Analysis of Tobacco-Specific Nitrosamines in the Common Smokeless Tobacco Afzal in Oman

OBJECTIVES
There is a lack of awareness regarding the carcinogenicity of Afzal, an illegal smokeless tobacco product (STP) widely used among the Omani youth. Previous research has shown that certain types of tobacco-specific nitrosamines (TSNAs) are associated with oral and lung cancers. This study therefore aimed to assess levels of four common TSNAs in a randomly selected sample of Afzal.


METHODS
This study was carried out at Sultan Qaboos University in Muscat, Oman, between April and September 2013. A random sample of Afzal was analysed for four types of TSNAs using high-performance liquid chromatography-tandem mass spectrometry. The four types of TSNAs analysed were 4-(N-nitrosomethylamino)-1-(3-pyridyl)-1-butanone (NNK), N-nitrosonornicotine (NNN), N-nitrosoanatabine (NAT) and N-nitrosoanabasine (NAB). As a reference product, a sample of laboratory-manufactured American moist snuff (Centers for Disease Control and Prevention, Atlanta, Georgia, USA) was also used to confirm the accuracy and precision of the analysis.


RESULTS
The analysis revealed total TSNA levels of 3.573 μg/g in the Afzal sample. Mean levels of NNN, NNK, NAT and NAB were 1.205, 1.015, 0.809 and 0.545 μg/g, respectively.


CONCLUSION
Levels of the two most abundant TSNAs (NNN and NNK) found in the Afzal sample exceeded international regulatory limits. Afzal users therefore need to be educated regarding the potential health risks associated with their STP use. Stricter implementation of current legislation is recommended to reduce the availability and usage of Afzal in Oman.


S
mokeless tobacco products (stps) represent a significant health risk and have been associated with oral and pancreatic cancers, oral lesions, coronary artery and peripheral vascular disease and adverse pregnancy outcomes. 1pproximately 28 carcinogens have been identified in STPs so far. 2 Tobacco-specific nitrosamines (TSNAs) are considered a potent class of carcinogens in STPs. 2 During the STP curing process, TSNAs form in the leaves and increase if the tobacco is subsequently fermented. 3Levels of TSNAs are also dependent on other factors, such as the basic pH level and the nitrite/ nitrate content of the product. 4The moisture content of the product and the related increase in microbial action are other causes of increased TSNA content in STPs. 3 There are four primary alkaloids in tobacco-nicotine, nornicotine, anatabine and anabasine-each of which can be nitrosated by nitrite to form the following TSNAs: 4-(N-nitrosomethylamino)-1-(3-pyridyl)-1butanone (NNK), N-nitrosonornicotine (NNN), N-nitrosoanatabine (NAT) and N-nitrosoanabasine (NAB), respectively. 4The International Agency for Research on Cancer classifies NNN and NNK as group one carcinogens. 5In contrast, levels of NAT and NAB are lower in tobacco and their carcinogenicity is thought to be much weaker than NNN or NNK. 4 Although NAB has been classified as a weak oesophageal carcinogen in rats, both NAB and NAT are classified as group three carcinogens, indicating that they are not classifiable as a human carcinogen. 5,6owever, the Food and Drug Administration in the USA considers nitrosamines as carcinogens which are not safe at any level. 7ifferent types of STPs have shown wide variations in their components and are therefore not equal in their delivery of carcinogenic TSNAs into the human body. 8The World Health Organization (WHO) reports that Swedish snus has the lowest level of nitrosaminethe most dangerous carcinogen-among STPs available on the global market. 9In contrast, the highest levels of TSNAs have been detected in Sudanese toombak. 10In the USA, the three most popular brands of snuff were found to have high concentrations of TSNAs. 11Afzal is a STP widely used by youth and teenagers in Oman. 12,13This may be perhaps due to its ease of use during manual work compared to smoked tobacco, its low cost and the limited public awareness of its harmful effects. 3High levels of certain heavy metals and cancer-enhancing anions in Afzal have previously been reported. 12,13This study aimed to analyse the levels of TSNAs in a random sample of Afzal in order to provide further evidence of the potential carcinogenic risk of this product.

Methods
This study was carried out at the Sultan Qaboos University (SQU) in Muscat, Oman, between April and September 2013.A single package of 4.00 kg of Afzal was purchased by the researchers from one source in order to maintain uniformity throughout the study.The Afzal sample was labelled, pH and moisture levels were tested and the sample was refrigerated as previously described. 12,13The analysis was undertaken within six months of manufacture of the product.In order to enhance the quality of the data and to confirm the accuracy and precision of the analysis, a sample of American moist snuff was utilised as a reference.The certified reference product-2 (CRP-2) is purposely manufactured for laboratory analysis by the Tobacco and Volatiles Branch of the Division of Laboratory Sciences at the Centers for Disease Control and Prevention ([CDC] Atlanta, Georgia, USA) and is not commercially available.This manufactured product is useful for comparing reference values obtained from different international analytical laboratories. 14,15amples of CRP-2 were thus prepared and analysed using the same methods as the Afzal samples [Figure 1].Simultaneously, the CRP-2 samples were dealt with in the analysis as a different STP sample matrix in order to prove the validity of the method used for other STPs.
Afzal and CRP-2 samples were dried separately in accordance with standard protocols from the CDC to determine moisture content in STPs. 16Approximately 15.00 g each of Afzal and CRP-2 were weighed separately in pre-weighed moisture dishes and placed uncovered in an oven at 99 ± 1 ºC for three hours.The samples were then removed from the oven, covered and cooled in a desiccator at room temperature for approximately 30 minutes to avoid drawing moisture from the air.After drying, the samples were ground separately to form a homogenised powder.
The TSNA analysis was performed according to methods described by Lawler et al. with minor modifications. 8Three Afzal samples and a single CRP-2 sample of 0.50 g each were transferred to 50 mL volumetric flasks and extracted using 10 mL of 100 mM aqueous ammonium acetate buffer.The samples were then shaken using an incubator shaker (KS 4000 i control shaker, IKA ® Werke GmbH & Co, Staufen im Breisgau, Germany) at 250 revolutions per minute (rpm) for one hour.Each extract was air-filtered twice with filter paper (Whatman ® grade 42 filtration paper, Sigma-Aldrich Corp., St. Louis, Missouri, USA) followed by a 0.22 µm pore-sized nylon syringe filter (Whatman ® GD/X syringe filter, Sigma-Aldrich Corp.).Analytical standards of NNN, NNK, NAT and NAB were purchased (Sigma-Aldrich Corp.) Methanol dilution was used to prepare five different concentrations of each analyte (10, 30, 50, 70 and 100 parts per billion [ppb]).Standard calibration curves were plotted for each of the TSNAs.For accuracy and precision, the relative standard deviation (RSD), limit of detection (LOD), limit of quantification (LOQ) and r-value of 10 replicate injections of 10 ppb each of the standards were calculated.
Analytical separation of the Afzal and the CRP-2 samples was performed using a high-performance liquid chromatography (HPLC) system (1290 Infinity LC System, Agilent Technologies, Santa Clara, California, USA) with a Zorbax SB C18 1.8 µm, 2.1 x 50 mm stationary phase column (Agilent Technologies).Eluent A (aqueous phase) was a five mM ammonium acetate solution, whereas eluent B (organic phase) was comprised of acetonitrile and water at a ratio of 70:30, along with 5 mM of ammonium acetate.The column temperature was kept at 40 ºC and the flow rate was constant at 0.4 mL/minute.The eluents were run in a gradient manner with a running time of five minutes.The detector used was a 6460 Triple Quadrupole LC/ MS (Agilent Technologies).All chromatographic data were processed using MassHunter Workstation software B.06.00 (Agilent Technologies).All chemicals used in the analysis were from Sigma-Aldrich Corp. and were of analytical grade.Deionised water from a Milli-Q ® Integral Water Purification System (EMD Millipore Corp., Bedford, Massachusetts, USA) was used throughout the analysis.Machine detection limits were calculated for each of the analysed TSNAs.Statistical analysis was performed using Excel spreadsheet software, Version 2010 (Microsoft Corp., Redmond, Washington, USA).As this study was a chemical analysis of two forms of STPs, it did not require ethical approval.

Results
The method used for the analysis of the Afzal samples showed high sensitivity and reliable validity, as indicated by the LOD, LOQ, RSD and other accuracy and linearity values of the calibration curves for all four TSNA types [Table 1].Concentrations of all four TSNAs were detected in both the Afzal and the CRP-2 samples [Table 2].TSNA levels for each of the three Afzal samples were very similar, reflecting the precision of the analysis.In the Afzal samples, mean analyte levels of NNN, NNK, NAT and NAB were 1.205 ± 0.019 µg/g, 1.015 ± 0.004 µg/g, 0.809 ± 0.007 µg/g and 0.545 ± 0.047 µg/g, respectively.For the CRP-2

*American moist snuff product purposely manufactured for laboratory analysis by the Tobacco and Volatiles Branch of the Division of Laboratory Sciences at the Centers for Disease Control and Prevention (Atlanta, Georgia, USA).
sample, mean levels of the same TSNAs were 1.793 ± 0.970 µg/g, 0.465 ± 0.280 µg/g, 1.668 ± 0.650 µg/g and 0.209 ± 1.110 µg/g, respectively.Levels of NNN were highest while NAB levels were lowest for both the Afzal and the CRP-2 samples.Overall, the CRP-2 sample contained higher mean total TSNA levels (4.135 µg/g) than the Afzal sample (3.573 µg/g).
The observed TSNA values for the CRP-2 sample were in line with certified reference values provided by the CDC, 15 validating the results of the analysis [Table 3].Chromatograms of the analysed TSNAs in all three Afzal samples were very similar, confirming the presence of the same analytes in each sample [Figure 2].Chromatograms of TSNAs in the CRP-2 sample were comparable to those of Afzal samples, although there were slight differences between each analyte [Figure 3].These chromatograms of TSNAs in CRP-2 showed comparable results, reflecting the effectiveness of the method used for the new STP sample matrix.

Discussion
Nitrosamines have been strongly associated with several human and animal cancers. 4While nitrosamines are present in many foods and items which come into contact with food in amounts ranging from 1-10 ng/g, the highest exposure to humans comes from tobacco use. 17These chemical compounds have been detected in the saliva of tobacco chewers and are believed to form in the digestive system. 17All STPs demonstrate a wide diversity in their chemical compositions which result from variations in geographical location, additives used and manufacturing methods. 4In order to avoid potentially diverse chemical compositions, one random sample of Afzal was chosen to conduct several chemical and toxicological tests.The presence of TSNAs in the Afzal samples was confirmed using a modified version of Lawler et al.'s method. 8vels of nitrosamines in green tobacco leaves are negligible; however, several factors may foster the formation of TSNAs in STPS, including the damping, curing and fermentation processes, storage temperatures, high nitrate/nitrite content, alkaline pH and high moisture levels. 18,19Most of these factors are present in Afzal.As reported previously, samples of Afzal were observed to have an alkaline pH (10.460), high moisture levels (52%) and a high nitrate concentration (8,792.200μg/g). 12In general, Afzal is stored illicitly in local STP shops or retail outlets where there is likely to be poor hygiene, a lack of air circulation and high temperatures which dramatically increase in the summer.Shi et al. confirmed that STP storage temperatures of above 30 °C increased TSNA formation significantly, as did high levels of nitrate. 20 study by the WHO revealed that storage at room temperature increases TSNA concentrations in STPs due to microbial action.9 Different manufacturing processes also play a role in TSNA level variations.High TSNA levels in American moist snuff and low TSNA levels in Swedish snus have been reported to be due to the fermentation process in the former and heat treatment or a pasteurisation-like process in the latter.9 Heat treatment gives rise to more sterile products by killing the bacteria implicated in the formation of nitrosamines, while fermentation leads to a high concentration of nitrites.21 Afzal is manufactured using a fermentation process; this may therefore facilitate the formation of TSNAs. Te high TSNA content in toombak has been attributed to its high nitrate content, alkaline pH and its manufacture via fermentation.10 In the current study, high levels of two specific forms of TSNAs-NNN and NNK-were noted.Measurements of NNK and NNN show significant variation from one country to another.Levels of NNN and NNK levels in Afzal noted in the current study were lower than those found in traditional American STPs (135.000μg/g and 17.800 μg/g, respectively) or dry weight toombak (3,085 μg/g and 7,870 μg/g, respectively); 4 however, these levels may nevertheless increase in Afzal blends due to the effect of the aforementioned factors that contribute to TSNA formation.Stepanov et al. reported that NNN and NNK levels in Indian khaini and zarda range between 1.740-76.900µg/g and 0.080-28.400µg/g, respectively, while lower ranges are reported for gutkha (0.090-1.090 µg/g and 0.040-0.430 In the current study, levels of NNN in Afzal were much higher than those observed in gutkha, but were lower than those found in khaini and zarda.In terms of NNK levels, the results of the current study were within the same range as those found in khaini and zarda, but were higher than that of gutkha.21 The NNK and NNN levels in toombak (516.000μg/g and 368.000 μg/g, respectively) were markedly higher than dry weight levels of Bangladeshi zarda (3.840 μg/g and 28.600 μg/g, respectively).22 Low levels of NNN were also noted in Indian tobacco products (18.600 μg/g). 23Concentrations of NNN and NNK in American moist snuff have been reported to be 42.600μg/g and 9.950 μg/g, respectively.23 Hearn et al. found that Alaskan Iq'mik had NNN and NNK dry weight levels of 2.700 µg/g and 0.340 µg/g, respectively.24 Handmade Pakistani gutkha has a lower total TSNA content compared with handmade gutkha.23 New STPs in the USA have decreased mean total dry weight TSNA levels (2.610 µg/g) in comparison to traditional STPs (7.420 µg/g); the same is true for specific NNN and NNK levels (2.050 versus 4.410 µg/g and 0.231 versus 1.200 µg/g, respectively).25 Nevertheless, these NNN and NNK levels are still 100-1,000 times higher than nitrosamine levels reported in food and beer.26 Both NNN and NNK are thought to be potent carcinogens.5 Balbo et al. reported a significant association between NNN in drinking water and the development of oesophageal and oral cancers in rats.27 In addition, NNK has been found to increase the risk of lung cancer. 28he NNN and NNK compounds are also known to form haemoglobin adducts in humans as well as in experimental animals.29,30 The use of toombak, which has high NNN and NNK levels as mentioned above, has been associated with cancers of the oral cavity.31,32 Many factors may influence the uptake or absorption of TSNAs in the oral cavity.TSNAs are highly water-soluble; therefore, as the moisture content of a STP increases, the levels of soluble TSNAs may also rise, resulting in enhanced absorption by the oral mucosa.33 The authors of the current study strongly believe that TSNAs are not safe at any level.As a result, it is strongly recommended that existing legislation in Oman regarding the illegal sale of Afzal be more strictly enforced.Furthermore, there is a need for national public health awareness programmes regarding the potential carcinogenic effects of Afzal.
For Swedish snus, maximum permissible levels of carcinogenic substances are laid out by the GOTHIATEK ® standard (Swedish Match, Stockholm, Sweden). 34According to this, the maximum permissible combined NNN and NNK level is 1.000 µg/g. 34etween 1983 and 2004, manufacturers of Swedish snus gradually decreased total TSNA levels to approximately 2.000 µg/g. 21A recent study revealed that the total TSNA level in an unused 1.000 g pouch of snus was approximately 0.830 µg/g, with quantifiable levels of NAT, NNK, NNN and NAB (0.268 µg/g, 0.191 µg/g, 0.344 µg/g and 0.025 µg/g, respectively). 33Therefore, total TSNA levels in the Afzal sample analysed in the current study were fivefold higher than the recommended GOTHIATEK ® limits.Specific TSNA levels in Afzal were six-fold, 2.4fold, 5.2-fold and 8.4-fold higher than the NAT, NNK, NNN and NAB levels, respectively, in snus.As such, TSNA levels in Afzal can be considered alarmingly high.The WHO recommends regulatory limits on the concentrations of selected carcinogens in tobacco products, including TSNAs. 9Their recommendations permit <2.00 µg/g of combined NNN and NNK levels on a dry weight basis. 9Combined levels of these two TSNAs in the studied Afzal sample (2.22 μg/g) exceeded this limit.This may have a negative effect on the health of Afzal users with frequent unmonitored use.Moreover, Afzal users may be exposed to higher levels of TSNAs as a result of varying self-determined pinch sizes; pinch sizes of Afzal are usually greater than the portions provided in packets of the more monitored and controlled Swedish snus. 12,13,33,34t is important to bear in mind that the results obtained by the current study and previous research on Afzal were derived from a single package of Afzal blend. 12,13The findings of future studies may therefore differ due to variations in the additives and manufacturing processes used.

Conclusion
This study revealed the presence of carcinogenic TSNAs in a single random sample of Afzal.Despite being present at relatively low levels in Afzal as compared to other STPs, two of the most potent carcinogenic TSNAs-NNN and NNK-still exceeded the regulatory limits proposed by the WHO.Total TSNA levels were five-fold higher than the limits recommended by the GOTHIATEK ® standard.These findings indicate that Afzal consumption may pose serious health risks for users.Consequently, existing legislation on the sale and availability of this STP should be enforced more rigorously.Increased public health education regarding the potential carcinogenic effects of Afzal is recommended.

Figure 1 :
Figure 1: Photograph of the reference product* (left) and the common smokeless tobacco Afzal (right) samples.These samples were analysed via high-performance liquid chromatography-tandem mass spectrometry to determine tobacco-specific nitrosamine levels.*American moist snuff product purposely manufactured for laboratory analysis by the Tobacco and Volatiles Branch of the Division of Laboratory Sciences at the Centers for Disease Control and Prevention (Atlanta, Georgia, USA).

Table 1 :
Accuracy, precision, reliability and linearity values for each type of tobacco-specific nitrosamine analysed in the Afzal samples

Table 2 :
Concentrations of tobacco-specific nitrosamine levels in three Afzal samples and one reference product* sample as determined by high-performance liquid chromatography-tandem mass spectrometry

Table 3 :
15mparison of certified reference tobaccospecific nitrosamine levels15with levels determined in the present study in one reference product* sample