('~Table T)~. The standard deviations of 111.91V and 1I1'.7'% respectively, appear highr' but are likely due to the fact that we analyzed non- - "cominercialcigarettes and! that the MH-DELA residue is unevenly `,°distributed when -the tobaccolplants are sprayed. The recovery 'rate ©ff' the '14C-labeled NDELA which was used as internal stand- r ~i ardvared betwee~n~ 50! and, 7~01$ i tob ~ tmbacco~, smak d~ ~~n,acco!, ean MH-DELA samples. The addition of 5.5 mgi of DELA to a mixture ~~,~'~'!'). .. ... ~: .... ~.. : .~ .: . ,,... .. . . . . containing I g of~ asco~rb~ic~~ ac~id~, 40~ g of Kentucky 11R~~1 t~o~b~acco~, ,.~ ,. and 4010, ml of ethyl acetate did not lead! to an increase in the NDELA yield'; when 5.5 mg of DELA were added to the traps col- lecting the mainstream smoke of the DELA-free USDA cigarette (L-8' )I , the amount of NDELA detected' in the final concentrate was insignificant. This assures that under normal conditions, NDELA is not artifa'ctua'lly formedlduringi the analytical work- up. O This study demonstrates that silyl,ated nitrosamines p c~f CIt . ~ - 8 - ~

are amenable to GC-TiEA detection with undiminished sensitivity. Table I'summariaes the findings for NDELA i~n tobacc© and .. . . v ,, . _ tobacco smoke of vari rv ,ous cigiarettes and other tobacco .4 ~: w z. i°t l the case of the hand-suckered tobaccoland the r1, k~ 1 e,~•... . ~:IDELA free tobacco, NDELA was not detected, (detection limit 0. . . ~ ...Y, ~ ppb in trobacco or 1,, ng~cigarettei a wery small~ge~k~obs ~rved at the retention time of NDELA was attributed to the internal w"::j~~:~1~I~Y~i~ , ~ ~ standard,, ~ N~DELA-1i4C'~)1.,_ A~ Frenchcigarette~ (purchased~ in ~1i975), ~„ y. ~eld i in the mainstream smoke did not i n the ~obacco nor NDELA which leads us to believe that the tobacco used for this „{ . .. ... ,. . . . . . ~ ., ,., .. .. • E. . . ..i. 5~ ir'. , product was not treated with MH-DELA. The two snuff samples .... 4 ,~ T . . .. ... . . ~ .. .. .. . .. . . .. „ . ,. ~ . . .. analyzed, showed NDELA values of 3.2 and 6.8 ppm, respectively. .111:,.~ For a popular 85 mm U.S. nonfilter cigarette we determined -43 ng of NDELA per cigarette in the sid'estream smoke, which is .{ ... .. . . . ..~ i., . . ..: =9. i• . , 7. .~. . . . ~i. ; i.+('.' T . ~.3.1~.' , Y L,.t.~ r.. generated in between puffs and'reSeased into the~y general en- t . , ' .- . : ., . , ....~. . . _ ~.... .„.~ ~..... ... ~ _~' > . ..: i• . .. . . . .k.:i , .wironment. ID!ISCLT!SSION Based on these findings one may expect that the MH-DELA Preparations sprayed a~rowth inhibitor onto the ~~~,,ssucker g~, tobacco, contain already NDELA. Table III shows that samples of. . . , . the MH-DELA preparations contained' between ©.6-1I.9 ppm of the nitrosamine. O Ch C!3 ~ od~ method developed ~ The~~ simp~~le~R r:e~pr~ucible~~ GC~-TEA~~ for the~ analysisiof N-nitrosodiethanollamine in tobacco products and' in. - 9 -

tobacco smoke demonstrated that~NDELAf in tobacco+ and its smoke derives from the treatment of tobacco with the sucker growth r :.y .- . . 4 i'~nhib~~itar male~ic~ hyd~raxide-di~ethanmlamine. ~ Alth~~o~ugh~.formula-~ ~ .NIC', -» ~~ a tlon~sf ' this widely used agricultural chemical contain~traces ' ' af NDELA, we expected~ thisnitrosamine to be primarily formed . . processing. The recommended effective dose of MH-DELA is 450! 1 JgtY A~ Sr ?3:3t~ tz i 6. n ` by nitrosatioof the amino alcohol residue duringtobacbo r mg per tmbacco plant of whiich 95 mg are DELA.''I'f one plant averages 160 g cured leaves the maximal residue on 1i kg tobacco ~:~~~ ~ thus, 1 kg cured tobacco may contain up, to 5.3'ug or 5.3 ppb of .. . ~ 1. 4 NDELA. We found in commercial U.S. tobaccae at least 108 8' ppb could be 59Omg DELA (20). The maximal amount of'NDELA found in the commercial MH-DELA preparation was 1' .9 ppm (Ta III )I ; b1c I Y ~'/^F'.~"y R £t.F_ ~y~r ~ . .t~ CVtX, ~ 6p ~-~ ;~ (Table II') , thus,, ma ximally 2.1$ of' the NDELA could derive from = the MEii-DELA preparation and most NDELA is formed during tobacco processing. This concept is also supported by the observation (3.2-6.8 ppm) than smoking tobacco does after curing (0.1-0!.4 . ~ that samples of~~ snuf~f~,~ ~which~~ went thr~ough, a~ long~ ~term; ..~ ., .. ...-.. . f~mentatio process,contai~n mueh higher amaunts0 f NDELA I ~:i F '-~t ~ . . ... ._. .... -.. .3 . . . _ . ,....,. y . _- . __. ~. ~ - - ppm).`Other corroborating evidence stems from the recent obr A, ,_ .} 4 '-+ , '" " " 'E;r servationn that levels of the carcinogenic, tobacco specific .N nitrosamines, NNN, NNK and NAT'are also significantly higher in snuf f [5. 5 to, 1:016 ppm, in 21 samples analyaed], than in 1P+ ~ cigarette tobac:co, [0.3 - 115 ppm in 151 sampl.es analyzed; 21114.Q N, .

The mainstream~smoke of U.S. cigarettes treated'iwith the I sucker growth~~ i~nhibitor~ MH-DELA: cont~~a~ine~d2~0~-4a' n~g, of NDELA per~ " cigarette. ' ` It has not been determined~as yet how much of' NDELA in .the tobacco is transfetedi into the smoke and how much 'i's pyrosynthesia~~ea3 8uring smok~ing,~. Bioassavs on rats (51.6) andl on Syrian golden hamsters (7') M, ~~~ ~. ',! have clearly demonstrated that NDELA is a relatively strong animal carcinogen. This analytical study has shown that resi- d~ual amounts~ of the~sucker growth ~~ inhib~it~or~~male~ic hydraz~~id~e- n diethanolamine on tobacco leaves give rise to"NDELA in"tobacco ~ :?:i'. _ C:! 1 and, its smoke. Snuff, which is increasingly used as a smoke between 3'~-2~-6~.~8 ppm o~'f~ ND'ELA~.~ ~ Thus,~ this~~ N-nitros~anuine!addis to, oral cavity of' its long-term: users ( 22 )', was shown to contai'n: substitute by young people and which is a carcinoge in the the carcinogenic potential of the tobacco amines in snuf f[ 5.5 to 1I© 6 ppm; 21]. specifi'c N-n'itros- Both the bioassay data for NDELA and this analytical study strongly suggest a review of the use of maleic hydraaid'e-di- ethanolamine preparations for tobacco crops. It appears that certain' potassiumisalt preparations of' MH would be' equally effective asaisuicke~~r growth;in~h~ibito~~r and could thus be~ used • in place of the diethanolamine preparations o MH (121).

ACTCNOWLBDGEMRNTS ~ ~ <. We are grateful to T.C. Tso, U.S.D.A., Beltsvi,lle, 'MD. for j ~the experimental cigarettes and expert advice. We thank S. n.".'r•}.`'~7d t+`.I ", .Adams for his technical support and C. Lettre for the modifi- .. . . , ~. cations of the GC. This study was supported by Grant 1 P01 - CA-29'5801 from , ~ the National Cancer Institute. Parts of this paper were pre- sented' at the 34'th Tobacco Chemists' Research Conference, Rich- ..~c~ mond, VA, Oct. 27-29, 1980. . . . ., . . , . .. .. . «x..e .. . l. ... .~ ~...~ ..n. ~~~. ~ti.te ~r~~'.. .. A ~ . This is No. II,xIX of "Chemical S'tudies on Tobacco Smoke." ~ ~ - . . , . . . . .J W td . .. ~ ,X{I, .,,. ,d.wl~el.i , wi,.. r: .. ...~..... t4:% . . . . -.. , ... . . . • ,.. . . _. ~. . __.. . -. ~ : -.. L.~:. . . ~.. ... .. ' ..._i :ISZ"::~. ... . . . ~ .... .. .. ~ b~ . ... ~"J..~.~.. ~. !

, - IARC~~ Monographs ~©n~~ ~~th~e Eva~luation~ o~f~ Carcinogeni~c~ Risk o Chemistry to Mani (1'978') N-Ni~trosodiethanolamine, 17 i, ^i ni 1 . `REFERENCES 'Vita, giounds; a New Class of Nitrosating and r y t . . . .. . - . , . ~ ~~:K Fine,'D.e. (1978~) , C-Nitro Com- Agents, Tox,icol. Lett. ~=J"s.~S Y ± ,. ~ (1979')1, 2-Bromo-2-nitropro- for Diethanolamine: 4; r pane-1,3-diol as a Nitrosating Agent A Model Study. Food Cosmet. Toxicol. 17,`"105-109. 'Druckrey, H., Freussmann, R., Ivankovic, S.~and Schmaehl, D.'(11967), Organotrope'carcinQgene Wirkungen'bei 65 wer- ^'a,. ~ . . . ~ . , . . , ~ schIedenen N-Nitroso-VerbindUngen an -BD-Ratten. Z. Rrebs- Reuberp- M.D1.""and Manning,; „W.B. '(19'80) , Potent Carcinogenicity of Nitrosodietlianolamine in Rats. R. and' Schmaehl, D~. (198'1i ) Dose-Re- sponse Study on the Carcinogenicity of N-Nitro~sodiethanol- amine (NDELA) ini Male Sprague-Dawley Rats. J. Cancer Res. Clin. Oncol. 9'9, A27., Biilfrich, J., Schmeltz, I, andlHoffmann, D. ('1978). El~~fecta of N-ri1itrosodie~thiano~l~amine~ and 1,1-diethanolhydra- zine in Syrian golden hamsters. Cancer-'Lett. 4, 55-6101. Q

Lij',insky, W. , Losikoff, A.M. and' Sansone, E.B. .(,1!9'S'1). Penetration ©f Rat Skin by Ni-Nitrosodiethanolamine and' u.N-Nitrosomorphoiline. ,,7'. Natl., Cancer. Inst. 66, 125-127. ,-Edwards, G.S., Peng, M., SAiegelhalder, B. and Kann, ~.. <,...,.: : :... . . . .. . . .,~,.. . ~ ~ .:. s,.' ' ,a, , . ~ ....'. -. s ..: ".;.~.~.,: i.. ! (',1i9791) Detection of N-Nitrosodiethanolamine in Humann clr:;Urine Followring, Application of a Contaminated Cosmetic. ;~~::halder, B. .09178).. yUrinar~y EXcretion of~ N-N:~itrosodli..- , ~ S ~ 2'17-222. ,~,0 . r . Preussmann, R., Wuertele, G., Eisenbrand, G. . . .. . .... . . ~.(: ethanolamine Administered Orally to Rats.Cance.r Lett. (1!981) ,,Effects of Maleic Hydrazide on Cig,arette Smoke Inhalation Toxicity in Syrian Golden Ham- United States Department of Agriculture (1979) ',"'The logic and Economic Assessment off'Maleic,Hydrazide." U.S.D.A. Techn. Bull. No. 1634, p. 106. 3 . , . . .. ::~ ~.~ PA. , p. 393'. Tsol, T.C'. (1972) '"Physiology and Biochemistry of Tobacco Plants.` Dowd~en, Hutchinson and Ross, Inc., Stroudsburg,. 14'. Steffens, G.L. (!1979) Influence of Growth Regulators and Herbicides on the Chemistry of Tobacco. Recent Adwan. To- bacco Sci.. 5, 133-1'6'3'. ~'

genic Agents in Unburned Processed Tobacco: I . .' Brunnemann, 1C «D!. ''and Wynder, E. L. .'(11975 ) Chemical studies on tobacco ~;. . ,., ..~.~. , _ smoke. XLIV. New separation techniques for classes smoke compounds. Recent Advan. Tobacco . Sci. 1: 97-122. 17. ..Brunnemann, K.D. and Hoffmann, D. (1974) The pH of Tobac- . . co Smoke. Food Cosmet. To:cicol« 12, 115-1,24 18. Bates, W.W., Griffith,R.B., Harlow, H.S., Senkus, W., and' Wakehami, H. (1968) Determination and Reporting of Total Particuliate Matter, Water in Total Particulate Matter and Nicotine. Tobacco Sci. 12, 192-196. - „ • . , ~'~`.~'-' V 19. International Comunittee for Cigar Smoke Study (1974) Oliver,,J.E..` _,(1981i)1. Agricultural Environmental Quality -,--'Machine Smoking of Cigars. Coresta Inf. Bull. 1, 31-34. .~ _, . . . . <>,. Institute, U.S'.D.A., Personal Communication. 21. Hoffmann, D. and Adams, J.D. ('11981) Carcinogenic Tobacco Specific N-Nitrosamines in Snuff and in the Saliva of Snuff Dippers.Submitted~. 22« Winn, Di.M., Blot, W.J., Shy, C. M. , PickTe, L.W., Toledo, M.A., and Fraumeni, Jr., J.F. ( 1'9'81i) Snuff Dipping and Oral Cancer Among Women in the Southern United!States. N=Nitrosodi- ~ ;: .: , ..,, -, -,:. , .. ,..~ W~ it T ,,.. . . . -Dimethylhydraaine« Cancer Lett.

ample # , ~ Tbbacco (ppb,) Mainstream Smoke . ._r :. 1us 5 . 7 . 3'0'• 1 2 ~,~: 74',°' 28 . 1 (i "e T Table I 1 J PRECISION Q OF NDELA ANALYSES* ~ 86' 8~ 2~8 3 ' : r, s1 rMean : Sigma « . y,. 1OI.6 ,Rel. Stdi. Dev. , ._. , __. 1'1!.9% tX S z; ~ . . • ... T~. . . } ... ... ...~ ..~...~~.5~ . ,~ ~ . . A . 4 Cw~ _a x: i' Y ii i n i]YC *Performed on the Kentucky IR1 Eteference cigarette. .,.. . 98.5' 36.3 99.8 0 28, .

w NITROSODIETHANOLAMINE (',NDELA)i IN TOBACCO AND TOBACCO SMOKE* 3i: Vx ~Q~~qtatil Table II -..._..,,~..~..:... , 8 9 =.3'0 . 84 ~ :51 n.d. ' .ni.d. 227. 290 19 „ 2'00 n.d. n.d. Tobacco** Mainstream (~ppb )Smoke ( ng/cig. )' , y Experimental Cigarettes Kentucky 1R1i, NF, 85 mm Kentucky 2R1, NF, 8'S mm Handsuckered 1970 Burley, NF, 85 mm MH-3'0i treated' 19'70 Burley, NF, 85 mm USDA M-6, Br~ight, pestic~~id~~e~~ treated NF, 85 mm USDA hr8, Bright, pesticide free***, NF , 8'5 mm b)Commerciial Products Cigarette A, NF, 85 mm Cigarette B, F', 8'5' mm French Cig. C, NF, 70 mm n.d. .., II,ittle Cigar, F, 85 mm .._.~.~.__y mm Large Cigar (7.7 g), NF, Y 1125, Snuff (Fine Cut), aged Snuf'f ( Fine Cut) , fresh Chewing To~~b~acco~~ A ~ Chewing Tobacco B 4' 19 108 ~ 3180 684!0 2'85 224 ni.d. = not detected (detection limit, see "Results") NF = non-filter F = f ilter *Data corrected for recovery using the isotope dilution nnethod. **Tobacco data are reported~ per dry weight. ***Otown in Prince Edward~ Islandi, Canada. ****S!idestream Smoke: 43' ng/cig.

i j~ s CS!'J : Table III' N'D!ELA IN MH-DELA PREPARATIONS Sample # ( year ) .._ ~ w . __. NDELl~ ('ppm ) (119'75')' 0.73 "059 1.~ c r.,a. ~"•v s c .., A, asf ,S...~aoaol T f ,93 ,

Modification of the GC-TEA Interface A: tatalytic FyrolyzerR B: Heated! Interface; l'; Graphite Ferrule, D: GC-Column. Gas Chromatograms of NDELA ('TMS ) a, .:~-Y... - . . . . ,~. *..y : ,. YYY! . '` r Mass Spectra Reference and Isolates from Tobacco and Tobacco Smoke Showing N~-Nitrosodiethanolamine as `the Di-Trimethylsilyl Derivative. , _ . _.

s 1A3T-„i ~ ' . .L. . .+/ . . t. ~ ~'1 o rl~~i:_ .~.I!.~:~.. ' t- I, .f Ag lJ e~J ~ r 'a . ., . . .. . h T~ 31~ A B i ^a, 4 -- -- - -- - - ~ - - -- -- - - - - Figure 1. .~ ; Y . Y nt y

/ .SS .~;.,.<... c.... Figure 2'. -H-~minutesl--~-{':.. I---I ~- 0 2 4 0 2 4 01 2 4 0.2 ...4 „ 4 NDELA (TMS),r ~r KENTUCKY' IRI KENTUCKY' IRI MH -34 ~' STANDARD TOBACCO MAINST!REAM' SOLUTIONi 3.21' ng EXTRACT' , SMOKE ( 1 m © 0 r. / V a. ON ! ,. W .~J h. 5 Q 5'

,~ '~q ~ ISOLATED, TOSACCO, 1',ii l,i '- ISOLATED,I TOBACCO SlViIOKE L I I ~sw , a 3s~~' _F~.~.- . . 20 40 E0' 80 10Q7 1~20 140' 160 1!80 C00 ~20 24Q ~M~ m~e 100' 1 ~ 73' 6O~REFEREMCE 40 ~ ~° 20 :~ 0 f (CHa)3 S'i'-0!-(CHj)2~, N-N=O' k' (CHa )a SJ-0"(CH2 )2~ 1oj /!6' 130 147 ' 263 ; 9 ,