Regulatory Toxicology and Pharmacology 70 (2014) 138–148
Contents lists available at ScienceDirect
Regulatory Toxicology and Pharmacology journal homepage: www.elsevier.com/locate/yrtph
Insights from analysis for harmful and potentially harmful constituents (HPHCs) in tobacco products Michael J. Oldham ⇑, Darren J. DeSoi, Lonnie T. Rimmer, Karl A. Wagner, Michael J. Morton Altria Client Services, 601 East Jackson Street, Richmond, VA 23261, USA
a r t i c l e
i n f o
Article history: Received 14 April 2014 Available online 25 June 2014 Keywords: Harmful and potentially harmful constituents Cigarettes Smokeless tobacco Assay variability
a b s t r a c t A total of 20 commercial cigarette and 16 commercial smokeless tobacco products were assayed for 96 compounds listed as harmful and potentially harmful constituents (HPHCs) by the US Food and Drug Administration. For each product, a single lot was used for all testing. Both International Organization for Standardization and Health Canada smoking regimens were used for cigarette testing. For those HPHCs detected, measured levels were consistent with levels reported in the literature, however substantial assay variability (measured as average relative standard deviation) was found for most results. Using an abbreviated list of HPHCs, statistically significant differences for most of these HPHCs occurred when results were obtained 4–6 months apart (i.e., temporal variability). The assay variability and temporal variability demonstrate the need for standardized analytical methods with defined repeatability and reproducibility for each HPHC using certified reference standards. Temporal variability also means that simple conventional comparisons, such as two-sample t-tests, are inappropriate for comparing products tested at different points in time from the same laboratory or from different laboratories. Until capable laboratories use standardized assays with established repeatability, reproducibility, and certified reference standards, the resulting HPHC data will be unreliable for product comparisons or other decision making in regulatory science. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction The US Food and Drug Administration (FDA) has established a list of 93 ‘‘harmful and potentially harmful constituents’’ (HPHCs) for tobacco products (FDA, 2012a). Currently, tobacco product
manufacturers are required to report to FDA the levels of an abbreviated HPHC list by brand and sub brand (18 HPHCs for cigarette smoke and 9 HPHCs for smokeless tobacco products) (FDA, 2012b). The purpose of this reporting is to allow FDA to ‘‘publish in a format that is understandable and not misleading to a lay
Abbreviations: 1,2,3,4,6,7,8-HpCDD, 1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin; 1,2,3,4,6,7,8-HpCDF, 1,2,3,4,6,7,8-heptachlorodibenzo-p-furan; 1,2,3,4,7,8,9-HpCDF, 1,2,3,4,7,8,9-heptachlorodibenzo-p-furan; 1,2,3,4,7,8-HxCDD, 1,2,3,4,7,8-hexachlorodibenzo-p-dioxin; 1,2,3,4,7,8-HxCDF, 1,2,3,4,7,8-hexachlorodibenzo-p-furan; 1,2,3,6,7,8HxCDD, 1,2,3,6,7,8-hexachlorodibenzo-p-dioxin; 1,2,3,6,7,8-HxCDF, 1,2,3,6,7,8-hexachlorodibenzo-p-furan; 1,2,3,7,8,9-HxCDD, 1,2,3,7,8,9-hexachlorodibenzo-p-dioxin; 1,2,3,7,8,9-HxCDF, 1,2,3,7,8,9-hexachlorodibenzo-p-furan; 1,2,3,7,8-PeCDD, 1,2,3,7,8-pentachlorodibenzo-p-dioxin; 1,2,3,7,8-PeCDF, 1,2,3,7,8-pentachlorodibenzo-p-furan; 2,3,4,6,7,8-HxCDF, 2,3,4,6,7,8-hexachlorodibenzo-p-furan; 2,3,4,7,8-PeCDF, 2,3,4,7,8-pentachlorodibenzo-p-furan; 2,3,7,8-TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; 2,3,7,8TCDF, 2,3,7,8-tetrachlorodibenzo-p-furan; A-a-C, 2-amino-9H-pyrido[2,3-b]indole; CDC, Centers for Disease Control and Prevention; CORESTA, Cooperation Centre for Scientific Research Relative to Tobacco; CRP2, CORESTA Reference Product 2; FDA, US Food and Drug Administration; GC/MS, gas chromatography–mass spectrometry; Glu-P1, 2-amino-6-methyldipyrido[1,2-a:30 ,20 -d]imidazole; Glu-P-2, 2-aminodipyrido[1,2-a:30 ,20 -d]imidazole; HPHC, harmful and potentially harmful constituent; HPLC, high performance liquid chromatography; ICP-MS, inductively coupled plasma mass spectroscopy; IQ, 2-amino-3-methylimidazo[4,5-f]quinoline; ISO, International Organization for Standardization; LC–MS/MS, liquid chromatography–mass spectroscopy–mass spectroscopy; LOD, limit of detection; LOQ, limit of quantification; MEA-a-C, 2-amino-3methyl-9H-pyrido[2,3-b]indole; MCS, mainstream cigarette smoke; GC/MS/MS, gas chromatograph–mass spectroscopy–mass spectroscopy; NDEA, N-nitrosodiethylamine; NDELA, N-nitrosodiethanolamine; NDMA, N-nitrosodimethylamine; NMEA, N-nitrosomethylethylamine; NMOR, N-nitrosomorpholine; NNK, 4-(methylnitrosamino)-1-(3pyridyl)-1-butanone; NNN, N-nitrosonornicotine; NPIP, N-nitrosopiperidine; NPYR, N-nitrosopyrrolidine; NSAR, N-nitrososarcosine; OCDD, octachlorodibenzo-p-dioxin; OCDF, octachlorodibenzo-p-furan; PAHs, polyaromatic hydrocarbons; PhIP, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine; total TCDD, total tetrachlorodibenzo-pdioxin; total-HpCDD, total heptachlorodibenzo-p-dioxin; total-HpCDF, total heptachlorodibenzo-p-furan; total-HxCDD, total hexachlorodibenzo-p-dioxin; total-HxCDF, total hexachlorodibenzo-p-furan; total-PeCDD, total pentachlorodibenzo-p-dioxin; total-PeCDF, total pentachlorodibenzo-p-furan; total-TCDF, total tetrachlorodibenzo-p-furan; TPM, total particulate matter; Trp-P-1, 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole; Trp-P-2, 1-methyl-3-amino-5H-pyrido[4,3-b]indole. ⇑ Corresponding author. Address: Regulatory Affairs, Altria Client Services, 601 East Jackson Street, Richmond, VA 23261, USA. Fax: +1 804 335 2081. E-mail addresses: [email protected], [email protected] (M.J. Oldham). http://dx.doi.org/10.1016/j.yrtph.2014.06.017 0273-2300/Ó 2014 Elsevier Inc. All rights reserved.
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148
person, and place on public display (in a manner determined by the Secretary) the list. . .’’ (Family Smoking Prevention and Tobacco Control Act, 2009, sec. 904). In addition, FDA has encouraged tobacco product manufacturers to include HPHC data in new product applications, although FDA has not been explicit in how it intends to use HPHC data when evaluating new tobacco products (FDA, 2011a,b,c). For both the purpose of consumer communication and potentially for the purpose of new tobacco product application evaluation, it is critical to understand and consider the limitations of tobacco product constituent analysis resulting from all sources of variability. Scientists and public health researchers have measured levels of chemical constituents to compare tobacco products for decades (Adams et al., 1987; Connolly et al., 2005; Ding et al., 2006; Gendreau and Vitaro, 2005; Hammond and O’Connor, 2008). Several large studies have demonstrated that measurements of tobacco and tobacco smoke constituents are not consistent as a result of assay, inter-laboratory, and temporal variability (Chepiga et al., 2000; Counts et al., 2004, 2005, 2006; Gaworski et al., 2011a; Morton and Laffoon, 2008; Oldham et al., 2012; Roemer et al., 2004). Estimates of assay and intra-laboratory variability can be made by comparing smoke chemistry analytical results for reference cigarettes that are manufactured as a single batch at a single point in time. For example, Gaworski et al. (2011a) included analyses of 52 mainstream cigarette smoke (MCS) constituents from the 1R4F and 2R4F reference cigarettes measured 107 times over a seven-year period at two different laboratories. The study found a general 10–15% relative standard deviation for the 39 of 52 smoke constituents that could be quantified, and it found significant differences between laboratories for some constituent measurements. It also included analyses of a control cigarette manufactured to the same specification 50 times over a seven-year period, which allowed assessment of temporal variability in MCS constituents. The authors reported variability to be greater for the control cigarette than for the reference cigarettes due to year-to-year variation in the tobacco crops used to manufacture the control cigarette. Recently, Purkis et al. (2012) summarized technical challenges and possible limitations for measurement of MCS constituents based on a review of Cooperation Centre for Scientific Research Relative to Tobacco (CORESTA) sponsored collaborative studies. Their review highlighted data variability issues (e.g., product variability; within and between laboratory variability) and the need for standardized analytical methodology. These studies clearly demonstrate that the inherent variability associated with constituent measurement in tobacco and tobacco smoke must be considered when comparing measured constituent levels between samples. The purpose of the current study was to evaluate the average relative standard deviation for all chemical constituents on a draft list of 96 HPHCs from MCS and smokeless tobacco products, which FDA published in 2011 (FDA, 2011d). The study samples were commercial cigarettes and smokeless tobacco products manufactured by Philip Morris USA and U.S. Smokeless Tobacco Company. In addition, samples from the same manufacturing lot were analyzed a second time for the constituents listed on an abbreviated list of HPHCs corresponding to those included in FDA’s draft guidance for constituent reporting (FDA, 2012b). This second analysis allowed assessment of short-term temporal analytical variability. These analyses confirm previous studies demonstrating the inherent variability in tobacco and tobacco smoke constituent measurement and extend these findings to the entire FDA draft HPHC list (FDA, 2011d). 2. Materials and methods Three contract laboratories analyzed tobacco products for all 96 HPHCs (Table 1) on the draft HPHC list (FDA, 2011d). The
139
chlorinated dioxins/furans measured in this study, which were listed as a single entity on the draft HPHC list (FDA, 2011d), are listed in Table 2. All three laboratories were ISO 17025 accredited, and all analytical methods for determination of HPHCs were on the laboratories’ ISO scope of accreditation at the time of testing. The three laboratories used were Arista Laboratories (Richmond, VA; Laboratory 1), Enthalpy Analytical, Inc. (Durham, NC; Laboratory 2), and Labstat International ULC (Ontario, Canada; Laboratory 3). Laboratories 1 and 2 were accredited by the American Association for Laboratory Accreditation, whereas Laboratory 3 was accredited by the Standards Council of Canada. Whether the HPHC was analyzed in cigarettes, smokeless tobacco products, or both, was based upon recommendations from the Tobacco Product Constituents Subcommittee of the Tobacco Products Scientific Advisory Committee (TPSAC, 2010). 2.1. Tobacco products Twenty commercial cigarette and 16 commercial smokeless tobacco products (Table 3) were assayed for all applicable HPHCs on the 96-item draft list (FDA, 2011d). The 3R4F reference cigarette (University of Kentucky) was used as the reference product for MCS constituents, and the CORESTA Reference Product 2 (CRP2; North Carolina State University) was used as the reference for smokeless tobacco constituents. In order to minimize analytical variability, all products were submitted to a single laboratory for a given method. Based upon previous experience in measuring MCS constituents (Counts et al., 2004, 2005; Morton and Laffoon, 2008; Gaworski et al., 2011b), three replicate analyses (each analysis may require multiple cigarettes) were chosen as sufficient to provide an estimate of each HPHC. The only exception was carbonyls in smoke, for which five replicates were used. The FDA draft guidance for reporting HPHCs (FDA, 2012b), which was published after this study started, recommended that seven replicates be used for each HPHC except smoke nicotine and carbon monoxide, for which twenty replicates were recommended. Therefore, to obtain the recommended number of replicates, for each tobacco product in this study a second set of samples was analyzed for the abbreviated list of HPHCs (FDA, 2012b) at the same three laboratories. For each tobacco product, a single manufacturing lot was used for all testing. 2.2. Smoking regimens MCS was generated under both ISO (2000b) and Health Canada (1999a) machine smoking regimens using commercially available linear and rotary smoking machines. Cigarettes were conditioned before smoking in accordance with ISO 3402 (ISO, 1999). MCS was generated and collected in basic accordance with ISO 3308 (ISO, 2000b) and ISO 4387 (ISO, 2000c). Deviations from the ISO standards were made when necessary in order to accommodate Health Canada smoking and in order to incorporate smoke traps for volatile or gas phase HPHCs. 2.3. HPHCs analytical assays for mainstream cigarette smoke Official methods (Health Canada; Centers for Disease Control and Prevention [CDC]) are cited, and other internally validated methods are briefly described for general understanding. 2.3.1. Nicotine and carbon monoxide in mainstream cigarette smoke Nicotine and carbon monoxide were determined using Health Canada Official Method T-115 (Health Canada, 1999a). Since this method also measures water and calculates tar, these data were also collected.
140
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148
Table 1 Draft list of harmful and potentially harmful constituents (FDA, 2011d) measured in mainstream cigarette smoke and smokeless tobacco products grouped by analytical method with available standardized analytical methods. Assay method constituent
MCS
Aromatic amines 1-Aminonaphthalene 2-Aminonaphthalene 4-Aminobiphenyl 2,6-Dimethylaniline o-Anisidine o-Toluidine
x x x x x x
Biological Aflatoxin B1 x x x x x x x
Heterocyclic aromatic amines A-a-C (2-amino-9H-pyrido ([2,3-b]indole) Glu-P-1 (2-amino-6-methyldipyridol[1,2-a:30 ,20 -d]imidazole) Glu-P-2 (2-aminodipyridol[1,2-a:30 ,20 -d]imidazole IQ (2-amino-d-methylimidazo[4,5-f]quinoline) MeA-a-C (2-amino-3-methyl)-9H-pyrido[2,3-b]indole) PhIP (2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine Trp-P-1 (3-amino1,4-dimethyl-5H-pyrido[4,3-b]indole) Trp-P-2 (1-methyl-3-amino-5H-pyrido[4,3-b]indole)
x xb,c xb,c xb,c x xb,c xb,c xb,c
N-Nitrosonornicotine (NNN) N-Nitrosopiperidine (NPIP) N-Nitrosopyrrolidine (NPYR) N-Nitrososarcosine (NSAR) Other Ammonia (listed as ammonia salts for STP) Anabasine Benzo[b]furan Caffeic acid Carbon monoxide Chlorinated dioxins/furansa Coumarin Dibenz[a,h]acridine Dibenz[a,j]acridine Dibenzo[c,g]carbazole Dibenzo[a,e]pyrene Dibenzo[a,h]pyrene Dibenzo[a,i]pyrene Dibenzo[a,l]pyrene Ethyl carbamate (urethane) Hydrazine Hydrogen cyanide Mercury 5-Methylchrysene Nicotine
Standardized analytical method
xf
Carbonyls Acetaldehyde Acetone Acrolein Crotonaldehyde Formaldehyde Methyl ethyl ketone Propionaldehyde
Metals Arsenic Beryllium Cadmium Chromium Cobalt Lead Nickel Selenium Nitrosamines N-Nitrosodiethylamine (NDEA) N-Nitrosodiethanolamine (NDELA) N-Nitrosodimethylamine (NDMA) N-Nitrosomethylethylamine (NMEA) N-Nitrosomorpholine (NMOR) 4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK)
STP
x
x x
x xb,c x xc xc x x x xd,e x x xe
x x x
xg xg xg x
x x x x
x xg x x
x c
x
CORESTA CORESTA CORESTA CORESTA
N° N° N° N°
75 72 75 72
(MCS) (STP) (MCS) (STP)
xf x xg x x
x xd,e x x
ISO 8454 (MCS) x
xb,c xb,c xb,c xe xc,d x xb,c xb xb,e x
xf
x x x
x
ISO 10315 (MCS) ISO 3400 (MCS) ISO 15517 (STP)
141
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148 Table 1 (continued) Assay method constituent
MCS
Nornicotine Phenolic compounds Catechol m-Cresol o-Cresol p-Cresol Phenol
x x x x x
Polycyclic aromatic hydrocarbons Benz[a]anthracene Benz[j]aceanthrylene Benzo[a]pyrene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[c]phenanthrene Chrysene Cyclopenta[c,d]pyrene Dibenz[a,h]anthracene Indeno[1,2,3-cd]pyrene Naphthalene
x x x x x x x x x x x
Radioactive Polonium-210 Uranium-235 Uranium-238
STP
Standardized analytical method
x
x
Semivolatiles 2-Nitropropane Acetamide Acrylamide Ethylbenzene Nitrobenzene Quinoline Styrene
xb,c x x x xb,c x x
Volatiles 1,3-Butadiene Acrylonitrile Benzene Ethylene oxide Furan Isoprene Nitromethane Propylene oxide Toluene Vinyl acetate Vinyl chloride
x x x x x x xb,c xb,c x xb,c xb,c
x x x x
ISO 22634 (MCS)
x xg x x x xf xf
CORESTA N° 70 (MCS) CORESTA N° 70 (MCS) CORESTA N° 70 (MCS)
CORESTA N° 70 (MCS)
CORESTA N° 70 (MCS)
CORESTA N° 70 (see CORESTA, 2013a); CORESTA N° 72 (see CORESTA, 2013b); CORESTA N° 75 (see CORESTA, 2012); ISO 3400 (see ISO, 1977); ISO 8454 (see ISO, 2007); ISO 10315 (see ISO, 2000a); ISO 15517 (see ISO, 2003); ISO 22634 (see ISO, 2008); MCS, mainstream cigarette smoke; STP, smokeless tobacco product. a See Table 2 for list of chlorinated dioxins/furans measured in mainstream cigarette smoke. b All values were less than limit of detection using ISO (2000b) smoking regimen. c All values were less than limit of detection using Health Canada (1999a) smoking regimen. d All values were less than limit of quantification using ISO (2000b) smoking regimen. e All values were less than limit of quantification using Health Canada (1999a) smoking regimen. f All values were less than limit of detection. g All values were less than limit of quantification.
2.3.2. Carbonyl in mainstream cigarette smoke Acetaldehyde, acetone, acrolein, crotonaldehyde, formaldehyde, methyl ethyl ketone, and propionaldehyde were determined using Health Canada Official Method T-104 (Health Canada, 1999b). 2.3.3. Nitrosamines in mainstream cigarette smoke N-Nitrosodiethylamine (NDEA), N-nitrosodiethanolamine (NDELA), N-nitrosodimethylamine (NDMA), N-nitrosomethylethylamine (NMEA), N-nitrosomorpholine (NMOR), 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), N-nitrosonornicotine (NNN), N-nitrosopiperidine (NPIP), N-nitrosopyrrolidine (NPYR), and N-nitrososarcosine (NSAR) were determined in MCS by collecting total particulate matter (TPM) on a 44-mm Cambridge filter pad. The pad was subsequently extracted in 10 mL of aqueous extraction solution containing internal standards. The solution was solvent-exchanged to methylene chloride and analyzed by gas chromatograph–mass spectrometry–mass spectrometry (GC/
MS/MS) for tobacco specific nitrosamines and volatile nitrosamines. NDELA was derivatized with N-methyl-n-(tert-butyldimethylsilyl)trifluoroacetamide/tert-butyldimethylchlorosilane prior to analysis and was analyzed in a separate analytical run from the other nitrosamines. 2.3.4. Polyaromatic hydrocarbons and heterocyclic compounds in mainstream cigarette smoke Benz[a]anthracene, benz[j]aceanthrylene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[a]pyrene, benzo[c]phenanthrene, chrysene, cyclopenta[c,d]pyrene, dibenz[a,h]acridine, dibenz[a,j] acridine, dibenz[a,h]anthracene, dibenzo[c,g]carbazole, dibenzo [a,e]pyrene, dibenzo[a,h]pyrene, dibenzo[a,i]pyrene, dibenzo[a,l] pyrene, indeno[1,2,3-cd]pyrene, 5-methylchrysene, and naphthalene were determined by passing MCS through a 92-mm filter pad. After addition of internal standard, polyaromatic hydrocarbons (PAHs) and heterocyclic compounds were extracted from
142
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148
Table 2 Chlorinated dioxins/furans measured in mainstream cigarette smoke.
Table 3 List of cigarettes and smokeless tobacco products tested.
Assay method constituent
MCS
Product name
1,2,3,4,6,7,8-HpCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF 1,2,3,4,7,8-HxCDD 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDD 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDD 1,2,3,7,8,9-HxCDF 1,2,3,7,8-PeCDD 1,2,3,7,8-PeCDF 2,3,4,6,7,8-HxCDF 2,3,4,7,8-PeCDF 2,3,7,8-TCDD 2,3,7,8-TCDF OCDD OCDF Total TCDD Total-HpCDD Total-HpCDF Total-HxCDD Total-HxCDF Total-PeCDD Total-PeCDF Total-TCDF
x xb,c xa,b xa,b xa,b xa,b xa,b xa,b xa,b xa,b xa,b xa,b xc,d xa,b x x xa,b xa,b x xb,c xc,d xa,b xa,b xc,d x
Cigarettes Marlboro 1000 s Box Marlboro 720 s Box Marlboro 720 s Gold Pack Box Marlboro Blend No. 27 Box Marlboro Box Marlboro Gold Pack 1000 s Box Marlboro Gold Pack Box Marlboro Gold Pack Soft Pack Marlboro Menthol Box Marlboro Menthol Gold Pack Box Marlboro Red Label Box Marlboro Silver Pack 1000 s Box Marlboro Silver Pack Box Marlboro Soft Pack Marlboro Special Blend (Gold Pack) 1000 s Box Marlboro Special Blend (Gold Pack) Box Marlboro Special Blend (Red Pack) 1000 s Box Marlboro Special Blend (Red Pack) Box Marlboro Virginia Blend Box Parliament (White Pack) Box
1,2,3,4,6,7,8-HpCDD, 1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin; 1,2,3,4,6,7,8HpCDF, 1,2,3,4,6,7,8-heptachlorodibenzo-p-furan; 1,2,3,4,7,8,9-HpCDF, 1,2,3,4,7,8, 9-heptachlorodibenzo-p-furan; 1,2,3,4,7,8-HxCDD, 1,2,3,4,7,8-hexachlorodibenzop-dioxin; 1,2,3,4,7,8-HxCDF, 1,2,3,4,7,8-hexachlorodibenzo-p-furan; 1,2,3,6,7,8HxCDD, 1,2,3,6,7,8-hexachlorodibenzo-p-dioxin; 1,2,3,6,7,8-HxCDF, 1,2,3,6,7, 8-hexachlorodibenzo-p-furan; 1,2,3,7,8,9-HxCDD, 1,2,3,7,8,9-hexachlorodibenzo-pdioxin; 1,2,3,7,8,9-HxCDF, 1,2,3,7,8,9-hexachlorodibenzo-p-furan; MCS, mainstream cigarette smoke; 1,2,3,7,8-PeCDD, 1,2,3,7,8-pentachlorodibenzo-p-dioxin; 1,2,3,7,8-PeCDF, 1,2,3,7,8-pentachlorodibenzo-p-furan; 2,3,4,6,7,8-HxCDF, 2,3,4,6,7,8-hexachlorodibenzo-p-furan; 2,3,4,7,8-PeCDF, 2,3,4,7,8-pentachlorodibenzo-p-furan; 2,3,7,8-TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; 2,3,7,8-TCDF, 2,3,7,8-tetrachlorodibenzo-p-furan; OCDD, octachlorodibenzo-p-dioxin; OCDF, octachlorodibenzo-p-furan; total TCDD, total tetrachlorodibenzo-p-dioxin; total-HpCDD, total heptachlorodibenzo-p-dioxin; total-HpCDF, total heptachlorodibenzo-p-furan; total-HxCDD, total hexachlorodibenzo-p-dioxin; total-HxCDF, total hexachlorodibenzo-p-furan; total-PeCDD, total pentachlorodibenzo-p-dioxin; total-PeCDF, total pentachlorodibenzo-p-furan; totalTCDF, total tetrachlorodibenzo-p-furan. a All values were less than limit of detection using ISO (2000b) smoking regimen. b All values were less than limit of detection Health Canada (1999a) smoking regimen. c Mean value was less than limit of quantification or one or more values were less than limit of detection using ISO (2000b) smoking regimen. d Mean value was less than limit of quantification or one or more values were less than limit of detection using Health Canada (1999a) smoking regimen.
the pad with methanol. The methanol extract was filtered, and a portion of the filtered extract was diluted with water and passed through a C18 cartridge. Analytes were eluted from the cartridge with cyclohexane and analyzed by gas chromatography–mass spectrometry (GC/MS). 2.3.5. Benzo[b]furan in mainstream cigarette smoke MCS was passed through a 92-mm Cambridge filter pad followed by two cryogenic impingers containing methanol. Following collection, the pad was placed into an Erlenmeyer flask and internal standard was added. The pad extract was combined with impinger solutions and quantified using GC/MS. 2.3.6. Aromatic amines in mainstream cigarette smoke 1-Aminonaphthalene, 2-aminonaphthalene, 4-aminobiphenyl, 2,6-dimethylaniline, o-anisidine, and o-toluidine were determined by collecting MCS on a 92-mm Cambridge filter pad. The pad was quartered and placed in an Erlenmeyer flask with 100 mL of 5% hydrochloric acid solution and shaken for 30 min. Extract was
Smokeless tobacco products Copenhagen Long Cut Copenhagen Long Cut Straight Copenhagen Long Cut Wintergreen Copenhagen Pouches Copenhagen Snuff Fine Cut Marlboro Snus Smooth Mint Red Seal Fine Cut Natural Red Seal Fine Cut Wintergreen Red Seal Long Cut Wintergreen Skoal Long Cut Mint Skoal Long Cut Straight Skoal Long Cut Wintergreen Skoal Original Fine Cut Wintergreen Skoal Pouches Mint Skoal Pouches Wintergreen Skoal X-tra Long Cut Wintergreen Blend
filtered and spiked with internal standards. The filtrate was washed with dichloromethane, made basic with sodium hydroxide solution and extracted with hexane. The hexane extract was dried with sodium sulfate, derivatized with pentafluoropropionic acid anhydride and trimethylamine, and then passed through a FlorisilÒ column (U.S. Silica Co., Berkeley Springs, WV). The aromatic amines in the hexane extract were then quantified by GC/MS. 2.3.7. Ammonia in mainstream cigarette smoke MCS was passed through a 44-mm Cambridge filter pad followed by one impinger containing 40 mL of 40 mM aqueous sulfuric acid. The filter pad was added to the impinger solution and mixed. The solution was filtered and analyzed by ion chromatography with conductivity detection. 2.3.8. Volatile organic compounds in mainstream cigarette smoke 1,3-Butadiene, acrylonitrile, benzene, ethylene oxide, furan, isoprene, nitromethane, propylene oxide, toluene, vinyl acetate, and vinyl chloride were determined in MCS by passing the smoke through a 44-mm Cambridge pad followed by one impinger containing 20 mL of methanol and internal standards. The impinger was cooled by a dry ice/isopropanol bath. While still cold, an aliquot was quickly transferred to a zero-headspace vial and then analyzed by GC/MS. 2.3.9. Semi-volatile organic compounds in mainstream cigarette smoke 2-Nitropropane, acetamide, acrylamide, ethylbenzene, nitrobenzene, quinoline, and styrene were determined in MCS by passing the
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148
smoke through a 44-mm Cambridge pad followed by one impinger containing 20 mL of methanol and internal standards. The impinger was cooled by a dry ice/isopropanol bath. The Cambridge pad was added to the impinger solution and shaken. An aliquot of the solution was analyzed by GC/MS. Since this method also measures pyridine, these data were also collected. 2.3.10. Chlorinated dioxins and furans in mainstream cigarette smoke The following chlorinated dioxins and furans were measured: 1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin (1,2,3,4,6,7,8-HpCDD); 1, 2,3,4,6,7,8-heptachlorodibenzo-p-furan (1,2,3,4,6,7,8-HpCDF); 1,2,3, 4,7,8,9-heptachlorodibenzo-p-furan (1,2,3,4,7,8,9-HpCDF); 1,2,3,4,7, 8-hexachlorodibenzo-p-dioxin (1,2,3,4,7,8-HxCDD); 1,2,3,4,7,8hexachlorodibenzo-p-furan (1,2,3,4,7,8-HxCDF); 1,2,3,6,7,8-hexachl orodibenzo-p-dioxin (1,2,3,6,7,8-HxCDD); 1,2,3,6,7,8-hexachlorodi benzo-p-furan (1,2,3,6,7,8-HxCDF); 1,2,3,7,8,9-hexachlorodibenzop-dioxin (1,2,3,7,8,9-HxCDD); 1,2,3,7,8,9-hexachlorodibenzo-p-furan (1,2,3,7,8,9-HxCDF); 1,2,3,7,8-pentachlorodibenzo-p-dioxin (1,2,3, 7,8-PeCDD); 1,2,3,7,8-pentachlorodibenzo-p-furan (1,2,3,7,8-PeCDF); 2,3,4,6,7,8-hexachlorodibenzo-p-furan (2,3,4,6,7,8-HxCDF); 2,3,4,7,8-pentachlorodibenzo-p-furan (2,3,4,7,8-PeCDF); 2,3,7, 8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD); 2,3,7,8-tetrachlorodibenzo-p-furan (2,3,7,8-TCDF); octachlorodibenzo-p-dioxin (OCDD); octachlorodibenzo-p-furan (OCDF); total heptachlorodi benzo-p-dioxin (total-HpCDD); total heptachlorodibenzo-p-furan (total-HpCDF); total hexachlorodibenzo-p-dioxin (total-HxCDD); total hexachlorodibenzo-p-furan (total-HxCDF); total pentachlorodibenzo-p-dioxin (total-PeCDD); total pentachlorodibenzo-pfuran (total-PeCDF); total tetrachlorodibenzo-p-dioxin (total TCDD); and total tetrachlorodibenzo-p-furan (total-TCDF). TPM was collected on Cambridge filter pads, which were spiked with labeled internal standards and Soxhlet-extracted for 16 h. The extract was subjected to an acid/base clean-up procedure followed by clean-up on micro columns of silica gel and alumina. The extract was spiked with 0.1 ng 13C-1,2,3,4-TCDD and 13C-1,2,3,4,7,8, 9-HCDD (to determine extraction efficiencies achieved for the 13Clabeled internal standards) and then concentrated to 20 lL. Analysis was conducted using GC-high resolution mass spectrometry. 2.3.11. Caffeic acid in mainstream cigarette smoke MCS was passed through a 44-mm Cambridge filter pad. The pad was shaken for 30 min with 40 mL of Type I water in an Erlenmeyer flask. The extract was then filtered with a 0.45 lm syringe filter, and aliquots were analyzed for caffeic acid using high performance liquid chromatography (HPLC) with ultraviolet light detection. 2.3.12. Hydrogen cyanide in mainstream cigarette smoke Hydrogen cyanide was determined using Health Canada Official Method T-107 (Health Canada, 1999c). 2.3.13. Hydrazine in mainstream cigarette smoke MCS was passed through a dinitrophenylhydrazine-treated 92-mm Cambridge pad followed by two impingers. Each impinger contained 40 mL of a solution consisting of methanol and aqueous citrate–phosphate buffer with added 2-nitrobenzaldehyde. Immediately after smoking, the filter pad was extracted with the impinger solutions and the extract was incubated for 30 min at 35 °C. An aliquot of extract was centrifuged and the resultant hydrazone (i.e., dinitrophenyl-hydrazone) was quantified by HPLC/MS/MS using an internal standard. 2.3.14. Heterocyclic amines in mainstream cigarette smoke 2-Amino-9H-pyrido[2,3-b]indole (A-a-C), 2-amino-6-methyldipyrido[1,2-a:30 ,20 -d]imidazole (Glu-P-1), 2-aminodipyrido[1, 2-a:30 ,20 -d]imidazole (Glu-P-2), 2-amino-3-methylimidazo[4,5-f]
143
quinoline (IQ), 2-amino-3-methyl-9H-pyrido[2,3-b]indole (MeAa-C), 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1), and 1-methyl-3-amino-5H-pyrido[4,3-b]indole (Trp-P-2) were determined by collecting MCS on a Cambridge filter pad. Internal standard was added to the pad, which was then extracted with formic acid. Extracts were filtered before analysis by HPLC/MS/ MS using positive electrospray ionization. 2.3.15. Metals in mainstream cigarette smoke Arsenic, beryllium, cadmium, chromium, cobalt, lead, nickel, and selenium were determined by collecting MCS condensate using an electrostatic precipitation unit equipped with a quartz collection tube. Collected material was rinsed from the quartz tube using methanol. The methanol was evaporated, and the resulting sample was digested using nitric acid, hydrogen peroxide, and heat. The digested samples were subsequently analyzed by inductively coupled plasma mass spectroscopy (ICP/MS) equipped with a dynamic reaction cell. 2.3.16. Phenolic compounds in mainstream cigarette smoke Catechol, m-cresol, o-cresol, p-cresol, and phenol were determined in MCS by collecting TPM on a 44-mm Cambridge filter pad. The pad was then extracted with 40 mL of 1% acetic acid. An aliquot of the TPM extract was filtered using a syringe filter and then diluted. The resulting extract was analyzed by HPLCfluorescence. 2.3.17. Polonium-210 in mainstream cigarette smoke Polonium-210 (210Po) was determined in MCS by collecting TPM on a 44-mm Cambridge filter pad. The pad was spiked with polonium-209 (209Po). The entire pad and sample was digested with concentrated hydrofluoric acid, and then heated to dryness. Aqueous dilute hydrochloric acid was then added to the residue. The sample was then deposited on silver or nickel foil in the presence of dilute hydrochloric acid. The resulting sample was analyzed by alpha spectrometry (209Po a-particle energy = 4.9 MeV and 210Po a-particle energy = 5.3 MeV). The calculated concentration of 210Po was based upon the determination of 209Po. 2.3.18. Ethyl carbamate in mainstream cigarette smoke Ethyl carbamate (urethane) was determined by collecting MCS on a Cambridge filter pad. The filter was extracted in 1:1 water/ methanol and then filtered. The sample was then derivatized with 9-xanthydrol in acid. The resulting samples were analyzed using HPLC-fluorescence. 2.4. Harmful and potentially harmful constituents analytical assays for smokeless tobacco Smokeless products were generally analyzed as subsamples from a composite sample that was created from a single lot of product. 2.4.1. Oven volatiles in smokeless tobacco Percent oven volatiles were determined in smokeless tobacco using Association of Analytical Communities International Official Method 966.02 (AOAC, 1990). 2.4.2. Nicotine and minor alkaloids in smokeless tobacco Nicotine (total), nornicotine, and anabasine were determined in smokeless tobacco using Health Canada Official Method T-301 (Health Canada, 1999d).
144
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148
2.4.3. pH in smokeless tobacco pH was determined in smokeless tobacco using the CDC method (CDC, 1999, 2009). 2.4.4. Metals in smokeless tobacco Arsenic, beryllium, cadmium, chromium, lead, nickel, and selenium were determined in smokeless tobacco by digesting tobacco samples using nitric acid in a closed-vessel microwave digestion. Digested samples were subsequently analyzed by ICP/MS. Since this method also measures cobalt, these data were also collected. 2.4.5. Mercury in smokeless tobacco Tobacco samples were digested in a microwave using nitric acid and potassium permanganate. Hydroxylamine hydrochloride was added after digestion, and the resulting solutions were analyzed for mercury using cold vapor atomic absorption spectrometry (flow injection). 2.4.6. Carbonyls in smokeless tobacco Acetaldehyde, crotonaldehyde, and formaldehyde were determined in smokeless tobacco by extracting tobacco with an aqueous/acetonitrile solution containing acidified dinitrophenylhydrazine. An aliquot of the extract was subsequently treated with base to stabilize the hydrazones. The samples were filtered and then analyzed using reversed-phase HPLC with UV detection. 2.4.7. Polyaromatic hydrocarbons in smokeless tobacco Benz[a]anthracene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[a]pyrene, chrysene, dibenz[a,h]anthracene, indeno [1,2,3-cd]pyrene, and naphthalene were determined in smokeless tobacco by extracting tobacco with methanol containing internal standards. The sample extracts were then diluted with water and passed through a solid-phase extraction cartridge. The analytes of interest were determined in the resulting solution with GC/MS. 2.4.8. Nitrosamines in smokeless tobacco NDEA, NDELA, NDMA, NMOR, NNK, NNN, NPIP, NPYR, and NSAR were determined in smokeless tobacco by extracting the tobacco in 10 mL of aqueous extraction solution containing internal standards. The solution was solvent-exchanged to methylene chloride and analyzed by GC/MS/MS for tobacco specific nitrosamines and volatile nitrosamines. NDELA was derivatized with N-methyl-n(tert-butyldimethylsilyl) trifluoroacetamide/tert-butyldimethylchlorosilane prior to analysis and was analyzed in a separate run. 2.4.9. Aflatoxin-B1 in smokeless tobacco Smokeless tobacco was extracted using methanol and water (4:1). Samples were filtered and diluted with phosphate buffered saline solution. Each sample was introduced to an immunoaffinity column, and the column was rinsed with water before elution of the aflatoxin-B1 with methanol. The eluate was diluted 1:1 with water and injected onto an HPLC-fluorescence system using postcolumn derivatization. 2.4.10. Ethyl carbamate in smokeless tobacco Ethyl carbamate (urethane) was determined in smokeless tobacco by extracting the tobacco samples in 1-propanol. Extracts were filtered and derivatized with 9-xanthydrol in acid. The resulting samples were analyzed using HPLC-fluorescence. 2.4.11. Coumarin in smokeless tobacco Coumarin was determined in smokeless tobacco by spiking each tobacco sample with internal standard and then extracting with 1:1 ethanol in water. Samples were filtered and then diluted with 1:1 methanol in water. Quantification was accomplished with
LC/MS/MS using positive electrospray ionization and multiple reactions monitoring mode. Three mass transition pairs (147.1/ 103.1; 147.1/91.1; 147.1/77.1) were monitored for identification, confirmation, and quantification. 2.4.12. Polonium-210 in smokeless tobacco 210 Po was determined in smokeless tobacco by spiking each sample with a traceable standard containing a known amount of 209 Po as a tracer. All sample matrices were digested with concentrated nitric acid until all organic material was dissolved, and then heated to dryness. Aqueous dilute hydrochloric acid was then added to the residue. The 209Po/210Po was then deposited on silver or nickel foil in the presence of dilute hydrochloric acid and alpha spectrometry performed on the foil (209Po a-particle energy = 4.9 MeV and 210Po a-particle energy = 5.3 MeV). Results for 210Po were calculated according to the amount of 209Po determined. 2.4.13. Uranium-238 in smokeless tobacco Uranium-238 (238U) was determined in smokeless tobacco using neutron activation. Samples were irradiated in a nuclear reactor (flux of 8 1012 neutrons/cm2/s) for up to 20 min. The samples were rotated during irradiation so that there was no horizontal flux variation (vertical flux variation was monitored with the individual flux monitors). Samples were allowed to decay for 6 d before counting using a high-resolution gamma-ray spectrometer (germanium detector connected to multichannel analyzer). Counting time varied between 20 and 30 min per sample. 238U was quantitatively identified by comparing the peak positions and areas results to library standards. 2.4.14. Uranium-235 in smokeless tobacco Uranium-235 (235U) was determined in smokeless tobacco using delayed neutron counting analysis. Samples were irradiated thermally in a nuclear reactor (flux of 8 1012 neutrons/cm2/s), and the delayed neutrons that were produced were counted to determine the concentration of 235U by comparison with standards. 2.5. Statistical analysis Analytes on the abbreviated HPHC list (FDA, 2012b) were tested at two different time points. The two different time points were compared using a two-way analysis of variance including both time and product as factors in the analysis. A statistically significant time effect is indicative of temporal variation in the analytical method.
3. Results and discussion Because the FDA draft list (FDA, 2011d) included several chemicals that were listed as a single entity (m-cresol, o-cresol and p-cresol, which were listed only as cresol), and it listed classes of chemicals (chlorinated dioxins and furans) as a single HPHC, the total number of HPHCs reported in this study was 113 for MCS and 37 for smokeless tobacco products. Three additional compounds (pyridine, tar, and water) are also reported for MCS because of the analytical methodology used. In addition to the HPHCs measured in smokeless tobacco products, cobalt, moisture, and pH were also measured in this study. As noted in Table 1, there are a limited number of standardized methods that only apply to a few HPHCs. None of the standardized methods noted in Table 1 were used by the testing laboratories.
145
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148 Table 4 Replicate-to-replicate variability of HPHCs using ISO (2000b) and Health Canada (1999a) smoking regimens. Constituent
Average %RSDa ISO
1,2,3,4,6,7,8-HpCDD 1,3-Butadiene 1-Aminonaphthalene 2,3,7,8-TCDF 2,6-Dimethylaniline 2-Aminonaphthalene 4-Aminobiphenyl 5-Methylchrysene A-a-C Acetaldehyde Acetamide Acetone Acrolein Acrylamide Acrylonitrile Ammonia Arsenic Benz[a]anthracene Benz[j]aceanthrylene Benzene Benzo[a]pyrene Benzo[b]fluoranthene Benzo[b]furan Benzo[c]phenanthrene Benzo[k]fluoranthene Cadmium Carbon monoxide Catechol Chrysene Crotonaldehyde Cyclopenta[c,d]pyrene Dibenz[a,h]anthracene Dibenzo[a,i]pyrene Ethyl carbamate (urethane) Ethylbenzene Ethylene oxide Formaldehyde Furan Hydrogen cyanide Indeno[1,2,3-cd]pyrene Isoprene Lead MeA-a-C Methyl ethyl ketone N-Nitrosodimethylamine (NDMA) N-Nitrosopiperidine (NPIP) N-Nitrosopyrrolidine (NPYR) NNK NNN Naphthalene Nicotine OCDD Phenol Polonium-210 Propionaldehyde Pyridine Quinoline Selenium Styrene Tar Toluene Total-HpCDD Total-HxCDD Total-TCDF Water m-Cresol o-Anisidine o-Cresol o-Toluidine p-Cresol
Health Canada
%
Number of products’ data used in %RSD
%
Number of products’ data used in %RSD
16.4 8.1 5.7 0.0 6.2 5.4 6.8 12.2 10.7 5.5 10.4 5.3 6.6 10.0 6.0 6.1 5.5 5.4 8.4 6.4 6.0 8.7 12.2 8.9 10.9 4.9 6.1 5.4 5.2 10.3 8.3 14.4 15.9
Contents lists available at ScienceDirect
Regulatory Toxicology and Pharmacology journal homepage: www.elsevier.com/locate/yrtph
Insights from analysis for harmful and potentially harmful constituents (HPHCs) in tobacco products Michael J. Oldham ⇑, Darren J. DeSoi, Lonnie T. Rimmer, Karl A. Wagner, Michael J. Morton Altria Client Services, 601 East Jackson Street, Richmond, VA 23261, USA
a r t i c l e
i n f o
Article history: Received 14 April 2014 Available online 25 June 2014 Keywords: Harmful and potentially harmful constituents Cigarettes Smokeless tobacco Assay variability
a b s t r a c t A total of 20 commercial cigarette and 16 commercial smokeless tobacco products were assayed for 96 compounds listed as harmful and potentially harmful constituents (HPHCs) by the US Food and Drug Administration. For each product, a single lot was used for all testing. Both International Organization for Standardization and Health Canada smoking regimens were used for cigarette testing. For those HPHCs detected, measured levels were consistent with levels reported in the literature, however substantial assay variability (measured as average relative standard deviation) was found for most results. Using an abbreviated list of HPHCs, statistically significant differences for most of these HPHCs occurred when results were obtained 4–6 months apart (i.e., temporal variability). The assay variability and temporal variability demonstrate the need for standardized analytical methods with defined repeatability and reproducibility for each HPHC using certified reference standards. Temporal variability also means that simple conventional comparisons, such as two-sample t-tests, are inappropriate for comparing products tested at different points in time from the same laboratory or from different laboratories. Until capable laboratories use standardized assays with established repeatability, reproducibility, and certified reference standards, the resulting HPHC data will be unreliable for product comparisons or other decision making in regulatory science. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction The US Food and Drug Administration (FDA) has established a list of 93 ‘‘harmful and potentially harmful constituents’’ (HPHCs) for tobacco products (FDA, 2012a). Currently, tobacco product
manufacturers are required to report to FDA the levels of an abbreviated HPHC list by brand and sub brand (18 HPHCs for cigarette smoke and 9 HPHCs for smokeless tobacco products) (FDA, 2012b). The purpose of this reporting is to allow FDA to ‘‘publish in a format that is understandable and not misleading to a lay
Abbreviations: 1,2,3,4,6,7,8-HpCDD, 1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin; 1,2,3,4,6,7,8-HpCDF, 1,2,3,4,6,7,8-heptachlorodibenzo-p-furan; 1,2,3,4,7,8,9-HpCDF, 1,2,3,4,7,8,9-heptachlorodibenzo-p-furan; 1,2,3,4,7,8-HxCDD, 1,2,3,4,7,8-hexachlorodibenzo-p-dioxin; 1,2,3,4,7,8-HxCDF, 1,2,3,4,7,8-hexachlorodibenzo-p-furan; 1,2,3,6,7,8HxCDD, 1,2,3,6,7,8-hexachlorodibenzo-p-dioxin; 1,2,3,6,7,8-HxCDF, 1,2,3,6,7,8-hexachlorodibenzo-p-furan; 1,2,3,7,8,9-HxCDD, 1,2,3,7,8,9-hexachlorodibenzo-p-dioxin; 1,2,3,7,8,9-HxCDF, 1,2,3,7,8,9-hexachlorodibenzo-p-furan; 1,2,3,7,8-PeCDD, 1,2,3,7,8-pentachlorodibenzo-p-dioxin; 1,2,3,7,8-PeCDF, 1,2,3,7,8-pentachlorodibenzo-p-furan; 2,3,4,6,7,8-HxCDF, 2,3,4,6,7,8-hexachlorodibenzo-p-furan; 2,3,4,7,8-PeCDF, 2,3,4,7,8-pentachlorodibenzo-p-furan; 2,3,7,8-TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; 2,3,7,8TCDF, 2,3,7,8-tetrachlorodibenzo-p-furan; A-a-C, 2-amino-9H-pyrido[2,3-b]indole; CDC, Centers for Disease Control and Prevention; CORESTA, Cooperation Centre for Scientific Research Relative to Tobacco; CRP2, CORESTA Reference Product 2; FDA, US Food and Drug Administration; GC/MS, gas chromatography–mass spectrometry; Glu-P1, 2-amino-6-methyldipyrido[1,2-a:30 ,20 -d]imidazole; Glu-P-2, 2-aminodipyrido[1,2-a:30 ,20 -d]imidazole; HPHC, harmful and potentially harmful constituent; HPLC, high performance liquid chromatography; ICP-MS, inductively coupled plasma mass spectroscopy; IQ, 2-amino-3-methylimidazo[4,5-f]quinoline; ISO, International Organization for Standardization; LC–MS/MS, liquid chromatography–mass spectroscopy–mass spectroscopy; LOD, limit of detection; LOQ, limit of quantification; MEA-a-C, 2-amino-3methyl-9H-pyrido[2,3-b]indole; MCS, mainstream cigarette smoke; GC/MS/MS, gas chromatograph–mass spectroscopy–mass spectroscopy; NDEA, N-nitrosodiethylamine; NDELA, N-nitrosodiethanolamine; NDMA, N-nitrosodimethylamine; NMEA, N-nitrosomethylethylamine; NMOR, N-nitrosomorpholine; NNK, 4-(methylnitrosamino)-1-(3pyridyl)-1-butanone; NNN, N-nitrosonornicotine; NPIP, N-nitrosopiperidine; NPYR, N-nitrosopyrrolidine; NSAR, N-nitrososarcosine; OCDD, octachlorodibenzo-p-dioxin; OCDF, octachlorodibenzo-p-furan; PAHs, polyaromatic hydrocarbons; PhIP, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine; total TCDD, total tetrachlorodibenzo-pdioxin; total-HpCDD, total heptachlorodibenzo-p-dioxin; total-HpCDF, total heptachlorodibenzo-p-furan; total-HxCDD, total hexachlorodibenzo-p-dioxin; total-HxCDF, total hexachlorodibenzo-p-furan; total-PeCDD, total pentachlorodibenzo-p-dioxin; total-PeCDF, total pentachlorodibenzo-p-furan; total-TCDF, total tetrachlorodibenzo-p-furan; TPM, total particulate matter; Trp-P-1, 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole; Trp-P-2, 1-methyl-3-amino-5H-pyrido[4,3-b]indole. ⇑ Corresponding author. Address: Regulatory Affairs, Altria Client Services, 601 East Jackson Street, Richmond, VA 23261, USA. Fax: +1 804 335 2081. E-mail addresses: [email protected], [email protected] (M.J. Oldham). http://dx.doi.org/10.1016/j.yrtph.2014.06.017 0273-2300/Ó 2014 Elsevier Inc. All rights reserved.
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148
person, and place on public display (in a manner determined by the Secretary) the list. . .’’ (Family Smoking Prevention and Tobacco Control Act, 2009, sec. 904). In addition, FDA has encouraged tobacco product manufacturers to include HPHC data in new product applications, although FDA has not been explicit in how it intends to use HPHC data when evaluating new tobacco products (FDA, 2011a,b,c). For both the purpose of consumer communication and potentially for the purpose of new tobacco product application evaluation, it is critical to understand and consider the limitations of tobacco product constituent analysis resulting from all sources of variability. Scientists and public health researchers have measured levels of chemical constituents to compare tobacco products for decades (Adams et al., 1987; Connolly et al., 2005; Ding et al., 2006; Gendreau and Vitaro, 2005; Hammond and O’Connor, 2008). Several large studies have demonstrated that measurements of tobacco and tobacco smoke constituents are not consistent as a result of assay, inter-laboratory, and temporal variability (Chepiga et al., 2000; Counts et al., 2004, 2005, 2006; Gaworski et al., 2011a; Morton and Laffoon, 2008; Oldham et al., 2012; Roemer et al., 2004). Estimates of assay and intra-laboratory variability can be made by comparing smoke chemistry analytical results for reference cigarettes that are manufactured as a single batch at a single point in time. For example, Gaworski et al. (2011a) included analyses of 52 mainstream cigarette smoke (MCS) constituents from the 1R4F and 2R4F reference cigarettes measured 107 times over a seven-year period at two different laboratories. The study found a general 10–15% relative standard deviation for the 39 of 52 smoke constituents that could be quantified, and it found significant differences between laboratories for some constituent measurements. It also included analyses of a control cigarette manufactured to the same specification 50 times over a seven-year period, which allowed assessment of temporal variability in MCS constituents. The authors reported variability to be greater for the control cigarette than for the reference cigarettes due to year-to-year variation in the tobacco crops used to manufacture the control cigarette. Recently, Purkis et al. (2012) summarized technical challenges and possible limitations for measurement of MCS constituents based on a review of Cooperation Centre for Scientific Research Relative to Tobacco (CORESTA) sponsored collaborative studies. Their review highlighted data variability issues (e.g., product variability; within and between laboratory variability) and the need for standardized analytical methodology. These studies clearly demonstrate that the inherent variability associated with constituent measurement in tobacco and tobacco smoke must be considered when comparing measured constituent levels between samples. The purpose of the current study was to evaluate the average relative standard deviation for all chemical constituents on a draft list of 96 HPHCs from MCS and smokeless tobacco products, which FDA published in 2011 (FDA, 2011d). The study samples were commercial cigarettes and smokeless tobacco products manufactured by Philip Morris USA and U.S. Smokeless Tobacco Company. In addition, samples from the same manufacturing lot were analyzed a second time for the constituents listed on an abbreviated list of HPHCs corresponding to those included in FDA’s draft guidance for constituent reporting (FDA, 2012b). This second analysis allowed assessment of short-term temporal analytical variability. These analyses confirm previous studies demonstrating the inherent variability in tobacco and tobacco smoke constituent measurement and extend these findings to the entire FDA draft HPHC list (FDA, 2011d). 2. Materials and methods Three contract laboratories analyzed tobacco products for all 96 HPHCs (Table 1) on the draft HPHC list (FDA, 2011d). The
139
chlorinated dioxins/furans measured in this study, which were listed as a single entity on the draft HPHC list (FDA, 2011d), are listed in Table 2. All three laboratories were ISO 17025 accredited, and all analytical methods for determination of HPHCs were on the laboratories’ ISO scope of accreditation at the time of testing. The three laboratories used were Arista Laboratories (Richmond, VA; Laboratory 1), Enthalpy Analytical, Inc. (Durham, NC; Laboratory 2), and Labstat International ULC (Ontario, Canada; Laboratory 3). Laboratories 1 and 2 were accredited by the American Association for Laboratory Accreditation, whereas Laboratory 3 was accredited by the Standards Council of Canada. Whether the HPHC was analyzed in cigarettes, smokeless tobacco products, or both, was based upon recommendations from the Tobacco Product Constituents Subcommittee of the Tobacco Products Scientific Advisory Committee (TPSAC, 2010). 2.1. Tobacco products Twenty commercial cigarette and 16 commercial smokeless tobacco products (Table 3) were assayed for all applicable HPHCs on the 96-item draft list (FDA, 2011d). The 3R4F reference cigarette (University of Kentucky) was used as the reference product for MCS constituents, and the CORESTA Reference Product 2 (CRP2; North Carolina State University) was used as the reference for smokeless tobacco constituents. In order to minimize analytical variability, all products were submitted to a single laboratory for a given method. Based upon previous experience in measuring MCS constituents (Counts et al., 2004, 2005; Morton and Laffoon, 2008; Gaworski et al., 2011b), three replicate analyses (each analysis may require multiple cigarettes) were chosen as sufficient to provide an estimate of each HPHC. The only exception was carbonyls in smoke, for which five replicates were used. The FDA draft guidance for reporting HPHCs (FDA, 2012b), which was published after this study started, recommended that seven replicates be used for each HPHC except smoke nicotine and carbon monoxide, for which twenty replicates were recommended. Therefore, to obtain the recommended number of replicates, for each tobacco product in this study a second set of samples was analyzed for the abbreviated list of HPHCs (FDA, 2012b) at the same three laboratories. For each tobacco product, a single manufacturing lot was used for all testing. 2.2. Smoking regimens MCS was generated under both ISO (2000b) and Health Canada (1999a) machine smoking regimens using commercially available linear and rotary smoking machines. Cigarettes were conditioned before smoking in accordance with ISO 3402 (ISO, 1999). MCS was generated and collected in basic accordance with ISO 3308 (ISO, 2000b) and ISO 4387 (ISO, 2000c). Deviations from the ISO standards were made when necessary in order to accommodate Health Canada smoking and in order to incorporate smoke traps for volatile or gas phase HPHCs. 2.3. HPHCs analytical assays for mainstream cigarette smoke Official methods (Health Canada; Centers for Disease Control and Prevention [CDC]) are cited, and other internally validated methods are briefly described for general understanding. 2.3.1. Nicotine and carbon monoxide in mainstream cigarette smoke Nicotine and carbon monoxide were determined using Health Canada Official Method T-115 (Health Canada, 1999a). Since this method also measures water and calculates tar, these data were also collected.
140
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148
Table 1 Draft list of harmful and potentially harmful constituents (FDA, 2011d) measured in mainstream cigarette smoke and smokeless tobacco products grouped by analytical method with available standardized analytical methods. Assay method constituent
MCS
Aromatic amines 1-Aminonaphthalene 2-Aminonaphthalene 4-Aminobiphenyl 2,6-Dimethylaniline o-Anisidine o-Toluidine
x x x x x x
Biological Aflatoxin B1 x x x x x x x
Heterocyclic aromatic amines A-a-C (2-amino-9H-pyrido ([2,3-b]indole) Glu-P-1 (2-amino-6-methyldipyridol[1,2-a:30 ,20 -d]imidazole) Glu-P-2 (2-aminodipyridol[1,2-a:30 ,20 -d]imidazole IQ (2-amino-d-methylimidazo[4,5-f]quinoline) MeA-a-C (2-amino-3-methyl)-9H-pyrido[2,3-b]indole) PhIP (2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine Trp-P-1 (3-amino1,4-dimethyl-5H-pyrido[4,3-b]indole) Trp-P-2 (1-methyl-3-amino-5H-pyrido[4,3-b]indole)
x xb,c xb,c xb,c x xb,c xb,c xb,c
N-Nitrosonornicotine (NNN) N-Nitrosopiperidine (NPIP) N-Nitrosopyrrolidine (NPYR) N-Nitrososarcosine (NSAR) Other Ammonia (listed as ammonia salts for STP) Anabasine Benzo[b]furan Caffeic acid Carbon monoxide Chlorinated dioxins/furansa Coumarin Dibenz[a,h]acridine Dibenz[a,j]acridine Dibenzo[c,g]carbazole Dibenzo[a,e]pyrene Dibenzo[a,h]pyrene Dibenzo[a,i]pyrene Dibenzo[a,l]pyrene Ethyl carbamate (urethane) Hydrazine Hydrogen cyanide Mercury 5-Methylchrysene Nicotine
Standardized analytical method
xf
Carbonyls Acetaldehyde Acetone Acrolein Crotonaldehyde Formaldehyde Methyl ethyl ketone Propionaldehyde
Metals Arsenic Beryllium Cadmium Chromium Cobalt Lead Nickel Selenium Nitrosamines N-Nitrosodiethylamine (NDEA) N-Nitrosodiethanolamine (NDELA) N-Nitrosodimethylamine (NDMA) N-Nitrosomethylethylamine (NMEA) N-Nitrosomorpholine (NMOR) 4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK)
STP
x
x x
x xb,c x xc xc x x x xd,e x x xe
x x x
xg xg xg x
x x x x
x xg x x
x c
x
CORESTA CORESTA CORESTA CORESTA
N° N° N° N°
75 72 75 72
(MCS) (STP) (MCS) (STP)
xf x xg x x
x xd,e x x
ISO 8454 (MCS) x
xb,c xb,c xb,c xe xc,d x xb,c xb xb,e x
xf
x x x
x
ISO 10315 (MCS) ISO 3400 (MCS) ISO 15517 (STP)
141
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148 Table 1 (continued) Assay method constituent
MCS
Nornicotine Phenolic compounds Catechol m-Cresol o-Cresol p-Cresol Phenol
x x x x x
Polycyclic aromatic hydrocarbons Benz[a]anthracene Benz[j]aceanthrylene Benzo[a]pyrene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[c]phenanthrene Chrysene Cyclopenta[c,d]pyrene Dibenz[a,h]anthracene Indeno[1,2,3-cd]pyrene Naphthalene
x x x x x x x x x x x
Radioactive Polonium-210 Uranium-235 Uranium-238
STP
Standardized analytical method
x
x
Semivolatiles 2-Nitropropane Acetamide Acrylamide Ethylbenzene Nitrobenzene Quinoline Styrene
xb,c x x x xb,c x x
Volatiles 1,3-Butadiene Acrylonitrile Benzene Ethylene oxide Furan Isoprene Nitromethane Propylene oxide Toluene Vinyl acetate Vinyl chloride
x x x x x x xb,c xb,c x xb,c xb,c
x x x x
ISO 22634 (MCS)
x xg x x x xf xf
CORESTA N° 70 (MCS) CORESTA N° 70 (MCS) CORESTA N° 70 (MCS)
CORESTA N° 70 (MCS)
CORESTA N° 70 (MCS)
CORESTA N° 70 (see CORESTA, 2013a); CORESTA N° 72 (see CORESTA, 2013b); CORESTA N° 75 (see CORESTA, 2012); ISO 3400 (see ISO, 1977); ISO 8454 (see ISO, 2007); ISO 10315 (see ISO, 2000a); ISO 15517 (see ISO, 2003); ISO 22634 (see ISO, 2008); MCS, mainstream cigarette smoke; STP, smokeless tobacco product. a See Table 2 for list of chlorinated dioxins/furans measured in mainstream cigarette smoke. b All values were less than limit of detection using ISO (2000b) smoking regimen. c All values were less than limit of detection using Health Canada (1999a) smoking regimen. d All values were less than limit of quantification using ISO (2000b) smoking regimen. e All values were less than limit of quantification using Health Canada (1999a) smoking regimen. f All values were less than limit of detection. g All values were less than limit of quantification.
2.3.2. Carbonyl in mainstream cigarette smoke Acetaldehyde, acetone, acrolein, crotonaldehyde, formaldehyde, methyl ethyl ketone, and propionaldehyde were determined using Health Canada Official Method T-104 (Health Canada, 1999b). 2.3.3. Nitrosamines in mainstream cigarette smoke N-Nitrosodiethylamine (NDEA), N-nitrosodiethanolamine (NDELA), N-nitrosodimethylamine (NDMA), N-nitrosomethylethylamine (NMEA), N-nitrosomorpholine (NMOR), 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), N-nitrosonornicotine (NNN), N-nitrosopiperidine (NPIP), N-nitrosopyrrolidine (NPYR), and N-nitrososarcosine (NSAR) were determined in MCS by collecting total particulate matter (TPM) on a 44-mm Cambridge filter pad. The pad was subsequently extracted in 10 mL of aqueous extraction solution containing internal standards. The solution was solvent-exchanged to methylene chloride and analyzed by gas chromatograph–mass spectrometry–mass spectrometry (GC/
MS/MS) for tobacco specific nitrosamines and volatile nitrosamines. NDELA was derivatized with N-methyl-n-(tert-butyldimethylsilyl)trifluoroacetamide/tert-butyldimethylchlorosilane prior to analysis and was analyzed in a separate analytical run from the other nitrosamines. 2.3.4. Polyaromatic hydrocarbons and heterocyclic compounds in mainstream cigarette smoke Benz[a]anthracene, benz[j]aceanthrylene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[a]pyrene, benzo[c]phenanthrene, chrysene, cyclopenta[c,d]pyrene, dibenz[a,h]acridine, dibenz[a,j] acridine, dibenz[a,h]anthracene, dibenzo[c,g]carbazole, dibenzo [a,e]pyrene, dibenzo[a,h]pyrene, dibenzo[a,i]pyrene, dibenzo[a,l] pyrene, indeno[1,2,3-cd]pyrene, 5-methylchrysene, and naphthalene were determined by passing MCS through a 92-mm filter pad. After addition of internal standard, polyaromatic hydrocarbons (PAHs) and heterocyclic compounds were extracted from
142
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148
Table 2 Chlorinated dioxins/furans measured in mainstream cigarette smoke.
Table 3 List of cigarettes and smokeless tobacco products tested.
Assay method constituent
MCS
Product name
1,2,3,4,6,7,8-HpCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF 1,2,3,4,7,8-HxCDD 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDD 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDD 1,2,3,7,8,9-HxCDF 1,2,3,7,8-PeCDD 1,2,3,7,8-PeCDF 2,3,4,6,7,8-HxCDF 2,3,4,7,8-PeCDF 2,3,7,8-TCDD 2,3,7,8-TCDF OCDD OCDF Total TCDD Total-HpCDD Total-HpCDF Total-HxCDD Total-HxCDF Total-PeCDD Total-PeCDF Total-TCDF
x xb,c xa,b xa,b xa,b xa,b xa,b xa,b xa,b xa,b xa,b xa,b xc,d xa,b x x xa,b xa,b x xb,c xc,d xa,b xa,b xc,d x
Cigarettes Marlboro 1000 s Box Marlboro 720 s Box Marlboro 720 s Gold Pack Box Marlboro Blend No. 27 Box Marlboro Box Marlboro Gold Pack 1000 s Box Marlboro Gold Pack Box Marlboro Gold Pack Soft Pack Marlboro Menthol Box Marlboro Menthol Gold Pack Box Marlboro Red Label Box Marlboro Silver Pack 1000 s Box Marlboro Silver Pack Box Marlboro Soft Pack Marlboro Special Blend (Gold Pack) 1000 s Box Marlboro Special Blend (Gold Pack) Box Marlboro Special Blend (Red Pack) 1000 s Box Marlboro Special Blend (Red Pack) Box Marlboro Virginia Blend Box Parliament (White Pack) Box
1,2,3,4,6,7,8-HpCDD, 1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin; 1,2,3,4,6,7,8HpCDF, 1,2,3,4,6,7,8-heptachlorodibenzo-p-furan; 1,2,3,4,7,8,9-HpCDF, 1,2,3,4,7,8, 9-heptachlorodibenzo-p-furan; 1,2,3,4,7,8-HxCDD, 1,2,3,4,7,8-hexachlorodibenzop-dioxin; 1,2,3,4,7,8-HxCDF, 1,2,3,4,7,8-hexachlorodibenzo-p-furan; 1,2,3,6,7,8HxCDD, 1,2,3,6,7,8-hexachlorodibenzo-p-dioxin; 1,2,3,6,7,8-HxCDF, 1,2,3,6,7, 8-hexachlorodibenzo-p-furan; 1,2,3,7,8,9-HxCDD, 1,2,3,7,8,9-hexachlorodibenzo-pdioxin; 1,2,3,7,8,9-HxCDF, 1,2,3,7,8,9-hexachlorodibenzo-p-furan; MCS, mainstream cigarette smoke; 1,2,3,7,8-PeCDD, 1,2,3,7,8-pentachlorodibenzo-p-dioxin; 1,2,3,7,8-PeCDF, 1,2,3,7,8-pentachlorodibenzo-p-furan; 2,3,4,6,7,8-HxCDF, 2,3,4,6,7,8-hexachlorodibenzo-p-furan; 2,3,4,7,8-PeCDF, 2,3,4,7,8-pentachlorodibenzo-p-furan; 2,3,7,8-TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; 2,3,7,8-TCDF, 2,3,7,8-tetrachlorodibenzo-p-furan; OCDD, octachlorodibenzo-p-dioxin; OCDF, octachlorodibenzo-p-furan; total TCDD, total tetrachlorodibenzo-p-dioxin; total-HpCDD, total heptachlorodibenzo-p-dioxin; total-HpCDF, total heptachlorodibenzo-p-furan; total-HxCDD, total hexachlorodibenzo-p-dioxin; total-HxCDF, total hexachlorodibenzo-p-furan; total-PeCDD, total pentachlorodibenzo-p-dioxin; total-PeCDF, total pentachlorodibenzo-p-furan; totalTCDF, total tetrachlorodibenzo-p-furan. a All values were less than limit of detection using ISO (2000b) smoking regimen. b All values were less than limit of detection Health Canada (1999a) smoking regimen. c Mean value was less than limit of quantification or one or more values were less than limit of detection using ISO (2000b) smoking regimen. d Mean value was less than limit of quantification or one or more values were less than limit of detection using Health Canada (1999a) smoking regimen.
the pad with methanol. The methanol extract was filtered, and a portion of the filtered extract was diluted with water and passed through a C18 cartridge. Analytes were eluted from the cartridge with cyclohexane and analyzed by gas chromatography–mass spectrometry (GC/MS). 2.3.5. Benzo[b]furan in mainstream cigarette smoke MCS was passed through a 92-mm Cambridge filter pad followed by two cryogenic impingers containing methanol. Following collection, the pad was placed into an Erlenmeyer flask and internal standard was added. The pad extract was combined with impinger solutions and quantified using GC/MS. 2.3.6. Aromatic amines in mainstream cigarette smoke 1-Aminonaphthalene, 2-aminonaphthalene, 4-aminobiphenyl, 2,6-dimethylaniline, o-anisidine, and o-toluidine were determined by collecting MCS on a 92-mm Cambridge filter pad. The pad was quartered and placed in an Erlenmeyer flask with 100 mL of 5% hydrochloric acid solution and shaken for 30 min. Extract was
Smokeless tobacco products Copenhagen Long Cut Copenhagen Long Cut Straight Copenhagen Long Cut Wintergreen Copenhagen Pouches Copenhagen Snuff Fine Cut Marlboro Snus Smooth Mint Red Seal Fine Cut Natural Red Seal Fine Cut Wintergreen Red Seal Long Cut Wintergreen Skoal Long Cut Mint Skoal Long Cut Straight Skoal Long Cut Wintergreen Skoal Original Fine Cut Wintergreen Skoal Pouches Mint Skoal Pouches Wintergreen Skoal X-tra Long Cut Wintergreen Blend
filtered and spiked with internal standards. The filtrate was washed with dichloromethane, made basic with sodium hydroxide solution and extracted with hexane. The hexane extract was dried with sodium sulfate, derivatized with pentafluoropropionic acid anhydride and trimethylamine, and then passed through a FlorisilÒ column (U.S. Silica Co., Berkeley Springs, WV). The aromatic amines in the hexane extract were then quantified by GC/MS. 2.3.7. Ammonia in mainstream cigarette smoke MCS was passed through a 44-mm Cambridge filter pad followed by one impinger containing 40 mL of 40 mM aqueous sulfuric acid. The filter pad was added to the impinger solution and mixed. The solution was filtered and analyzed by ion chromatography with conductivity detection. 2.3.8. Volatile organic compounds in mainstream cigarette smoke 1,3-Butadiene, acrylonitrile, benzene, ethylene oxide, furan, isoprene, nitromethane, propylene oxide, toluene, vinyl acetate, and vinyl chloride were determined in MCS by passing the smoke through a 44-mm Cambridge pad followed by one impinger containing 20 mL of methanol and internal standards. The impinger was cooled by a dry ice/isopropanol bath. While still cold, an aliquot was quickly transferred to a zero-headspace vial and then analyzed by GC/MS. 2.3.9. Semi-volatile organic compounds in mainstream cigarette smoke 2-Nitropropane, acetamide, acrylamide, ethylbenzene, nitrobenzene, quinoline, and styrene were determined in MCS by passing the
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148
smoke through a 44-mm Cambridge pad followed by one impinger containing 20 mL of methanol and internal standards. The impinger was cooled by a dry ice/isopropanol bath. The Cambridge pad was added to the impinger solution and shaken. An aliquot of the solution was analyzed by GC/MS. Since this method also measures pyridine, these data were also collected. 2.3.10. Chlorinated dioxins and furans in mainstream cigarette smoke The following chlorinated dioxins and furans were measured: 1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin (1,2,3,4,6,7,8-HpCDD); 1, 2,3,4,6,7,8-heptachlorodibenzo-p-furan (1,2,3,4,6,7,8-HpCDF); 1,2,3, 4,7,8,9-heptachlorodibenzo-p-furan (1,2,3,4,7,8,9-HpCDF); 1,2,3,4,7, 8-hexachlorodibenzo-p-dioxin (1,2,3,4,7,8-HxCDD); 1,2,3,4,7,8hexachlorodibenzo-p-furan (1,2,3,4,7,8-HxCDF); 1,2,3,6,7,8-hexachl orodibenzo-p-dioxin (1,2,3,6,7,8-HxCDD); 1,2,3,6,7,8-hexachlorodi benzo-p-furan (1,2,3,6,7,8-HxCDF); 1,2,3,7,8,9-hexachlorodibenzop-dioxin (1,2,3,7,8,9-HxCDD); 1,2,3,7,8,9-hexachlorodibenzo-p-furan (1,2,3,7,8,9-HxCDF); 1,2,3,7,8-pentachlorodibenzo-p-dioxin (1,2,3, 7,8-PeCDD); 1,2,3,7,8-pentachlorodibenzo-p-furan (1,2,3,7,8-PeCDF); 2,3,4,6,7,8-hexachlorodibenzo-p-furan (2,3,4,6,7,8-HxCDF); 2,3,4,7,8-pentachlorodibenzo-p-furan (2,3,4,7,8-PeCDF); 2,3,7, 8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD); 2,3,7,8-tetrachlorodibenzo-p-furan (2,3,7,8-TCDF); octachlorodibenzo-p-dioxin (OCDD); octachlorodibenzo-p-furan (OCDF); total heptachlorodi benzo-p-dioxin (total-HpCDD); total heptachlorodibenzo-p-furan (total-HpCDF); total hexachlorodibenzo-p-dioxin (total-HxCDD); total hexachlorodibenzo-p-furan (total-HxCDF); total pentachlorodibenzo-p-dioxin (total-PeCDD); total pentachlorodibenzo-pfuran (total-PeCDF); total tetrachlorodibenzo-p-dioxin (total TCDD); and total tetrachlorodibenzo-p-furan (total-TCDF). TPM was collected on Cambridge filter pads, which were spiked with labeled internal standards and Soxhlet-extracted for 16 h. The extract was subjected to an acid/base clean-up procedure followed by clean-up on micro columns of silica gel and alumina. The extract was spiked with 0.1 ng 13C-1,2,3,4-TCDD and 13C-1,2,3,4,7,8, 9-HCDD (to determine extraction efficiencies achieved for the 13Clabeled internal standards) and then concentrated to 20 lL. Analysis was conducted using GC-high resolution mass spectrometry. 2.3.11. Caffeic acid in mainstream cigarette smoke MCS was passed through a 44-mm Cambridge filter pad. The pad was shaken for 30 min with 40 mL of Type I water in an Erlenmeyer flask. The extract was then filtered with a 0.45 lm syringe filter, and aliquots were analyzed for caffeic acid using high performance liquid chromatography (HPLC) with ultraviolet light detection. 2.3.12. Hydrogen cyanide in mainstream cigarette smoke Hydrogen cyanide was determined using Health Canada Official Method T-107 (Health Canada, 1999c). 2.3.13. Hydrazine in mainstream cigarette smoke MCS was passed through a dinitrophenylhydrazine-treated 92-mm Cambridge pad followed by two impingers. Each impinger contained 40 mL of a solution consisting of methanol and aqueous citrate–phosphate buffer with added 2-nitrobenzaldehyde. Immediately after smoking, the filter pad was extracted with the impinger solutions and the extract was incubated for 30 min at 35 °C. An aliquot of extract was centrifuged and the resultant hydrazone (i.e., dinitrophenyl-hydrazone) was quantified by HPLC/MS/MS using an internal standard. 2.3.14. Heterocyclic amines in mainstream cigarette smoke 2-Amino-9H-pyrido[2,3-b]indole (A-a-C), 2-amino-6-methyldipyrido[1,2-a:30 ,20 -d]imidazole (Glu-P-1), 2-aminodipyrido[1, 2-a:30 ,20 -d]imidazole (Glu-P-2), 2-amino-3-methylimidazo[4,5-f]
143
quinoline (IQ), 2-amino-3-methyl-9H-pyrido[2,3-b]indole (MeAa-C), 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1), and 1-methyl-3-amino-5H-pyrido[4,3-b]indole (Trp-P-2) were determined by collecting MCS on a Cambridge filter pad. Internal standard was added to the pad, which was then extracted with formic acid. Extracts were filtered before analysis by HPLC/MS/ MS using positive electrospray ionization. 2.3.15. Metals in mainstream cigarette smoke Arsenic, beryllium, cadmium, chromium, cobalt, lead, nickel, and selenium were determined by collecting MCS condensate using an electrostatic precipitation unit equipped with a quartz collection tube. Collected material was rinsed from the quartz tube using methanol. The methanol was evaporated, and the resulting sample was digested using nitric acid, hydrogen peroxide, and heat. The digested samples were subsequently analyzed by inductively coupled plasma mass spectroscopy (ICP/MS) equipped with a dynamic reaction cell. 2.3.16. Phenolic compounds in mainstream cigarette smoke Catechol, m-cresol, o-cresol, p-cresol, and phenol were determined in MCS by collecting TPM on a 44-mm Cambridge filter pad. The pad was then extracted with 40 mL of 1% acetic acid. An aliquot of the TPM extract was filtered using a syringe filter and then diluted. The resulting extract was analyzed by HPLCfluorescence. 2.3.17. Polonium-210 in mainstream cigarette smoke Polonium-210 (210Po) was determined in MCS by collecting TPM on a 44-mm Cambridge filter pad. The pad was spiked with polonium-209 (209Po). The entire pad and sample was digested with concentrated hydrofluoric acid, and then heated to dryness. Aqueous dilute hydrochloric acid was then added to the residue. The sample was then deposited on silver or nickel foil in the presence of dilute hydrochloric acid. The resulting sample was analyzed by alpha spectrometry (209Po a-particle energy = 4.9 MeV and 210Po a-particle energy = 5.3 MeV). The calculated concentration of 210Po was based upon the determination of 209Po. 2.3.18. Ethyl carbamate in mainstream cigarette smoke Ethyl carbamate (urethane) was determined by collecting MCS on a Cambridge filter pad. The filter was extracted in 1:1 water/ methanol and then filtered. The sample was then derivatized with 9-xanthydrol in acid. The resulting samples were analyzed using HPLC-fluorescence. 2.4. Harmful and potentially harmful constituents analytical assays for smokeless tobacco Smokeless products were generally analyzed as subsamples from a composite sample that was created from a single lot of product. 2.4.1. Oven volatiles in smokeless tobacco Percent oven volatiles were determined in smokeless tobacco using Association of Analytical Communities International Official Method 966.02 (AOAC, 1990). 2.4.2. Nicotine and minor alkaloids in smokeless tobacco Nicotine (total), nornicotine, and anabasine were determined in smokeless tobacco using Health Canada Official Method T-301 (Health Canada, 1999d).
144
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148
2.4.3. pH in smokeless tobacco pH was determined in smokeless tobacco using the CDC method (CDC, 1999, 2009). 2.4.4. Metals in smokeless tobacco Arsenic, beryllium, cadmium, chromium, lead, nickel, and selenium were determined in smokeless tobacco by digesting tobacco samples using nitric acid in a closed-vessel microwave digestion. Digested samples were subsequently analyzed by ICP/MS. Since this method also measures cobalt, these data were also collected. 2.4.5. Mercury in smokeless tobacco Tobacco samples were digested in a microwave using nitric acid and potassium permanganate. Hydroxylamine hydrochloride was added after digestion, and the resulting solutions were analyzed for mercury using cold vapor atomic absorption spectrometry (flow injection). 2.4.6. Carbonyls in smokeless tobacco Acetaldehyde, crotonaldehyde, and formaldehyde were determined in smokeless tobacco by extracting tobacco with an aqueous/acetonitrile solution containing acidified dinitrophenylhydrazine. An aliquot of the extract was subsequently treated with base to stabilize the hydrazones. The samples were filtered and then analyzed using reversed-phase HPLC with UV detection. 2.4.7. Polyaromatic hydrocarbons in smokeless tobacco Benz[a]anthracene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[a]pyrene, chrysene, dibenz[a,h]anthracene, indeno [1,2,3-cd]pyrene, and naphthalene were determined in smokeless tobacco by extracting tobacco with methanol containing internal standards. The sample extracts were then diluted with water and passed through a solid-phase extraction cartridge. The analytes of interest were determined in the resulting solution with GC/MS. 2.4.8. Nitrosamines in smokeless tobacco NDEA, NDELA, NDMA, NMOR, NNK, NNN, NPIP, NPYR, and NSAR were determined in smokeless tobacco by extracting the tobacco in 10 mL of aqueous extraction solution containing internal standards. The solution was solvent-exchanged to methylene chloride and analyzed by GC/MS/MS for tobacco specific nitrosamines and volatile nitrosamines. NDELA was derivatized with N-methyl-n(tert-butyldimethylsilyl) trifluoroacetamide/tert-butyldimethylchlorosilane prior to analysis and was analyzed in a separate run. 2.4.9. Aflatoxin-B1 in smokeless tobacco Smokeless tobacco was extracted using methanol and water (4:1). Samples were filtered and diluted with phosphate buffered saline solution. Each sample was introduced to an immunoaffinity column, and the column was rinsed with water before elution of the aflatoxin-B1 with methanol. The eluate was diluted 1:1 with water and injected onto an HPLC-fluorescence system using postcolumn derivatization. 2.4.10. Ethyl carbamate in smokeless tobacco Ethyl carbamate (urethane) was determined in smokeless tobacco by extracting the tobacco samples in 1-propanol. Extracts were filtered and derivatized with 9-xanthydrol in acid. The resulting samples were analyzed using HPLC-fluorescence. 2.4.11. Coumarin in smokeless tobacco Coumarin was determined in smokeless tobacco by spiking each tobacco sample with internal standard and then extracting with 1:1 ethanol in water. Samples were filtered and then diluted with 1:1 methanol in water. Quantification was accomplished with
LC/MS/MS using positive electrospray ionization and multiple reactions monitoring mode. Three mass transition pairs (147.1/ 103.1; 147.1/91.1; 147.1/77.1) were monitored for identification, confirmation, and quantification. 2.4.12. Polonium-210 in smokeless tobacco 210 Po was determined in smokeless tobacco by spiking each sample with a traceable standard containing a known amount of 209 Po as a tracer. All sample matrices were digested with concentrated nitric acid until all organic material was dissolved, and then heated to dryness. Aqueous dilute hydrochloric acid was then added to the residue. The 209Po/210Po was then deposited on silver or nickel foil in the presence of dilute hydrochloric acid and alpha spectrometry performed on the foil (209Po a-particle energy = 4.9 MeV and 210Po a-particle energy = 5.3 MeV). Results for 210Po were calculated according to the amount of 209Po determined. 2.4.13. Uranium-238 in smokeless tobacco Uranium-238 (238U) was determined in smokeless tobacco using neutron activation. Samples were irradiated in a nuclear reactor (flux of 8 1012 neutrons/cm2/s) for up to 20 min. The samples were rotated during irradiation so that there was no horizontal flux variation (vertical flux variation was monitored with the individual flux monitors). Samples were allowed to decay for 6 d before counting using a high-resolution gamma-ray spectrometer (germanium detector connected to multichannel analyzer). Counting time varied between 20 and 30 min per sample. 238U was quantitatively identified by comparing the peak positions and areas results to library standards. 2.4.14. Uranium-235 in smokeless tobacco Uranium-235 (235U) was determined in smokeless tobacco using delayed neutron counting analysis. Samples were irradiated thermally in a nuclear reactor (flux of 8 1012 neutrons/cm2/s), and the delayed neutrons that were produced were counted to determine the concentration of 235U by comparison with standards. 2.5. Statistical analysis Analytes on the abbreviated HPHC list (FDA, 2012b) were tested at two different time points. The two different time points were compared using a two-way analysis of variance including both time and product as factors in the analysis. A statistically significant time effect is indicative of temporal variation in the analytical method.
3. Results and discussion Because the FDA draft list (FDA, 2011d) included several chemicals that were listed as a single entity (m-cresol, o-cresol and p-cresol, which were listed only as cresol), and it listed classes of chemicals (chlorinated dioxins and furans) as a single HPHC, the total number of HPHCs reported in this study was 113 for MCS and 37 for smokeless tobacco products. Three additional compounds (pyridine, tar, and water) are also reported for MCS because of the analytical methodology used. In addition to the HPHCs measured in smokeless tobacco products, cobalt, moisture, and pH were also measured in this study. As noted in Table 1, there are a limited number of standardized methods that only apply to a few HPHCs. None of the standardized methods noted in Table 1 were used by the testing laboratories.
145
M.J. Oldham et al. / Regulatory Toxicology and Pharmacology 70 (2014) 138–148 Table 4 Replicate-to-replicate variability of HPHCs using ISO (2000b) and Health Canada (1999a) smoking regimens. Constituent
Average %RSDa ISO
1,2,3,4,6,7,8-HpCDD 1,3-Butadiene 1-Aminonaphthalene 2,3,7,8-TCDF 2,6-Dimethylaniline 2-Aminonaphthalene 4-Aminobiphenyl 5-Methylchrysene A-a-C Acetaldehyde Acetamide Acetone Acrolein Acrylamide Acrylonitrile Ammonia Arsenic Benz[a]anthracene Benz[j]aceanthrylene Benzene Benzo[a]pyrene Benzo[b]fluoranthene Benzo[b]furan Benzo[c]phenanthrene Benzo[k]fluoranthene Cadmium Carbon monoxide Catechol Chrysene Crotonaldehyde Cyclopenta[c,d]pyrene Dibenz[a,h]anthracene Dibenzo[a,i]pyrene Ethyl carbamate (urethane) Ethylbenzene Ethylene oxide Formaldehyde Furan Hydrogen cyanide Indeno[1,2,3-cd]pyrene Isoprene Lead MeA-a-C Methyl ethyl ketone N-Nitrosodimethylamine (NDMA) N-Nitrosopiperidine (NPIP) N-Nitrosopyrrolidine (NPYR) NNK NNN Naphthalene Nicotine OCDD Phenol Polonium-210 Propionaldehyde Pyridine Quinoline Selenium Styrene Tar Toluene Total-HpCDD Total-HxCDD Total-TCDF Water m-Cresol o-Anisidine o-Cresol o-Toluidine p-Cresol
Health Canada
%
Number of products’ data used in %RSD
%
Number of products’ data used in %RSD
16.4 8.1 5.7 0.0 6.2 5.4 6.8 12.2 10.7 5.5 10.4 5.3 6.6 10.0 6.0 6.1 5.5 5.4 8.4 6.4 6.0 8.7 12.2 8.9 10.9 4.9 6.1 5.4 5.2 10.3 8.3 14.4 15.9
