effect of storage conditions on nitrosated, acylated, and ... · [cancer research 49, 5895-5900,...

[CANCER RESEARCH 49, 5895-5900, November 1, 1989)

Effect of Storage Conditions on Nitrosated, Acylated, and Oxidized PyridineAlkaloid Derivatives in Smokeless Tobacco Products1

Roger A. Andersen,2 Harold R. Burton, Pierce D. Fleming, and Thomas R. Hamilton-Kemp

United States Department of Agriculture, Agricultural Research Service [R. A. A., P. D. F.], Department of Agronomy fR. A. A., H. R. B., P. D. F.J, and Department ofHorticulture [T. R. H-K.], University of Kentucky, Lexington, Kentucky 40546

ABSTRACT

Very large concentration increases in nitrite (34-fold), nitrosated pyr-idine alkaloids, and related 4-(methylnitrosamino)-l-(3-pyridyl)-l-butan-one (NNK) (14- to 33-fold) occurred in moist snuff during storage at24Â°Cfor 52 weeks, whereas, decreases in all parent and some acylated

pyridine alkaloids were observed in the same material. Nitrite concentrations in dry snuff decreased up to 90% during storage; increased contentsof nitrosated alkaloids and NNK of 30 to 80% were also observed.Storage effects on chewing tobacco included a 75% increase in nitrite andsmall increases of nitrosated alkaloids and NNK. Sums of parent alkaloids in moist snuff decreased 24 and 54% after storage for 24 weeks at24 and 32Â°C,respectively, while sums of alkaloid derivatives increased,up to 36-fold for nitrosated alkaloids and NNK, 92% for acylated, and133% for oxidized components. Levels of A"-nitrosonornicotine, NNK,and ,\"-nit rosoiiimtabim>after 52 weeks' storage at 24Â°Cwere 547, 41,

and 296% higher, respectively, in ambient air-exposed moist snuff thanin the nonexposed counterpart.

A mathematical model was evaluated and used to calibrate nonlineargas chromatography alkali bead detector response to the individualcomponents. This permitted the use of a single analysis for all requiredindividual compounds over a wide concentration range.

INTRODUCTION

EpidemiolÃ³gica! investigations have indicated that the use ofsnuff, particularly the practice of snuff dipping, results in ahigher incidence of oral cancer in humans (1). Snuff, tobaccojV'-nitroso-pyridine alkaloids, and related NNK3 in snuff in

duced oral cavity tumors in F344 rats (2).Several commercial U. S. smokeless tobacco products includ

ing snuff and chewing tobaccos were analyzed for nitrosatedpyridine alkaloids and were found to contain 3-90 Â¿ig/gNNN,0.1-8 Mg/g NNK and 0.5-7 Mg/g NAT plus NAB (3). In similarstudies of several U. S. snuff brands, products contained up to64 Mg/g NNN, 14 Mg/g NNK, 215 Mg/g NAT and 7 Mg/g NAB(4, 5). Other suspected carcinogenic agents analyzed in thelatter investigations were the volatile nitrosamines NDMA,NPYR, NMOR, and NDELA which were either not detectedor present at less than 0.2 ng/g, the volatile aldehydes, i.e.,formaldehyde, acetaldehyde, and crotonaldehyde, which wereall present at less than 8 Mg/g>traces of PAH documented bythe presence in some of these products of up to 0.09 ^g/g BaPas an indicator of PAH and 1.22 pCi/g of 2lÂ°Po,an a-particle

emitter known to deposit in and irradiate soft tissues. It appeared probable that nitrosated pyridine alkaloids were themost abundant carcinogens in these smokeless tobaccos.

Received 3/27/89; revised 7/5/89: accepted 8/4/89.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1This investigation is in connection with a project at the Kentucky AgriculturalExperiment Station. The results are published as KAES Journal Article 89-3-66.

2To whom requests for reprints should be addressed.'The abbreviations used are: NNK, 4-(methylnitrosamino)-l-(3-pyridyl)-l-

butanone; NNN, W-nitrosonornicotine; NAT, A"-nitrosoanatabine; NAB, N'-

nitrosoanabasine; PAH, polynuclear aromatic hydrocarbons; GC, gas chromatography; FNN, W-formylnornicotine; FAT. JV'-formylanatabine; ANN, W-acetyl-nornicotine; AAT, /V'-acetylanatabine; BNN, A"-n-butanoylnornicotine; HNN,W-n-hexanoylnornicotine; ONN, W-n-octanoylnornicotine.

Effects of prolonged storage on contents of the nitrosatedalkaloids NNN and NNK and several of their nitrogenousprecursors in leaves of cured burley tobacco genotypes varyingin total alkaloid content and a commercial cultivar KY 14 werereported (6, 7). yV'-Acyl-pyridine alkaloids were recently quan

tified in leaves of these same burley tobaccos during field growthand air curing (8). It was postulated that acylation, nitrosation,and oxidation reactions competed for the same available parentpyridine alkaloid precursors during the curing and postharvestprocessing of tobacco (8). Concentrations of NNN, NNK, andNAT in a snuff product increased after removal of the productfrom an air-tight aluminum envelope and exposure to ambientroom air for up to 14 days (9). Effects of prolonged storage onconcentrations of alkaloid nitrosamines and other derivativesin smokeless tobacco products have not been documented,however.

Recent analyses of nitrosated and acylated pyridine alkaloidsin tobacco performed in our laboratory (6, 8, 10, 11) haveinvolved the use of a GC detector utilizing an alkali (rubidium)bead of the type introduced by Kolb and Bischoff in 1974 (12).The use of this nitrogen-specific detector coupled with a fusedsilica capillary column permitted simultaneous GC analyses ofup to about 15 nitrosated and acylated pyridine alkaloids. Inour initial studies, nonlinear response was a problem whichnecessitated frequency detector calibrations for multilevel concentrations at narrow concentration ranges of about 0.5 ng ofsought compound.

In the present study, effects of prolonged storage conditionson nitrosated and other pyridine alkaloids and nitrite nitrogenwere determined in reference smokeless tobaccos which variedin tobacco composition and moisture content. Comparisons ofnitrosamine and other alkaloid derivative contents were madeamong the three smokeless tobaccos. An investigation of theGC alkali bead detector responses to pyridine alkaloid derivatives was simultaneously carried out to simplify componentcalibrations and extend the range of nonlinear calibration byuse of an appropriate mathematical model.

MATERIALS AND METHODS

Smokeless Tobacco Products and Storage Conditions. Three referencesmokeless tobacco research products, namely, loose-leaf chewing tobacco, (1S1), dry' snuff (1S2), and moist snuff (1S3) were obtained from

the Tobacco and Health Research Institute, University of Kentucky,Lexington, KY. The moisture content and principal components of thetobacco products were: 1S1, moisture (20%), Wisconsin air-cured leaf(17%), Pennsylvania air-cured leaf (15%), crushed burley stems (6%);1S2, moisture (8.7%), dark air-cured stem (33%), dark-fired leaf (23%),fire-curved Virginia leaf (20%), flue-cured stem (15%); 1S3, moisture(53%), dark fire-cured leaf (26%), dark air-cured leaf (7.8%), burleystems (3.7%). Packagings used were: 1S1, 3-oz foil pouch; 1S2, 4.6-ozmetal can; 1S3, 1.2-oz plastic can.

Approximate one-half pound lots of each reference product wereremoved from their packagings and transferred to 1-liter mason jarswith rubber sealed screw tops for storage in the dark in electronicallycontrolled environment chambers at 24 Â±1Â°Cand 32 Â±1Â°Cfor 24 and

52 weeks beginning in November, 1986. Four replicate samples (one

5895

on April 13, 2017. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

ALKALOIDS IN SMOKELESS TOBACCO

per jar) were used for each storage treatment. Seals were broken at 2-week intervals throughout storage and the tobacco samples were remixed during a time period of about 30 s. The jars were then resealedimmediately. In addition, four original packaging containers of eachreference product were kept in their original sealed containers andstored for 52 weeks at 24Â°Cor 32Â°C.The contents of one container

was considered as one replicate sample. Fifteen-g samples were removedat the beginning of storage (0 time) and after the desired storage period,placed in paper bags, and then frozen at â€”¿�70Â°Covernight. The bag

contents were then freeze dried, ground to 100-200 mesh size, andequilibrated to ambient moisture overnight in darkness at ambienttemperature on a laboratory bench. All samples were then stored insealed plastic containers at â€”¿�70Â°Cuntil analyzed.

Reference Compounds. Nicotine, confinine, and 2,3'-dipyridyl were

obtained from Aldrich (Milwaukee, Wl). Nornicotine and NNK wereobtained from Chemsyn Science Laboratories (Lenexa, KS). Anatabine,NNN, NAT, FNN, FAT, ANN, AAT, BNN, HNN, and ONN weresynthesized as previously described (8, 10).

Chemical Analyses. NNK, Â¿V'-nitroso-and Â¿V'-acylpyridinealkaloids,and oxidized pyridine alkaloids, i.e., confinine and 2,3'-dipyridyl were

extracted, partitioned, and analyzed by GC analysis as previouslydescribed (8), but with minor modification. Briefly, 1-g tobacco sampleswere extracted with citrate-phosphate buffer containing ascorbic acidat pH 4.5. After extraction, the mixture was adjusted to pH 5.0 andpartitioned with ethyl acetate. The ethyl acetate phase was subsequentlypartitioned with l N HC1. The acidic aqueous phase was adjusted topH 5.0 before final partitioning with chloroform. An aliquot of thechloroform phase containing an added and known amount of azoben-zene was injected into a Hewlett-Packard Model 5880A GC in splitlessmode utilizing a 30-m x 0.25-mm fused silica DB-5 capillary column,temperature programming, and a nitrogen-specific detector. Identifications of Chromatographie peaks obtained with tobacco samples werebased on cochromatography in the quantitative GC analysis and confirmation of GC-mass spectrometry fragmentation patterns with authentic compounds. For the latter purpose a Finnegan 705 ion trapdetector and a Hewlett-Packard 5985A system in electron impact modewere used with GC conditions similar to those used for the quantitativeanalysis. Quantitations were carried out by internal standardizationwith azobenzene as before (8), except that response factors for pyridinealkaloid derivatives were determined by these steps: (a) detector responses using a given bead were plotted versus seven or more amountsof a standard compound ranging from 0 to 1500 ng that were injectedinto the GC, (b) curvilinear regression (13) was used to determine thebest fit mathematical function to describe the resultant empirical curve;in our experiments this was the power model y = ox*,and (c) a response

factor was obtained by taking the derivative of the mathematical modelfor a given detector and bead response. Development of the rationalefor using an appropriate model is described in "Results and Discussion." Corrections were made for recoveries of alkaloid related com

pounds carried through the entire extraction procedure as previouslydescribed (8). Reported recoveries ranged from 16 to 99% (8, 10).

Parent (major) alkaloids, i.e., nicotine, nornicotine, and anatabinewere determined by a capillary GC procedure utilizing an FID ornitrogen-specific detector (14). Nitrite was determined by a spectropho-tometric procedure (15). Calcium was determined as previously described (8). Quantitative results for chemical components in tobaccoduring storage were normalized on the basis of calcium content at thestart of storage. Results were subjected to an analysis of variance.

RESULTS AND DISCUSSION

GC QuantitÃ¤ten using Curvilinear Regression. Least squaresregressions of values obtained for amounts of standard alkaloidderivatives injected into the GC and the corresponding peakareas obtained with a specific detector bead were performedusing the following mathematical models: (a) polynomial ofdegree n - 1-5, (b) parabolic, (c) power, (d) exponential, (e)logarithmic, and (/) hyperbolic. In all cases, the highest correlations (r2 values) were obtained using the power model y =

ox6. This model was used to predict known standard alkaloid

derivative concentrations within Â±2.5% of actual values. Valuesfor constants a and b were calculated from at least seven x-ypaired values that were determined empirically (ng compoundinjected into the GCâ€”GC peak are unit response) over a rangeof 0-1000 ng of a specific compound. Computations wereperformed from a logarithmically transformed expression ofthe power model, i.e., Iny = blnx + ina. For example, theconstants obtained for responses of two detector beads to standard NNK were calculated from paired values of x-y. Thefollowing values were obtained: Bead 1, a = 69.2, b = 1.18based on x-y paired values (23.4-2,855; 46.8-6,467; 93.6-14,652;188-33,258;375-75,347;750-170,703;l,500-386,735);Bead 2, a = 287.8, b = 0.59 based on paired values (not shown).Both beads exhibited nonlinear responses. Regression curveswere plotted with the aid of a graphics computer program usingthe power model and the calculated constants for given beadswith each standard alkaloid derivative. These curves were usedto calibrate contents of alkaloid derivatives in tobacco samples.

The development and use of regression curves of GC detectorresponses versus content of a specific pyridine alkaloid usingthe power model permitted a single simultaneous analysis ofall the required individual compounds over a wider concentration range than could be used previously (6). This was anadvantage in the present study because about 15 compoundswere determined in a single analysis.

Nitrite, Nitrosated, and Parent Alkaloids. Results for effectsof storage duration at 24"C on contents of nitrite, nitrosated,

and parent alkaloids are given in Table 1. Trends are similar tothose obtained at 32Â°C.

Very large increases (up to 34-fold) in nitrite concentrationsoccurred in moist snuff stored at 24Â°Ccompared to 0 time

controls. Accumulation was maximal at 24 weeks, and thendecreased during the final period. In contrast, decreases innitrite were found in stored dry snuff. Maximal nitrite nitrogencontent in stored moist snuff was 4,456 /Â¿g/g,while highestcontents in dry snuff and chewing tobacco were much lower.

Each nitrosated alkaloid showed very large increases in content in moist snuff after 24 and 52 weeks' storage. The range

of individual nitrosamine concentrations after storage of moistsnuff was 67 to 989 ^g/g, and maximal individual nitrosamineaccumulations were 32-, 13-, and 15-fold higher than 0 timestorage values for NNN, NNK, and NAT, respectively. Although 0 time levels of individual nitrosamines in dry snuffwere larger than in moist snuff, much smaller increases inNNN, NNK, and NAT occurred in dry snuff after storage.Increased magnitudes of nitrosated alkaloids were found instored chewing tobacco similar to those in dry snuff. However,individual nitrosamine concentrations in stored chewing tobacco were much smaller.

Each parent alkaloid in moist snuff declined in content duringstorage and was lowest after 52 weeks' storage. Nicotine, the

most abundant alkaloid, declined 39%; nornicotine and anatabine levels decreased 76 and 95%, respectively. There was ahighly significant negative correlation (r = â€”¿�1.0)between con

tents of nicotine and NNN during storage of moist snuff.Although little or no changes in nicotine concentrations wereobserved in stored dry snuff and chewing tobacco, there weredeclines of nornicotine (12%) in dry snuff, and declines ofnornicotine (24%) and anatabine (15%) in chewing tobacco.

The larger changes in nitrite, individual nitrosated, and parent alkaloids in moist snuff compared to dry snuff or chewingtobacco during storage (Table 1) could have resulted fromdifferences in the tobacco blend formulations or the much

5896



ALKALOIDSIN SMOKELESS TOBACCO

Table 1 Effect of storage at 24Â°C on concentrations ofnitrosated and parent pyridine alkaloid and nitrite in smokeless tobaccos

Pyridine alkaloids and derivatives(^g/g)Â°Product

(%moisture)Moist

snuff(53)LSD.05*Dry

snuff(8.9)LSD.05Chewing

tobacco(19)LSD.05LSD.05

(overall)Storegeduration(weeks)024520245202452NO2-N,(KgVgf132.54,455.71,281.7105.477.013.37.63.41.62.81.50.255.2NNN19.4413.7641.419.4116.1169.4157.213.92.84.43.20.411.2NitrosatedNNK5.466.973.93.384.4151.8145.910.10.01.00.40.25.5NAT62.2989.2850.458.5238.8329.8296.316.412.417.39.20.631.8Nicotine19,44015,23911,76677110,96011,97411,9303536,8157,7367,3341,058710ParentNornicotine326.796.676.811.1255.3269.5224.026.0295.0333.9225.420.418.3Anatabine296.116.513.95.7191.3208.8196.5NS'229.2270.2195.615.615.2" Each entry is the mean value obtained by analysis of four replicate samples.* LSD.05, least significant difference (P = 0.05).c NS, not significant.

Table 2 Effect of storage at 24'C on acylated and oxidized pyridine alkaloid concentrations in smokeless tobaccos

Pyridine alkaloids and derivatives(Â¿ig/g)Â°Product

(9Ãmoisture)Moist

snuff(53)LSD.05*Dry


tobacco(19)LSD.05Storageduration(weeks)024520245202452LSD.05

(overall)FNN248.4188.2219.810.3197.8260.2244.814.190.0125.058.16.39.7ANN69.9433.0595.628.139.657.359.83.421.429.014.71.914.9BNN3.32.44.10.34.97.26.60.40.61.70.80.20.3AcylatedHNN26.810.713.00.714.519.017.21.15.17.83.90.40.7OxidizedONN46.714.418.91.126.132.829.02.27.612.25.50.91.4FAT149.286.280.76.296.2119.7108.55.439.751.325.42.04.5AAT25.712.121.94.119.524.621.41.23.94.92.70.32.3Cotinine952.1945.21151.347.8889.61155.81245.4146.2327.5442.7240.622.381.42,3'-Dipyridyl935.71608.92260.194.61004.61411.81376.782.7533.9721.8489.627.467.4*Each entry is the mean value obtained by analysis of four replicate samples.*LSD.05, least significant difference (P = 0.05).

Table 3 Effect of storage temperature and duration on sums of parent and derivatized pyridine alkaloid concentrations in smokeless tobaccos

Storage mode S of pyridine alkaloids and derivatives (jig/g)0

Product (%moisture)Moist

snuff(53)LSD.05*Dry



(overall)Duration

(weeks)024245252024245252024245252TemperatureCO24/322432243224/322432243224/3224322432Parent19,98515,2409,10311.92413,49891111,20012,00312,38412,13111,522NS7,1237,8318,3737,5335,951403769Nitrosated87.01,469.83,108.31.565.72,356.4129.1439.2651.0573.3599.4433.634.215.222.718.212.812.10.873.1Acylated570.0747.11,094.7954.11,077.439.6398.7520.6462.2487.4355.024.1168.2231.8179.6111.2108.07.425.7Oxidized1,887.82,554.14,389.43,411.43,985.7164.11.894.22,567.62,522.02,622.12,084.

1168.2861.41,164.51,075.2730.1735.139.2130.7

" Each entry is the mean value obtained by summations of analyses for individual compounds of an alkaloid type in four replicated samples.* LSD.05, least significant difference (P = 0.05); NS = not significant.

higher moisture (53%) in moist snuff compared to the 8.9 and19% moisture contents of dry snuff and chewing tobacco,respectively. Prolonged storage effects on chemical componentconcentrations are similar to air curing effects in many ways.In an earlier study, nitrite, NNN, NNK, and NAT concentrations were elevated to a much larger extent in hurley tobaccocured under controlled environments for 20 days at 32Â°C-83%relative humidity, than after curing at either 24Â°C-70%relative

humidity or 15Â°C-50%relative humidity (11, 16). Nitrosated

alkaloid composition changes of a similar nature occurredduring short-term storage of these tobaccos.4 Although the

effects of temperature were not separable from those of relativehumidity in these former studies, our present results related to

4 H. R. Burton, L. P. Bush, and M. V. Djordjevic. Influence of temperature

and humidity on the accumulation of tobacco specific nitrosamines in storedburley tobacco, submitted for publication.

5897




temperature effects (Table 3) do not support the view that thesecomponent composition changes are caused solely by temperature variations. Also, total alkaloid concentrations in homogenized leaf-cured hurley tobacco were smaller after curing at16% moisture than after curing at 12 or 8% moisture (7). Weare presently investigating controlled environmental storageeffects on moist snuff reference tobaccos with different levelsof moisture at the same temperature. The nicotine, total nitrogen, and nitrate nitrogen content of dry snuff in our presentstudy was greater than that of moist snuff or chewing tobaccoat the onset of storage. Since these components could serve asprecursors of nitrite or nitrosated alkaloids, it does not appearthat their initial concentrations fully account for the largerchanges of nitrite, nitrosated alkaloids, and parent alkaloids inmoist snuff that we observed. Moist snuff contains dark air-cured tobacco leaf, whereas, dry snuff contains dark air-curedstem. There are several other tobacco type compositional differences among the smokeless tobacco blends, and undoubtedly,these have some effect on the storage-related changes in chemical composition. We believe, however, that moisture contentdifferences contribute most significantly to this effect.

Our results (Table 1) indicated that at some time periodduring storage, a decrease in nitrite from earlier levels tookplace in each smokeless tobacco product. Highest nitrite concentrations were found in moist snuff and chewing tobacco afterthe 24-week storage period, whereas, that in dry snuff was at 0time storage. The tendency for eventual declines in nitriteduring storage is also consistent with results we obtained earlierthat showed nitrite disappearance in homogenized leaf-curedtobacco was a function of increased storage time ( 17). Nitritein postharvest or processed tobacco is believed to form and mayaccumulate from the action of bacterial nitrate reducÃaseonnitrate in the tobacco. Nitrate nitrogen contents in the smokeless tobaccos used in the present study ranged from 2 to 7 mg/g on a wet weight basis at the beginning of the storage treatments. The reduction reaction occurs under some specific conditions, e.g., presence of specific bacterial genera and strains(18, 19), anaerobiosis (18, 19), high nitrate concentrations (19),and a slightly acid pH (19). It is probable that some or all nitritemay be further reduced under specific conditions by bacterialnitrite reducÃasein denitrificalion reaclions during Ihe postharvest processing of tobacco (18, 19) and lhal denilrificalion maycompete with nilrosalion of parenl pyridine alkaloids duringformalion of NNN, NNK, and NAT. The demonslraled lend-ency for nilrile concenlralions lo sometimes change in differentdireclions and rales during slorage compared lo concenlralionsof nilrosaled pyridine alkaloids, makes il necessary lo determineaccumulÃ¢tions of nilrosaled alkaloids direct ly for purposes ofquali iy conlrol of tobacco producÃs.Reductions of carcinogenicnitrosamine formalions in smokeless lobaccos may be feasibleby Iwo approaches lhal are based on Ihe resulls of Ihe presenlsludy, namely, (a) conlrol of Ihe produci moislure conlenlduring slorage, and (b) pasleurizalion of Ihe lobacco followedby slorage in slerilized and sealed conlainers.

Acylated and Oxidized Alkaloids. Resulls for effecls of slorageal 24"C on conlenls of acylaled and oxidized alkaloids in

smokeless lobaccos (Table 2) showed Irends lhal were similarlo Ihose for storage al 32Â°C.Concenlralions of acylaled com

pounds generally decreased in moisl snuff during slorage exceplin Ihe case of ANN. The general decrease in concenlralions ofacylaled alkaloids (excepl for ANN) in moisl snuff duringslorage (Table 2) conlrasls wilh Ihe irend for NNK and eachnilrosaled alkaloid (Table 1). Since Ihe nilrite content of moistsnuff was elevated significanlly during slorage, it may be spec

ulated thai acyl substiluted alkaloids in moist snuff may haveserved as precursors for nilrosaled alkaloids in reaclion sequences similar lo lhal delermined for Ihe lerliary amino groupin nicoline (20). Il should be noled, however, lhal Ihe biochemical reaclions affecling acylaled pyridine alkaloid synlhesis anddegradalion are noi known al presenl (8). Increased conlenlsof oxidized compounds occurred during slorage of moist snuff.

In conlrasl lo resulls for moisl snuff, concenlralions of acylaled compounds increased in dry snuff during slorage; increased conlenls of oxidized compounds were also observed.Each acylaled and oxidized alkaloid in chewing lobacco increased in concenlration at a slorage duralion of 24 weeks,compared lo Iheir respeclive 0 time conlrols. Further slorage,however, caused a decline in ine accumulalion of each corresponding alkaloid derivalive lo levels lower lhan al O lime(excepl for BNN). The generally small increases in concenlralions of acylaled alkaloids lhal occurred in dry snuff and chewing lobacco during slorage may indicale lhal Ihe acylaled alkaloids conlribuled much less lo nilrosamine formalion andaccumulalion in Ihese smokeless lobacco producÃslhan was Ihecase for moisl snuff.

Colinine and 2,3'-dipyridyl were Ihe most abundant alkaloid

related compounds excepl for nicoline measured in each smokeless lobacco al all slorage duralions. Among acylaled alkaloids,FNN, ANN, ONN, and FAT were Ihe mosl abundanl.

Sums of Alkaloids by Type. Resulls for effecls of slorageduralion and slorage lemperalure on conlenls of sums of parenl,nilrosaled, acylaled, and oxidized alkaloids are given in Table3. Compared lo O lime conlrols, concentrations of sums ofparent alkaloids in moist snuff decreased after slorage. Inconlrasl, sums of nilrosated, acylaled, and oxidized alkaloidsin moisl snuff increased during slorage; moreover, slorage al32Â°Ccaused larger accumulÃ¢tions of Ihese alkaloids lhan slorage at 24Â°C.Sums of nitrosated alkaloids increased 16- and 35-fold in moist snuff after slorage for 24 weeks al 24 and 32Â°C,

respeclively. Corresponding increases after slorage for 52 weekswere 17- and 26-fold. Sums of nitrosated pyridine alkaloidschanged (increased) most among the alkaloid groups.

In conlrasl to results from moisl snuff, concentralions ofsums of parenl alkaloids in dry snuff did noi change significanlly among Ihe slorage duralion-lemperalure irealmenls. Aswas Ihe case for moisl snuff, sums of nilrosaled, acylaled, andoxidized alkaloids in dry snuff generally increased during slorage; sums of nilrosaled alkaloids increased 48 and 31% afterslorage for 24 weeks at 24 and 32Â°C,respectively; a corresponding 36% increase occurred after 52 weeks al 24Â°C.Thus, nilro

saled alkaloids underwenl much smaller concentration changesin dry snuff than in moist snuff.

Sums of nitrosaled alkaloids in chewing lobacco increasedafter storage for 24 weeks, whereas small decreases occurredafter 52 weeks.

Our results thai demonslrale a slorage lemperature effecl (24versus 32Â°C)on changes in sums of alkaloids in moist snuff

(Table 3) are generally consistenl wilh Ihe resulls we oblainedearlier for changes in NNN conlenls in air-cured4 and homogenized leaf-cured hurley tobacco slored al 20 and al 30Â°C(7).

In Ihe cases of dry snuff and chewing lobacco, Ihe Irends ofresulls for slorage lemperalure-dependenl changes of sums ofparenl alkaloids are less clear and consislenl wilh earlier resulls.Sums of nilrosaled and acylaled alkaloids in Ihese lobaccostended to be lower or unchanged at Ihe 32Â°Cslorage lemperalure lhan al 24Â°C.

Ambient Air Exposure. Component concentrations in smokeless tobaccos stored for 52 weeks at 24Â°Cwith periodic exposure

5898



ALKALOIDSIN SMOKELESS TOBACCO

Table 4 Effect of ambient air exposure during storage on nitrite and pyridine alkaloid contents in smokeless tobacco stored 52 weeks at 24Â°C

Product (%moisture)Moist

snuff(53)LSD.05*Dry



(overall)Status

of exposure to ambient air duringstorageExposedSealedExposedSealedExposedSealedPyridine

alkaloids and derivatives(*ig/g)Â°NOj-N(i"g/g)1,281.71,492.7292.47.66.80.51.51.4NS158.2NNN641.499.130.5157.2123.615.23.22.9NS17.4NitrosatedNNK73.952.3NS'145.9108.411.50.40.5NS85.7NAT850.4214.5129.7296.3241.618.09.28.8NS70.92

ofParent11,92414,4041,58512,13111,576NS7,5336,3918751,9411of Oxidized3,411.44,373.7238.72,622.12,034.4197.2730.1751.5NS168.62

ofAcylated954.0729.055.1487.9386.326.2111.2109.0NS33.2

" Each entry is the mean value obtained either by analyses of a component in four replicate samples or by summations of analyses for individual components of an

alkaloid type in four replicate samples.* LSD.05, least significant difference (P = 0.05).c NS, not significant.

to fresh ambient air are compared to those in tobaccos storedin their original sealed packagings without exposure to freshlyintroduced ambient air (Table 4). Trends are generally similarto those obtained for storage at 32"C.

Nitrite-nitrogen content in moist snuff exposed to freshlyintroduced ambient air during storage did not differ significantly from moist snuff stored in sealed containers. In general,accumulations of nitrosated and acylated alkaloids were higherin moist snuff stored for 52 weeks with exposure to freshambient air than without such exposure. Thus, levels of NNN,NNK, and NAT after 52 weeks' storage were 547, 41, and

296% higher, respectively, in fresh ambient air-exposed moistsnuff than in nonexposed counterparts. Acylated alkaloids afterstorage were higher in ambient air-exposed than in nonexposedmoist snuff, whereas parent and oxidized alkaloids were somewhat lower.

Exposure of dry snuff to ambient air during storage resultedin higher 52-week storage levels of nitrite-yv, individual nitrosated alkaloids, and sums of oxidized and acylated alkaloidsthan were determined in nonexposed (sealed) dry snuff. Sumsof parent alkaloids were not significantly different, however.Comparison of component concentrations in chewing tobaccoafter storage in the presence or absence of freshly introducedambient air during storage revealed no significant differences.

The different results for smokeless tobacco materials in Table4 reflect effects of storage in different atmospheres. Thus,exposure of tobaccos to ambient air periodically probably provided contact with microorganisms that infected exposed tobacco and caused a chemical change unlike that in the sametobacco stored in sealed containers with little or no air exchange. The atmospheres may have also differed in contents ofoxygen and other component gases. The rationale for investigating effects of ambient air exposure was based on the probablerole of specific bacteria in the reduction of endogenous nitratein tobacco to nitrite or other reduction products and a furtherpossible role of anaerobiosis on reduction of nitrate and otherchemical reactions (18,19). Indeed, the elevations of nitrosatedalkaloid concentrations in moist and dry snuffs and the parentalkaloid decrease in moist snuff after storage in atmospheresadmixed with ambient air suggest that significant changes inbacterial flora or atmospheric composition took place. Thelarger accumulations of NNN in moist and dry snuff storedwith exposure to ambient air compared to NNN contents inthese tobaccos stored in sealed containers may indicate thefeasibility of limiting nitrosamine buildup in tobacco productsby technology that controls bacterial entry and growth duringall phases of storage.

ACKNOWLEDGMENTS

The authors thank T. G. Sutton, U. S. Department of Agriculture,Agricultural Research Service, University of Kentucky, and A. H.Vaught of the Tobacco and Health Research Institute, University ofKentucky for their skillful contributions in preparing tobacco materials.We are grateful to J. D. Crutchfield and T. A. Becherer for analyticalservices.

REFERENCES

1. Winn, D. M., Blot, W. J., Shy, C. M., Pickle, L. W., Toledo, M. A., andFraumeni, J. F., Jr. Snuff dipping and oral cancer among women in theSouthern United States. N. Engl. J. Med., 304: 745-749, 1981.

2. Hecht, S. S., Rivenson, A., Braley, J., DiBello, J., Adams, J. D., andHoffmann, D. Induction of oral cavity tumors in F344 rats by tobacco-specific nitrosamines and snuff. Cancer Res., 46:4162-4166, 1986.

3. Hoffmann, D., Brunnemann, K. D., Adams, J. D., and Hecht, S. S. Formation and analysis of A'-nitrosamines in tobacco products and their endogenousformation in tobacco consumers. In: I. K. O'Neill, R. C. von Borstel, J. E.Long, C. J. Miller, and H. Bartsch (eds.). ^-Nitroso Compounds: Occurrence,Biological Effects and Relevance to Human Cancer, pp. 743-762. 1ARCScientific Publications No. 57. Lyon, France: International Agency for Research on Cancer, 1984.

4. Hoffmann, D., Harley, N. N., Fisenne, J., Adams, J. D., and Brunnemann,K. D. Carcinogenic agents in snuff. J. Nati. Cancer Inst., 76:435-437, 1986.

5. Hoffmann, D.. Adams. J. D., Lisk, D., Fisenne, I., and Brunnemann, K. D.Toxic and carcinogenic agents in dry and moist snuff. J. Nati. Cancer Inst.,79:1281-1286, 1987.

6. Andersen, R. A., and Kemp, T. R. Accumulation of 4-(Ar-methyl-A'-nitrosa-mino)-l-(3-pyridyl)-l-butanone in alkaloid genotypes of hurley tobacco during postillinosi processing: comparisons with A"-nitrosonornicotine andprobable nitrosamine precursors. Cancer Res., 45: 5287-5293, 1985.

7. Andersen, R. A., Kasperbauer, M. J., Burton. H. R., Hamilton, J. L., andYoder, E. E. Changes in chemical composition of homogenized leaf-curedand air-cured burley tobacco stored in controlled environments. J. Agrie.Food Chem., 30: 663-668, 1982.

8. Andersen, R. A., Fleming, P. D., Burton, H. R., Hamilton-Kemp, T. R., andSutton, T. G. .V Ai \ I and iV'-nitroso-pyridine alkaloids in alkaloid lines ofburley tobacco during growth and air curing. J. Agrie. Food Chem.. 37: 44-50, 1989.

9. Hoffmann, D., and Adams, J. D. Carcinogenic tobacco-specific A'-nitrosamines in snuff and in the saliva of snuff dippers. Cancer Res., 41: 4305-4308, 1981.

10. Burton, H. R., Andersen, R. A., Fleming, P. D., and Walton, L. R. Changesin chemical composition of burley tobacco during senescence and curing. 2.Acylated pyridine alkaloids. J. Agrie. Food Chem., 36: 579-584, 1988.

11. Burton, H. R., Childs, G. H., Jr., Andersen, R. A., and Fleming, P. D.Changes in chemical composition of burley tobacco during senescence andcuring. III. Tobacco specific nitrosamines. J. Agrie. Food Chem., 37: 426-430, 1989.

12. Kolb, 11.. and Bischoff, J. A new design of a thermionic nitrogen andphosphorus detector for gas Chromatograph). J. Chromatogr. Sci., 12: 625-629, 1974.

13. Sokal, R. R., and Rohlf, F. J. Biometry. Multiple and curvilinear regression,pp. 617-690, Ed. 2. San Francisco: W. H. Freeman and Co., 1981.

14. Severson, R. F., McDuffie, K. L., Arrendale, R. F., Gwynn, G. R., Chaplin,J. F., and Johnson, A. W. Rapid method for the analysis of tobacco nicotinealkaloids. J. Chromatogr., 21: 111-121, 1981.

15. Crutchfield, J., and Burton, H. R. Improved method for the quantificationof nitrite in plant materials. Anal. Lett., 22: 555-571, 1989.

5899




16. Andersen, R. A.. Burton. H. R.. Fleming, P. D.. Hamilton-Kemp. T. R.. andGay, S. S. Effects of air-curing environments on alkaloid-derived nitrosa-mines in hurley tobacco. In: H. Bartsch, I. K. O'Neill, and R. Schulte-I In m.iim (eds.). Relevance of A'-Nitroso Compounds to Human Cancer:Exposures and Mechanisms, IARC Scientific Publication No. 84. pp. 451-455. Lyon, France: International Agency for Research on Cancer, 1987.

17. Andersen, R. A., and Kasperbauer. M. J. Post-harvest treatment and theaccumulation of nitrite and A"-nitrosonornicotine in hurley tobacco. In: I. K.O'Neill, R. C. von Borstel, C. T. Miller, J. Long, and H. Bartsch (eds.). A'-

Nitroso Compounds: Occurrence, Biological Effects, and Relevance to Human Cancer, IARC Scientific Publication No. 57, pp. 877-883. Lyon. France:

International Agency for Research on Cancer, 1984.18. CalmÃ©is,S.. Ohshima. H.. and Bartsch, H. Nitrosamine formation by deni

trifying and non-denitrifying bacteria: implication of nitrite reducÃaseandnitrate reducÃasein nitrosation catalysis. J. Gen. Microbio!., 134: 221-226,1988.

19. Parsons, L. L., Smith, M. S., Hamilton, J., and MacKown. C. T. Nitratereduction during curing and processing of hurley tobacco. Tobacco Sci., 30:100-103, 1986.

20. Hecht, S. S., Chen, C. B., Hirota, N., Ornaf, R. M., Tso, T. C., and Hoffmann,D. Tobacco-specific nilrosamines: formation from nicotine in vitro and duringtobacco curing and carcinogenicity in Strain A mice. J. Nati. Cancer Inst..60:819-824. 1978.

5900on April 13, 2017. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from


1989;49:5895-5900. Cancer Res Roger A. Andersen, Harold R. Burton, Pierce D. Fleming, et al. ProductsOxidized Pyridine Alkaloid Derivatives in Smokeless Tobacco Effect of Storage Conditions on Nitrosated, Acylated, and

Updated version

http://cancerres.aacrjournals.org/content/49/21/5895

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

[email protected] at

To request permission to re-use all or part of this article, contact the AACR Publications


http://cancerres.aacrjournals.org/content/49/21/5895

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

mailto:[email protected]


effect of storage conditions on nitrosated, acylated, and ... · [cancer research 49, 5895-5900,...

Documents