Abstract

Tobacco-specific nitrosamines (TSNAs) are carcinogenic and are present in cured tobacco leaves. This study was designed to elucidate the mechanisms of TSNAs formation under warm temperature storage conditions. Results showed that nitrogen oxides (NOx) were produced from nitrate and nitrite in a short period of time under 45°C and then reacted with alkaloids to form TSNAs. Nitrite was more effective than nitrate in promoting TSNAs formation during 45°C storage which may be due to the fact that nitrite can produce a large amount of NOx in comparison with nitrate. Presence of activated carbon effectively inhibited the TSNAs formation because of the adsorption of NOx on the activated carbon. The results indicated that TSNAs are derived from a gas/solid phase nitrosation reaction between NOx and alkaloids. Nitrate and nitrite are major contributors to the formation of TSNAs during warm temperature storage of tobacco.

1. Introduction

Tobacco-specific nitrosamines (TSNAs) are a group of important and toxic components of tobacco and tobacco smoke [1, 2]. TSNAs mainly consist of N-nitrosonornicotine (NNN), N-nitrosoanatabine (NAT), N-nitrosoanabasine (NAB), and 4-(methyl nitrosamino)-1-(3-pyridyl)-1-butanone (NNK). NNN and NNK are strong carcinogens [3, 4]. TSNAs are produced via nitrosation of tobacco alkaloids during the curing and storage of tobacco leaves (Figure 1) [58]. In China, both air-cured and flue-cured tobacco leaves are generally stored for approximately 18 months in warehouses before being processed for cigarette production in order to reduce the unfavorable smells [9].

Many studies reported about the factors that influence TSNAs formation during air-curing [7, 1012]. Cui [10] found that TSNAs levels in the leaf lamina and midrib increased substantially during the fourth to seventh weeks of air-curing stage. During this stage nitrate is reduced to nitrite via microbial activity, and the resulting nitrite, in turn, is involved in nitrosation reactions with the naturally existing alkaloids during air-curing, then leading to the formation of TSNAs (Figure 2) [11].

However, to our knowledge there are few reports available on the mechanisms of TSNAs formation during storage stage. TSNAs levels may increase several fold in comparison to the levels in freshly air-cured leaves [13, 14], but the mechanisms of TSNAs formation during storage are not clear. It has been reported that, as the storage temperature increased, TSNAs and nitrite contents increased, and the most rapid increase in TSNAs occurred during the warm temperature season [15, 16]. The interactions between temperature and abundant nitrate in cured tobacco leaf could be responsible for TSNAs formation during storage. Treatment of the tobaccos with streptomycin and rifampicin did not inhibit nitrosamine formation during storage indicating that TSNAs formation was not influenced by microorganisms during the storage process (Figure 2) [14]. How the nitrate and nitrite trigger the increase in TSNAs during warm temperature storage is still not clear (Figure 2). The objectives of this study were to elucidate the mechanisms of TSNAs formation during the storage time of cured tobacco leaf and to verify the hypothesis that nitrogen oxides produced from nitrate and nitrite in tobacco are responsible for the formation of TSNAs during storage.

2. Experiments

2.1. Plant Materials

Tobacco samples were grown in 2015 and cured locally. Leaves from the middle stalk positions were collected. Flue-cured tobacco (N. tabacum cv. “Hongda”) was from Midu county, Yunnan province. Leaf samples were prepared by removing the stems, cut to strips, mixed thoroughly, sealed in plastic bags, and then stored at 4°C for the tests. The moisture content of tobacco was 12%.

2.2. Methods
2.2.1. Direct Addition of Nitrate and Nitrite to Tobacco

The treatments included four levels of NaNO3 (10, 20, 30, and 40 mg/mL) and NaNO2 (5, 10, 15, and 20 mg/mL), which were from 10 mL solution each, equal to 73, 146, 219, and 292 mg of and 33, 67, 100, and 133 mg of . The solutions were sprayed onto each of eight 20 g flue-cured tobacco samples. The samples were then placed to an ambient environment for air-drying. The moisture content of tobacco samples after air-drying was 13%. Then the samples were stored in an airtight vacuum desiccator (15 cm diameter) at 45°C with relative humidity of 70% for 15 d. The sample sprayed with the same volume of deionized water was used as the control. The moisture content of tobacco samples after treatment was 11.8%.

2.2.2. Separating the Tobacco and Nitrate/Nitrite Sources in a Closed System under 45°C Storage Conditions

A vacuum desiccator with a porcelain plate was used to form a closed system in which tobacco cuts and added nitrate and nitrite could be separated during storage. Five mL of each aqueous solution of NH4NO3, KNO3, NaNO3, and NaNO2 at 1 mol/L (equal to 310 mg of and 230 mg of ) and 5 mL volume of deionized water (control) were sprayed onto the medical gauze pads (5 cm × 5 cm). After air-drying, the sprayed gauzes were placed in the bottom of the vacuum desiccator together with each 20 g sample of flue-cured tobacco. The vacuum desiccators were then tightly closed and sealed with petroleum jelly. The containers were then placed into a chamber at a temperature of 45°C and a relative humidity of 70% and stored for 15 d. The moisture content of tobacco samples after 45°C treatments was 11.1%.

2.2.3. Effect of Indirect Addition of Nitrate and Nitrite on Nitrogen Oxides Formation in a Closed System with Flue-Cured Tobacco

Eight treatments in this experiment were divided to two groups: Group I: (1) tobacco; (2) tobacco and NaNO3; (3) tobacco and NaNO3 + 2.0 g activated carbon (AC); (4) tobacco and NaNO3 + 10.0 g AC and Group II: (1) tobacco; (2) tobacco and NaNO2; (3) tobacco and NaNO2 + 2.0 g AC; (4) tobacco and NaNO2 + 10.0 g AC. The amount of added NaNO3 or NaNO2 was 0.8 g and 0.4 g for treatments 2 to 4 (equal to 583 mg of and 266 mg of ) in each group. The weight ratios of tobacco, , and AC were 68 : 1 : 3 and 68 : 1 : 17, respectively. And the weight ratios of tobacco, , and AC were 150 : 1 : 7 and 150 : 1 : 37, respectively. NaNO3 and NaNO2 were dissolved in 5 mL deionized water and then sprayed onto the gauze pads. After air-drying, the gauze pads were placed on the bottom of the vacuum desiccator with 40 g sample of flue-cured tobacco leaf in each desiccator. AC sample was made from wood with the particle size being 75 μm, produced by Zhengzhou Tianhe Water Purification Material Co. Ltd. in Henan province. The surface area was 901.7 m2/g, and the total pore volume was 0.518 cm3/g with the micropore, mesopore, and macroporous volume being 0.372 cm3/g, 0.129 cm3/g, and 0.016 cm3/g, respectively. AC was activated in 100°C oven for 30 min, then was wrapped in a medical gauze, and suspended under the porcelain plate of the vacuum desiccator for the relevant treatments. All vacuum desiccators were tightly closed, sealed, and placed into a chamber at a temperature of 45°C and a relative humidity of 70% for 24 h.

The AC used in the above experiments were taken out and then were placed in the new airtight desiccator, respectively. Two grams of AC was put into another container as control. All airtight desiccators were placed into a chamber at a temperature of 60°C and a relative humidity of 70% for 15 min and 90 min.

2.2.4. Effects of Activated Carbon (AC) on TSNAs Formation in Flue-Cured Tobacco in Response to Nitrate and Nitrite Addition

Twelve treatments in this experiment were divided into two groups: Group I: (1) tobacco (4°C); (2) tobacco (warm temperature control 45°C); (3) tobacco and NaNO3 separately; (4) tobacco and NaNO3 + 1.0 g AC separately; (5) tobacco and NaNO3 + 5.0 g AC separately; and (6) tobacco and NaNO3 + 10.0 g AC separately and Group II: (1) tobacco (4°C); (2) tobacco (warm temperature control 45°C); (3) tobacco and NaNO2 separately; (4) tobacco and NaNO2 + 1.0 g AC separately; (5) tobacco and NaNO2 + 5.0 g AC separately; and (6) tobacco and NaNO2 + 10.0 g AC separately. The amount of NaNO3 or NaNO2 added to each sample of treatments 3 to 6 was 0.3 g (equal to 219 mg of and 200 mg of ). The weight ratios of tobacco, , and AC were 91 : 1 : 4, 91 : 1 : 22, and 91 : 1 : 45, respectively. And the weight ratios of tobacco, , and AC were 100 : 1 : 5, 100 : 1 : 25, and 100 : 1 : 50, respectively. NaNO3/NaNO2 were dissolved in 5 mL deionized water and then were sprayed onto the gauze pads. After air-curing, the gauzes were placed on the bottom of the vacuum desiccators with 20 g sample of flue-cured tobacco leaf in each desiccator. All treatments were performed in the vacuum desiccator as described above. The activated AC also was used as described above. The vacuum desiccators were put into a chamber with a constant temperature of 45°C and a relative humidity of 70%, respectively. The moisture content of cured tobacco samples after treatment was 11.6%.

2.3. Chemical Analyses
2.3.1. TSNAs Measurements

In each experiment, tobacco samples were lyophilized, ground to powder, sieved through a 0.25 mm screen, and then measured for the content of NNN, NNK, NAT, and NAB. TSNAs contents were determined at the Beijing Cigarette Factory according to the method of SPE-LC-MS/MS [12, 17, 18].

2.3.2. Nitrate, Nitrite, and Alkaloid Measurements

NO3-N and NO2-N were quantified according to the method of Crutchfield and Grove [19]. The individual alkaloids were analyzed with a gas chromatograph as described by Jack and Bush [20]. Methyl tert-butyl ether was used as the extraction solvent with N-hexadecane as the internal standard.

2.3.3. Nitrogen Oxides Analysis

The first step was diluting the standard gas with a dynamic gas calibrator (Model 146 i, Thermo Scientific, USA EPA) to give a concentration within the operational range of the instrument. The high purity nitric oxide (NO) and nitrogen dioxide (NO2) standards in N2 (component content: 69.8 ppm, gas sample number: L120911099, National Institute of Metrology/National Standard Material Research Center, Beijing, China) were configured into nitrogen oxides gas of a low concentration (1 ppm) by 146i calibrator. It was necessary to modify the original procedure by configuring the standard gas again if the levels of NOx exceeded the limit of detection. The NO and NOx in the air were defaulted to the zero point by the instrument.

After experiment, the airtight vacuum desiccators were taken out and then connected with a vacuum pump (DOA-P504-BN, GAST Manufacturing, A Unit of Idex Corporation, MI, USA) to extract the gas into a gas collecting bag. After 25 s, the gas bag was pulled out and connected with the NO-NO2-NOX analyzer (Model 42i, Thermo Scientific, USA EPA, Reference method RFNA-1289-074) for NOx analysis by chemiluminescence detection.

2.4. Statistical Analyses

Analysis of variance (ANOVA) and least significant difference (LSD) of TSNAs and NOx values were performed at the 0.05 level of significance. Data were statistically analyzed with SPSS 20.0. Figures were drawn with Origin 8.5. All treatments were randomly designed in triplicate.

3. Results and Discussion

3.1. Effect of Nitrate Concentration on TSNAs Formation in Flue-Cured Tobacco

The mean contents of nicotine, nornicotine, anabasine, and anatabine of tobacco used in this experiment were 19.8, 1.2, 0.3, and 1.5 mg/g, respectively, and the NO3-N and NO2-N content were correspondingly 119 μg/g and 10 μg/g. With the increasing amount of added, both individual and total TSNAs contents increased significantly (Figure 3). After the addition of 0.4 g NaNO3 (equal to 292 mg of ), the NAT and total TSNAs contents increased by 340% and 311%, respectively, in comparison with the control.

No significant change of NNN was observed in the tobacco samples when less than 0.2 g of NaNO3 (146 mg of ) was added. However, when the amount of added increased to 219 mg (0.3 g NaNO3), the NNN content was approximately double that of 0.2 g NaNO3 addition. Although the NAB content was the lowest of the four individual TSNAs, a significant increase in NAB content occurred as a result of the addition of NaNO3 reached to 0.2 g. The NNK content also increased significantly as nitrate addition increased.

3.2. Effect of Nitrite Concentration on TSNAs Formation in Flue-Cured Tobacco

Table 1 shows that as the concentration of added increased, the individual and total TSNAs contents showed rapid and dramatic increases, and the relative increments were greater for the samples with greater amounts of added . When tobacco samples were treated with 133 mg of (0.2 g of NaNO2), total TSNAs content was 117 μg/g, at a 207% increment compared with the control (0.56 μg/g). For the individual TSNA, levels of NNN, NNK, NAT, and NAB showed 257-, 38-, 203-, and 110-fold increases, respectively. The results showed that was more effective in promoting TSNAs formation in the cured leaf than during warm temperature storage.

The higher TSNAs levels in burley tobacco are partly due to the relatively higher levels of TSNAs precursors, such as alkaloids and oxide of nitrogen, that are present in the leaf tissue [2123]. It is reported that nitrite which is derived from the bacterially mediated reduction of nitrate is considered to be the limiting factor in TSNAs formation in air-cured tobacco during air-curing [6, 24]. In this research, the addition of nitrate and nitrite in flue-cured tobacco to the levels which are equivalent to those in burley tobacco can increase the TSNAs concentration comparable to burley tobacco especially coupled with warm temperature. Results indicated that the great amount of nitrosating species available is a major contributor to the formation of TSNAs observed in cured tobacco during storage.

3.3. Effects of Indirect Addition of Nitrate/Nitrite on TSNA Formation in Flue-Cured Tobacco during Warm Temperature Storage

TSNAs contents increased as the storage temperature increased [15, 16], and the abundance of nitrite and nitrate could be a major contributor. To clarify how nitrate or nitrite affects the formation of TSNAs under warm temperature, indirect addition of nitrate or nitrite experiments was carried out.

After tobacco leaf treated with 1 mol/L NaNO2, total TSNAs content increased almost by 54 times compared with that in the control sample (Figure 4). When tobacco leaf is placed separately with gauze pad containing 310 mg nitrate, obvious increases were observed both in individual and in total TSNA content compared with the control. Data presented here suggested that even though the tobacco sample is placed separately with nitrate or nitrite, TSNA content in tobacco also increased. The formation of TSNAs in storage triggered by nitrate and nitrite is likely a gas phase reaction.

3.4. Effect of Indirect Addition of Nitrate/Nitrite on NOx Formation in a Closed System with Flue-Cured Tobacco

Flue-cured tobacco leaves could generate trace concentrations of NOx under 45°C after 24 h treatment (Figures 5 and 6). As a main component, NO accounted for more than 95% of the NOx produced from the tobacco sample. As 0.8 g of NaNO3 (equal to 583 mg of nitrate) was added in the vacuum desiccator, the concentration of NO and NOx increased significantly (); NOx reached 0.6 ppm (Figure 5(c)). When 0.4 g of NaNO2 (equal to 266 mg of nitrite) was added, NOx concentrations in the desiccator dramatically increased to 8.3 ppm, which was 24.6-fold greater than the control (Figure 6(c)).

It is interesting that when 2 g of AC was added to the system (the weight ratio of tobacco, , and AC is 68 : 1 : 3; the weight ratio of tobacco, , and AC is 150 : 1 : 7), a 97% decrease of NOx level was observed in comparison to the added nitrate treatment (Figure 5(c)), and almost 7.9 ppm of NOx were adsorbed by AC (Figure 6(c)). Meanwhile, similar adsorption effects were observed for NO. The concentration of NO and NOx decreased significantly () as the addition of AC increased, indicating strong adsorption for NOx generated from nitrate and nitrite by AC.

Table 2 presents the desorption rate of NOx from the AC samples which were used for the adsorption experiments and were significantly higher than those from control group. The desorption rates of NOx and NO were obviously increased with the prolonging of treating time and the increase of NOx adsorption observed in Figures 5 and 6. Results indicated that NOx were indeed adsorbed by AC. As the weight ratio of tobacco, , and AC is 150 : 1 : 7, NOx decreased by 7.9 ppm (Figure 6(c)); at the same time 0.632 ppm of NOx was desorbed by the AC after treatment at 60°C for 90 min. Since the desorption rate of NOx in AC has close relationship with the temperature and time [25], further research is needed to determine the effect of desorbing temperature and time on the desorption rate of NOx adsorbed by AC.

During flue-curing process, direct-fired systems allow combustion products, specifically NOx, to mix with the air and expose the green tobacco leaves to these gases [26]. The previous research showed that TSNAs may be formed by nitrosation via the elevated levels of NOx in the air surrounding the leaves during the curing process [27]. However, there were few reports about the relationship between NOx and TSNAs during storage process. Data in Figure 4 suggested that TSNAs in storage is likely a gas phase reaction. This result showed that NOx can be generated from NO2-N and NO3-N under 45°C after 24 h. Nitrite was more effective than nitrate in the production of NOx which in turn would probably promote the formation of TSNAs.

3.5. Effect of AC on TSNAs Formation in Flue-Cured Tobacco in Response to Nitrate and Nitrite Added

Having a very porous structure and special surface properties, AC has been used to trap TSNAs in tobacco solution [28, 29]. Lin et al. [30] reported that AC made from coconut shells exhibited a high affinity for TSNAs. The impregnated sorbent ZnAC can remove 73% of the TSNAs in solution, offering a cost-effective candidate for industrial applications [29].

As shown in Figures 7(e) and 8(e), total TSNAs content in tobacco incubated separately with (treatment 3) at 45°C increased markedly to 25 μg/g, a 59-fold increase over the control (treatment 2). By treating tobacco with , however, total TSNAs increased to 1 μg/g, also significantly higher than the control. The huge difference of TSNAs content between tobacco treated with and could be attributed to the high concentration of NOx which were emitted from under the same condition.

However, after adding 1 g of AC to the separating system of tobacco and , the weight ratio of tobacco, , and AC was 100 : 1 : 5, both individual and total TSNAs contents of tobacco decreased markedly, with total TSNAs content decreasing to 13 μg/g, 48% lower than in treatment 3, and NNN, NAT, NAB, and NNK decreasing by 47.9%, 47.6%, 67.3% and 41%, respectively. When the addition of AC increased to 5 g, the weight ratio of tobacco, and AC reached to 100 : 1 : 25, total TSNAs levels further decreased to 1.2 μg/g, indicating a crucial role of AC in inhibiting TSNAs formation by adsorbing the NOx generated from NO2-N and NO3-N. The results also showed no significant differences in reducing TSNAs levels by increasing AC usage greater than 5 g (Figures 7 and 8). The maximum inhibitory effect of AC on TSNAs formation occurred at an AC/ ratio of 25/1. The above results indicated that AC may have the potential to be used as adsorbent agent to reduce TSNAs formation during tobacco leaf storage. Our results strongly support the theory that TSNAs formation in storage is a gas phase reaction, and the NOx which is produced from nitrate and nitrite could intermediate the TSNAs formation during warm temperature postcured storage of tobacco. Since NOx can be produced from nitrate and nitrite under warm temperature in a short period, it can readily react with alkaloids to form TSNAs. This result also explained why the most rapid increase in TSNAs occurred during the warm temperature season [16].

Decreasing NOx level by AC adsorption significantly reduced TSNAs formation of tobacco which indicated that the removal of NOx from storage environment could be an effective way to inhibit TSNAs formation in storing tobacco leaf. Therefore, controlling the storage environment and scavenging gaseous nitrosation agents would be crucial to reduce or inhibit TSNAs formation during leaf storage.

4. Conclusion

The results proved that TSNAs are derived from a gas/solid phase nitrosation reaction between NOx and alkaloids during storage. Nitrogen oxides produced from nitrate and nitrite are responsible for the formation of TSNAs during storage under warm temperature. Presence of activated carbon in the tobacco storage containers effectively inhibited the TSNAs formation due to the adsorption of NOx on the activated carbon.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this paper.

Authors’ Contributions

Jun Wang and Huijuan Yang contributed equally to this work and should be considered co-first authors.