Identifying the Geographical Origin of Tobacco Leaf by Strontium and Lead Isotopic with Mineral Elemental Fingerprint

The primary aim of this paper was to identifying the geographical origin of tobacco leaves based on stable isotopic and mineral elemental fingerprint. We collected eighty-one tobacco leaf samples from Argentina, Brazil, Zimbabwe, the U.S., Zambia, and China. And nine mineral element contents and four strontium and lead isotope ratios of the tobacco leaves were determined by thermal ionization mass spectrometry (TIMS) and inductively coupled plasma mass spectrometry (ICP-MS). After variance and stepwise discriminant analysis, the discriminant functions of the tobacco leaf’s geographical origin were established. The results indicate that: (1) the contents of six mineral elements including Cu, Zn, Cr, Ni, Cd, and Pb, together with four strontium and lead isotope ratios containing 87 Sr/ 86 Sr, 208 Pb/ 204 Pb, 207 Pb/ 204 Pb, and 206 Pb/ 204 Pb, were significantly different among six countries. (2) Different countries presented some characteristic mineral elemental and isotopic fingerprint. The even contents of mineral elements from Zambian tobacco leaf were much lower than the other countries, among which four elements consisting of Zn, Cr, As, and Cd were not detected. The three average lead isotope ratios including 208 Pb/ 204 Pb, 207 Pb/ 204 Pb, and 206 Pb/ 204 Pb from Zimbabwe tobacco leaves were far higher than the other countries, and the range of which was unoverlapped. (3) The effective identification of the geographical origin of tobacco leaf was accomplished by Fisher stepwise discriminant analysis and the characteristic tracing elements consisted of Cu, Zn, Cr, Ni, Cd, Pb, 87 Sr/ 86 Sr, 207 Pb/ 204 Pb, and 206 Pb/ 204 Pb. Based on the established discriminant functions, the original and cross-validation accuracy towards different geographical origins of tobacco leaves were 98.8% and 95.1%, respectively. The study shows that the strontium and lead isotopic with mineral elemental fingerprints is a potential effective method to identify the geographical origin of tobacco leaves from different countries.


Introduction
With the globalization of the economy, agricultural products' transnational and cross-regional circulation are increasingly frequent, and the quality and safety of agricultural products have become a common concern of the world today [1,2]. e geographical origin of agricultural products is one of the important quality and safety elements. Even more, some laws and regulations expressly require the labeling of the geographical origin.
e European Union regulations allow the application of the following geographical indications to a food product: protected designation of origin (PDO), protected geographical indication (PGI), and traditional specialties guaranteed (TSG) [3,4].
Many studies have been conducted to investigate the geographical origin of agricultural products, mainly through DNA, near-infrared spectroscopy, stable isotopes, mineral elements, feature organic content, etc., to nd speci c indicators that can characterize regional information [3,4]. Among them, stable isotope ngerprints are a natural label of crops, closely related to their growing environment and location, and do not change in the process. ey can provide independent and unchangeable identi cation information for its origin traceability. erefore, stable isotope detection technology combined with mineral element content analysis is the most effective method to trace the origin of agricultural products [5]. At present, this technology has been widely used in the origin traceability of wine [6], beer [7], spring water [8], tea [9], coffee [10], rice [11], lamb [12], beef [13], poultry meat [14], edible vinegar [15], wheat [16], and cocoa [17]. It has provided strong technical support for the protection of geographical indications of world-renowned agricultural products and processed products.
Tobacco leaves are important economic crops in the world, and their quality is closely related to genetic factors, cultivation measures, modulation techniques, origin soil and climatic conditions, etc., among which the effect of the production areas such as soil and climatic conditions on the quality of tobacco leaves is highly significant [18]. Due to the high taxes and fees on tobacco, smuggling and illegal sales are very active. ese activities violate legal regulations and lead to severe financial and tax losses. Taking the European Union as an example, illegal sales cause tax and customs losses of 10 billion every year. Tobacco leaves are the foundation of the cigarette industry. e quality and authenticity of the origin directly affect the quality and authenticity of cigarettes. erefore, the unauthentic origin of tobacco leaves is one of the primary forms of cigarette counterfeiting [19].
In the world, China is the world's largest producer and importer of tobacco leaves, while Zimbabwe, Brazil, and the United States are the world's largest tobacco leaf exporters. In international trade, counterfeit tobacco often appears, which seriously affects the financial revenue of the government and the legitimate rights and interests of consumers.
ere are many researches about tobacco origin tracing, mainly based on volatile components of tobacco leaves [20], metabolites [21], nitrogen-containing compounds [22], pollen content [23], X-ray fluorescence [24], near-infrared spectroscopy [25], and color fractal [26], but these methods are easily affected by factors such as fertilization type, redrying process, picking period, and storage conditions. Isotope ratio characteristics generally do not change due to chemical and physical changes experienced, and they are an effective indicator for judging the origin of animal and plant products [27]. Because of this, many researchers use stable isotope technology to carry out exploratory research on the traceability of tobacco leaves or cigarettes. For example, the position specific isotope analysis (PSIA) was used to study the intrinsic relationship between the hydrogen isotope ratio in nicotine and the origin of tobacco leaves [28], and/or use the compound specific isotope analysis (CSIA) to explore the ratio of carbon, hydrogen, and oxygen isotopes in nicotine to trace the tobacco [29], and/or by bulk stable isotope analysis (BSIA) to study lead isotope ratio to trace tobacco in different provinces of China [30]. e aims of this study were to develop reliable analysis technology that could establish the geographical origin of tobacco leaves based on their trace mineral elemental "fingerprint" and the stable isotope ratios of strontium and lead. Here we present the findings from tobacco samples originating from Argentina, Brazil, Zimbabwe, the U.S., Zambia, and China.

Sample Collection and Preparation.
e 81 tobacco leaf samples were primary-cured tobacco and red-cured sheet tobacco collected from 6 different countries in 2013-2014, including 12 from Argentina, 10 from Brazil, 17 from Zimbabwe, 6 from the United States, 8 from Zambia, and 28 from China.
Samples thus prepared were ground to obtain the powder by a Cyclotec sample mill (SMF2002, SUPOR, China).

Elemental Analysis.
e elemental analysis is based on YC/T 380-2010 "Tobacco and Tobacco products chrome, nickel, arsenic, selenium, cadmium, lead assay inductive coupling plasma mass spectrometry" [31]. A microwave digestion system (MW3000, Anton Paar, Austria) was used to digest the pretreated tobacco samples, and the digestion process was as follows: a certain amount of samples (0.2∼0.3 g) and 65% HNO 3 solution (5 mL, BV-III, Beijing Institute of Chemical Reagents) and 30% H 2 O 2 solution (2 mL, BV-III, Beijing Institute of Chemical Reagents) were added into a Teflon digestion vessel for digestion. In the 60-min digestion process, the temperature gradually rose to 190°C. Finally, the digested liquid was diluted with ultrapure water to 50 mL and stored in a plastic storage bottle before analysis. e concentrations of nine elements (Cu, Zn, Co, Cr, Ni, As, Se, Cd, and Pb) were determined by inductively coupled plasma mass spectrometry (ICP-MS, ELAN-DRC-e, Perki-nElmer, USA).

Sr and Pb Isotope Analysis.
e measurement of the isotopic ratio of the Sr and Pb is based on GB/T 17672-1999 "Determinations for isotopes of lead, strontium, and neodymium in rock samples" [32]. Pretreatment method: 0.5 g powder was weighed into a quartz beaker, and 65% HNO 3 solution was added to dissolve it overnight. It was dried and then heated at 250°C for 2 h and at 550°C for 4 h in a muffle furnace, after that 1 ml 65% HNO 3 was added to dissolve and then it was transferred to the Teflon vessel and finally evaporated to be tested. Separation and purification of Sr and Pb are completed using the AG50W-X8 cation exchange resin (Pharmaceutical Group Chemical Reagent Co., Ltd. China) and the AG1-X8 anion exchange resin (Bio-Rad Laboratories, UK).

Statistics.
e statistical analysis of the data was used by SPSS 22.0 software (IBM, US). e analysis of variance (ANOVA) was carried out for each element. Firstly, the variance homogeneity test for each variable was carried out to check whether the total variances of each analytical index groups were consistent. en, at the significance level p < 0.05, Duncan's multiple comparison was performed to determine the significant difference between the individual regions when the F value was significant in ANOVA.
e Fisher stepwise discriminant analysis is employed to classify the geographical origin. e robustness of the classification model was evaluated by the original validation and cross-validation tests.

Difference Analysis of Mineral Element Content in Tobacco from Different Countries.
e content of 9 mineral elements in tobacco leaves from Argentina, Brazil, Zimbabwe, the United States, Zambia, and China was analyzed by ANOVA. e measurement and statistical results are shown in Table 1, Figure 1 e contents of Cu, Zn, Ni, Cd, and Pb are relatively higher than the other minerals in all tobacco leaf samples with average values of 7.89, 24.69, 2.78, 1.15, and 1.21 mg/kg, and the contents of Co, Cr, As, and Se have relatively lower average, with values of 0.46, 0.56, 0.35, and 0.19 mg/kg, respectively. Among them, the element contents of Cu, Zn, Cr, Ni, Cd, and Pb were significantly different among the 6 countries (p < 0.001, F value ˃7). In contrast, the element contents of Co, As and Se were not significantly different (p > 0.001).
e contents of mineral elements in tobacco leaf samples from different countries have their own characteristics. e content of Cr in Argentine tobacco leaves is significantly  higher than that of the other 5 countries, with an average of 0.83 mg/kg. e content of Pb in Brazilian tobacco leaves is lower than that of the other 5 countries, with an average of 0.09 mg/kg, and the content of other elements is in medium level. e content of Cu, Co, and Se in Zimbabwe tobacco leaves was significantly higher than those of the other 5 countries, with the average values of 10.52, 0.65, and 0.25 mg/kg, respectively. e content of Zn in American tobacco leaves was the highest, with an average value of 40.15 mg/kg, and the content of Co was the lowest, with an average value of 0.14 mg. It is worth noting that the content of mineral elements in tobacco leaves in Zambia is generally low, and the content of Cu and Ni is significantly lower than that of the other 5 countries, with an average of 0.79 and 0.24 mg/kg, respectively; At the same time, Zn, Cr, As, and Cd are not detected. e contents of Ni, As, Se, Cd, and Pb in tobacco leaves in China were significantly higher than those in the other five countries, with average values of 5.27, 0.75, 0.25, 2.91, and 2.60 mg/kg, respectively.

Characteristics of Sr and Pb Isotopes in Tobacco from Different Countries.
e results of 4 isotopic ratio measurements in different countries are shown in Table 2, and the box plots are shown in Figure 10 -13. e difference between the isotopic ratio between the six countries was significant (p < 0.001, F > 116

Traceability of Tobacco Origin.
From the above analysis, it can be seen that the fingerprints of mineral elements, Sr and Pb isotope ratios in tobacco leaves in 6 countries all show specific characteristics. To further explore the analytical methods of isotope and mineral element fingerprints for traceability of tobacco and to find element indicators that characterize regional characteristics. In this paper, Fisher's Stepwise Discriminant Analysis is used to construct the origin traceability model. 9 mineral elements, 4 Sr and Pb isotopes ratios in the tobacco were introduced into the discriminant model. e essential variables were screened out for modeling after the Fisher's Stepwise Discriminant Analysis, such as group mean equality test, covariance matrix, step statistics, and eigen values tests. e homogeneity testing results of the group's average showed that there were significant differences in 9 indexes of tobacco, such as Cu, Zn, Cr, Ni, Cd, Pb, 87 Sr/ 86 Sr, 207 Pb/ 204 Pb, and 206 Pb/ 204 Pb, among different countries (see Table 1 and 2 for F values differences between groups). Among them, 87 Sr/ 86 Sr, 207 Pb/ 204 Pb, and 206 Pb/ 204 Pb have the most obvious differences, and their F values reach 122.238, 161.295, and 129.181, respectively, which is particularly effective in distinguishing the origin. e 5 discriminant functions formed by these 9 indicators, these functions variances, accounted for 70.0%, 18.5%, 8.9%, 2.2%, and 0.4% of the total variance, respectively, and the cumulative variance accounted for 100% of the total variance. Based on the above, the discriminant function of tobacco origin for 6 countries was finally constructed (see Table 3). In practical application, the Many researches have shown that stable isotope ratios and mineral elements, combined with multivariate statistical analysis have proven to be particularly useful in tracing the geographical origin of the crops [3,[33][34][35]. ese tracers   International Journal of Chemical Engineering reflect the characteristics of the soil, climate, fertilizer, processing, and storage of the crops [1][2][3][4].
In the origin discriminant functions, the absolute value of the coefficient reflects the element weights in the origin determination [3,34]. e larger the absolute value of the coefficient, the greater the impact on the origin determination. From Table 3, the absolute value of the coefficients of 87 Sr/ 86 Sr, 207 Pb/ 204 Pb, and 206 Pb/ 204 Pb are absolutely bigger than the others. is indicates that these three isotope ratios play a key role in the traceability of the  origin. e results are consistent with previous reports. In these researches, 87 Sr/ 86 Sr, 207 Pb/ 204 Pb and 206 Pb/ 204 Pb mainly reflect the elemental profile of the soil and vary from region to region, related to the "terroir" of tobacco [33,35]. As shown in Table 3, the coefficient of Zn and Cu is very small. It indicates that Zn and Cu are not only affected by soil of origin but also related to external impurities in tobacco growing and processing [3,[17][18][19][20]34]. In tobacco growing and processing, Zn is an essential trace element of tobacco. Zinc fertilizer needs to be applied in the process of tobacco planting. And, a Bordeaux mixture prepared with CuSO 4 is commonly used in plant fungicides. e initial verification and a cross-verification method validate the model in identifying tobacco production, as shown in Table 4. In the initial verification results, all tobacco leaves from Argentina, Brazil, Zimbabwe, the United States, and Zambia were correctly classified, and the initial discrimination accuracy was 100%. One Chinese tobacco leaf was misclassified as Argentina, and the initial discrimination accuracy was 96.4%. In the cross-validation results, tobacco leaves from Argentina, Brazil, Zimbabwe, the United States, and Zambia were all correctly classified, and the cross-discrimination accuracy rate was 100%. 4 Chinese tobacco leaves were misclassified (3 from Argentina and 1 from the United States). e cross-discrimination accuracy rate is 85.7%. In conclusion, based on Fisher's stepwise discriminant analysis, the initial determination accuracy rate of tobacco leaf origin is 98.8%. e cross verification accuracy rate is 95.1%, which can realize the origin identification of most tobacco leaves. e first three discriminant function scores of the tobacco leaf samples are plotted as a scatterplot ( Figure 14). It can be intuitively seen that tobacco leaves from different countries have their own spatial distribution characteristics, and Argentina and Brazil have some crossovers.

Conclusions
In this study, the contents of 9 mineral elements and the ratios of 4 strontium and lead isotopes in 81 tobacco leaf    rough Fisher's stepwise discriminant analysis, the effective identification of the origin of tobacco leaves is realized via 9 characteristic traceability indicators (Cu, Zn, Cr, Ni, Cd, Pb, 87 Sr/ 86 Sr, 207 Pb/ 204 Pb, and 206 Pb/ 204 Pb). e initial validation and cross-validation accuracy of the function for tobacco leaf origin were 98.8% and 95.1%, respectively. e combination of Sr and Pb stable isotope ratio and mineral element content can effectively discriminate the origin of tobacco leaves, and the discrimination accuracy is very high.

Data Availability
e original data used to support the findings of this work can be obtained from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.