Analysis of volatility characteristics of five jujube varieties in Xinjiang Province, China, by HS‐SPME‐GC/MS and E‐nose

Abstract In this study, headspace solid‐phase microextraction coupled with gas chromatography‐mass spectrometry (HS‐SPME‐GC/MS) was used to identify individual volatile compounds in five jujube varieties, and E‐nose was used to identify their flavor. The results showed that a total of 45 volatile compounds were detected by GC‐MS in the five varieties, and the proportion of acids was the highest (38.29%–54.95%), followed by that of aldehydes (22.94%–47.93%) and esters (6.33%–26.61%). Moreover, different varieties had obviously different volatile components. E‐nose analysis showed that the R7 and R9 sensors were more sensitive to the aroma of jujube than other sensors. The strong response of R7 sensor was attributed to terpenes (or structurally similar substances) in jujube fruit, such as 1‐penten‐3‐one, 2‐octenal, (E)‐2‐heptanaldehyde, and (E)‐2‐hexenal and that of R9 sensor was attributed to the cyclic volatile components such as benzaldehyde, benzoic acid, and methyl benzoate. The multivariate data analysis (PCA, OPLS‐DA, and HCA) of the results of GC/MS and E‐nose showed that the five varieties could be divided into three groups: (1) Ziziphus jujuba Mill. cv. Huizao (HZ) and Z. jujuba cv. Junzao (JZ). Acids were the main volatile components for this group (accounting for 47.44% and 54.95%, respectively); (2) Z. jujuba cv. Hamidazao (HMDZ). This group had the most abundant volatile components (41), and the concentrations were also the highest (1285.43 µg/kg); (3) Winter jujube 1 (Z. jujuba cv. Dongzao, WJ1) and Winter jujube 2 (Z. jujuba cv. Dongzao, WJ2). The proportion of acids (38.38% and 38.29%) and aldehydes (40.35% and 38.19%) were similar in the two varieties. Therefore, the combination of headspace solid‐phase microextraction coupled with gas chromatography‐mass spectrometry and E‐nose could quickly and accurately identify the volatile components in jujube varieties from macro‐ and microperspectives. This study can provide guidance for the evaluation and distinguishing of jujube varieties and jujube cultivation and processing.


| INTRODUC TI ON
Jujube (Ziziphus jujuba Mill.), a member of Rhamnaceae family, originates from the Yellow River basin in China. It has been cultivated for more than 4000 years . Jujube fruit is favored by many consumers for the abundance of components such as triterpenoids, phenolic acids, flavonoids, and polyphenols that have anti-inflammatory, antiallergic, and antioxidant effects (Cheng et al., 2020;Jiang et al., 2019). Moreover, jujube products, such as fruit wines and jam, are considered healthy foods that are increasingly preferred by consumers around the world (Wojdylo et al., 2016). The aroma of jujube gives a special flavor characteristic to jujube products. However, most researches focus on the preservation and processing of jujube fruit (Cheng et al., 2020), while few studies focus on the volatile components in jujube fruit.
Volatile components are the secondary metabolites of fruit. The release of volatile components brings unique flavor profile for fruit (Janzantti & Monteiro, 2014). Therefore, volatile components can be used to evaluate the fruit quality (Liu, Du, et al., 2019). However, the volatile components may significantly vary in varieties (Barros-Castillo et al., 2021;, and the volatile components of a variety may also vary in production areas (Li et al., 2013;Spizzirri et al., 2019). Therefore, a comprehensive exploration of the volatile components of jujube fruit is essential, which is helpful for the raw material cultivation and selection for fruit processing industry (Wang, Wang, Deng, et al., 2019).
Headspace solid-phase microextraction (HS-SPME) is commonly used to extract the volatile components from fruit using fused silica fibers. Compared with simultaneous distillation extraction and volatile oil extraction, HS-SPME can avoid thermochemical influence on volatile components during the extraction process. Moreover, the headspace solid-phase microextraction coupled with gas chromatography-mass spectrometry (HS-SPME-GC/MS) provides a rapid and accurate tool for qualitative and quantitative analysis of the volatile components in fruit . Galindo et al. (2015) used this method to detect the volatile components in jujube fruit, finding that there were a total of 18 volatile components detected, and the aldehydes and acids such as hexanal, (E)-2-hexanal, benzaldehyde, hexanoic acid, and decanoic acid accounted for 95%.  also used this method to detect the volatile components in jujube fruit, finding that there was significant difference in volatile components among different varieties.
The electronic nose (E-nose) is a bionic equipment with multiple chemical sensors that can accurately evaluate the volatility characteristics of fruit (Yang et al., 2016). The combination of HS-SPME-GC/MS and E-nose can analyze the volatile components of fruit from macroscopic and microscopic perspectives. Currently, this method has been widely used for shelf-life assessment (Wang, Baldwin, et al., 2019;Yang et al., 2020), freshness assessment (Dou et al., 2020;Pennazza et al., 2013), and fruit processing monitoring (Xu et al., 2019;Yang et al., 2016). Chen et al. (2018) used this method to analyze the volatile components in jujube fruit, finding that there were significant differences in the volatility characteristics among different varieties.
Xinjiang Uygur Autonomous Region of China is the main production area of jujube in the world, but the volatile components of different jujube varieties in Xinjiang have not been determined. Therefore, in this study, five jujube varieties mainly cultivated in this area were used as test materials, to identify the volatile components using HS-SPME-GC/MS and E-nose, and the differences in volatile components among different varieties were determined by using multivariate data analysis. This study contributes to the breeding of excellent jujube varieties and the fruit processing industry. to the semired stage (the color of 50% of fruit surface became red) at room temperature of 25 ± 3℃, with relative humidity of 40 ± 5%.

| Sample preparation
Then, the fruit of each variety (1 kg) was peeled, stoned, cut into small pieces (3 mm in thickness), frozen in liquid nitrogen, and stored at −80℃.

| HS-SPME extraction of volatile components
The volatile components of jujube fruit were extracted and determined according to the method proposed by Song et al. (2019), using an aged SPME extraction head (50/30 μm PDMS/CAR/DVB).
Fruit sample (4 g) was equilibrated in a 20 ml sample vial at 50℃ for 30 min, and the temperature of the solid-phase microextraction apparatus was set at 50℃. After that, vials were stirred at 300 rpm and headspace extraction was performed for 30 min.

| Quantification of volatile components by GC/ MS
The extracted volatile components were analyzed with GC/MS (SQ-456-GC-MS, Scion), and a DB-WAX (30.0 m × 250 μm, 0.25 μm) chromatographic column was used. The temperature of the column incubator started at 40℃. After 3 min, the temperature increased to 100℃ at 6℃ /min. Then, it increased to 230℃ at 10℃/min and lasted for 6 min. The temperature of sample inlet was 200℃, and the temperature of detector was 250℃. The carrier gas was N 2 at a flow rate of 0.8 ml/min. The MS cleavage was performed using electron ionization mode with an electron energy of 70 eV. The temperature of ion source was 200℃. The interface temperature was 250℃. The mass was in the range of 33-400 u.

| E-nose analysis of the volatility characteristic of jujube fruit
E-nose analysis was performed with the method of Chen et al. (2018).
Fruit samples were crushed and sieved through a 60 mesh. After that, the powder (3.0 g) was transferred in 15-ml headspace vial, sealed, and equilibrated at 30℃ for 30 min. After equilibration, the samples were placed in an E-nose device equipped with 10 sensors (PEN3, AirSense) for detection, and the responses in 1 min were recorded.

| Data analyses
Data were analyzed using SPSS 18.0 (IBM). The concentrations of volatile components were expressed as mean ±standard (standard deviation). Duncan's test was used to detect the significant difference in the data of GC/MS and E-nose analysis for different varieties. GC/MS and E-nose data were analyzed using PCA, OPLS-DA, and HCA, and plotting was performed using Origin 2018 (Origin Lab Co.).
In this study, 36, 41, 37, 33, and 31 kinds of volatile components were identified in HZ, HMDZ, JZ, WJ1, and WJ2, respectively, of which 24 kinds of components were found in the five varieties ( Figure 1 and Table 1 components in jujube fruit, accounting for more than 50%. It seems that there are more acids in jujube fruit cultivated in Xinjiang, China, which may be attributed to the growth environment.

| Multivariate data analysis by GC/MS
PCA analysis can reduce the dimension of the original data and retain the variability to separate samples (Khalil et al., 2017).   To verify the results, HCA analysis was performed using the squared Euclidean distance method (Wang, Wang, Deng, et al., 2019). The heat map shows the distribution of main volatile components of jujube fruit ( Figure 2c). Similar to the distribution in OPLS-DA analysis, the quantitative results were also divided into three groups by hierarchical clustering: (1) HMDZ, (2) HZ and JZ, and (3) WJ1 and WJ2, which verified the accuracy of previous multivariate analysis.

| E-nose analysis
The E-nose is very sensitive to fruit aroma. A slight change in the aroma can be detected by the sensor (Yang et al., 2016). The E-nose equipped with 10 metal oxide semiconductors (sensors) was used to analyze fruit aroma in this study. Among the sensors, only R2 (broadly sensitive), R6 (sensitive to methane), R7 (sensitive to terpenes and sulfides), R8 (sensitive to alcohols, aldehydes, and ketones), and R9 (sensitive to aromatics and organic sulfides) responded. The response value of R3 (ammonia, sensitive to aromatic components), R4 (mainly sensitive to hydrogen), R5 (sensitive to alkanes, aromatics, and small polar compounds), and R10 (sensitive to high concentration aliphatic compounds) were about 1 (Figure 3), which meant almost no response.
The response value of R7, R9, R6, and R2 was 4.73-44.45, 4.25-29.36, 2.25-15.07, and 2.32-12.18, respectively, which were similar to the results of Chen et al. (2018) and Song et al. (2019). According to the response value, HMDZ had the strongest aroma. The strong response of R7 and R9 sensors indicates that HMDZ contains terpenes and sulfides.
Moreover, the response of R1 (benzene and structural analogues) was only found in HMDZ, which might be caused by the aromatic com-

| Combining analysis of HS-SPME-GC/MS and E-nose
The comparison of the results of GC/MS and E-nose (Figures 2 and 4) showed that the classification results were consistent: (1) HMDZ, (2) HZ and JZ, and (3) WJ1 and WJ2. Among the sensors, R7 and R9 had stronger responses, indicating that jujube fruit might contain a variety of terpenes and aromatic components (or structural analogues). The responding volatile components were identified by GC/MS. For example, 1-penten-3-one, 2-octenal, (E)-2-heptanal, and (E)-2-hexenal (terpenes and structural analogues) might cause the response of R7 . The response of R9 to HMDZ was the strongest, followed by that to JZ and HZ. According to the quantitative results, it was speculated that the response of R9 (sensitive to aromatic substances) might be due to the volatile components such as benzaldehyde, benzoic acid, and methyl benzoate, and the high response value to HMDZ might be due to benzaldehyde (239.2583 μg/kg). In addition, it was noticed that R6 (broadly sensitive to methane) also responded, which might be due to the volatile components containing methyl such as methyl valerate, methyl hexanoate, methyl decanoate, ethyl benzoate, methyl laurate, acetoin, and 3-octanone.

| CON CLUS ION
In this study, a total of 45 kinds of volatile components were identified from the fruit of five jujube varieties (JZ, HZ, HMDZ, WJ1, and WJ2) cultivated in Xinjiang, China, and quantified by HS-SPME-GC/ MS and E-nose. HMDZ has the most abundant volatile components, and aldehydes and acids are the main volatile components. WJ1 and WJ2 have few differences in volatile components, which may be due to that they belongs to the same variety. Moreover, their volatile components are similar to those of HMDZ, and acids and aldehydes are the main components. In HZ and JZ, acids are the main volatile components, and the concentration is almost twice that of aldehydes. This study provides guidance for the selection of raw materials and jujube fruit processing. It also verifies that HS-SPME-GC/MS and E-nose technology can quickly and accurately identify the flavor differences among jujube varieties.

CO N FLI C T S O F I NTE R E S T
The authors have no affiliation with any organization with a direct or indirect financial interest in the subject matter discussed in the manuscript.