Characterization of volatile compounds in pickled and dried mustard (Brassica juncea, Coss.) using optimal HS-SPME-GC-MS

ABSTRACT

Pickled and dried mustard (PDM; Brassica juncea, Coss) is a traditional fermented vegetable product that is widely consumed as a food ingredient with aromatic flavour in China. Here, we aimed to determine the volatile compounds in PDM by headspace solid-phase micro-extraction combined with gas chromatography-mass spectrometry (HS-SPME-GC-MS). The results showed that a total of 41 volatile compounds were identified and quantified in PDM by optimized HS-SPME-GC-MS. Approximately 15 compounds were considered to be the dominant volatile compounds in PDM with relatively high concentrations. These findings make the volatile composition characteristics in PDM clear for the first time.

RESUMEN

En China, la mostaza encurtida y deshidratada (PDM; Brassica juncea, Coss) constituye un producto vegetal tradicional fermentado, ampliamente consumido, cuyo sabor es aromático. El presente artículo da cuenta de un estudio que se propuso determinar el perfil de compuestos volátiles contenidos por la PDM mediante el uso de cromatografía de gases, espectrometría de masas, microextracción en fase sólida y la técnica de espacio de cabeza (HS-SPME-GC-MS). Empleando la técnica HS-SPME-GC-MS, los resultados de los análisis dan cuenta de la presencia de un total de 41 compuestos volátiles que pudieron ser identificados y cuantificados en la PDM. Se considera que aproximadamente 15 de estos compuestos son los compuestos volátiles dominantes en la PDM, comprobándose que se encuentran en concentraciones relativamente elevadas. Estos hallazgos dejan en claro, por primera vez, las características de la composición volátil de la PDM.

KEYWORDS:

• Pickled and dried mustard
• Brassica juncea, Coss
• headspace solid-phase micro-extraction (HS-SPME)
• gas chromatography-mass spectrometry (GC-MS)
• volatile compound

PALABRAS CLAVE:

• Mostaza encurtida y deshidratada
• Brassica juncea, Coss
• Espacio de cabeza microextracción en fase sólida (HS-SPME)
• Cromatografía de gases-Espectrometría de masas (GC-MS)
• Compuesto volátil

1. Introduction

Pickled and dried mustard (PDM, also called Meigancai in Chinese or mandarine) is a traditional fermented vegetable product made of Brassica juncea, Coss, and it is widely consumed as a food ingredient with aromatic flavour in China (Li et al., 2012). PDM can be added to pork, chicken, fish, and cowpeas during cooking by various methods, including steaming and stir-frying. Particularly, steamed pork with PDM is nutritious and delicious and is a famous Chinese dish throughout the world. Traditionally, PDM is homemade in the beginning of every winter or spring, and the detailed procedure was described by Huang, Huang, and Feng (2012). In short, fresh mustard is cut into pieces, pickled with salt, and then dried in the sun or with the help of other heat sources, producing the PDM products. Moreover, the products can be stored at room temperature for two years (Huang et al., 2012).

The main quality attributes to PDM are due to its typical flavour and taste, while its aroma resulting from its volatile compounds could initially influence the consumers’ acceptance. Zhao, Tang, and Ding (2007) analysed volatile compounds in fresh mustard and their pickles with different pickling times, and Xu, Hu, Wu, and Deng (2013) compared volatile compounds in eight kinds of high salt pickled mustards. They all found that glucosinolates enzymatically decomposed compounds accounted for the major proportion in the volatile compounds. In contrast, PDM is a kind of product that is not only pickled but also dried afterwards from fresh mustard. Furthermore, drying and dewatering affect the quality of the food products, such as texture, taste, aroma, and nutritional properties (Yang et al., 2016). Hence, the characteristics of volatile compounds in PDM may also be different. Luo, Yu, Hu, and Qiu (2016) analysed the flavour components of pickled mustard dried by different methods, and they found that freeze drying method could protect flavour substances in pickled mustard better than using microwave drying. However, to the best of our knowledge, no other reports have focused on the volatile compounds in PDM.

In recent years, headspace solid-phase micro-extraction (HS-SPME) and gas chromatography-mass spectrometry (GC-MS) have been commonly used in the literature for the extraction of volatile compounds and for further quantification of the aroma produced, respectively (Cuevas-Glory, Pino, Santiago, & Sauri-Duch, 2007; Panighel & Flamini, 2014; Xu et al., 2016). In this way, Yao et al. (2015) evaluated the volatile profile of Sichuan dongcai, a traditional salted vegetable, during fermentation with a salting time from one to three years, Yang et al. (2016) found that the flavour of Flammulina velutipes significantly changed during the hot air drying process, Tian, Li, Qin, Yu, and Ma (2015) reported that it was an effective identification tool for the quality discrimination of beef seasonings, and Cheng et al. (2015) successfully characterized the aroma-active profiles of fruits from three different bayberry cultivars, etc. However, there are some factors that could influence the HS-SPME efficiency and further affect the results of GC-MS analysis (Fan, Li, Xue, Hou, & Xue, 2017).

Accordingly, in this study, we aimed to determine and identify the volatile compounds of PDM by HS-SPME-GC-MS. We also optimized the HS-SPME condition for better GC-MS analysis of the volatile compounds in PDM. The findings of this study could provide consumers some guidance for selecting PDM products and may lay a theoretical foundation for future studies of PDM, further expanding its applications.

2. Materials and methods

2.1. Samples

In this study, five commercially available brands of PDM samples (Figure 1) were analysed, and all of them were collected in Hangzhou, Zhejiang Province, China, in June 2016: XH, from Shaoxing Xianheng Food Co., Ltd., Shaoxing, China; GGX, from Jinyun County Quanyou Food Co., Ltd., Jinhua, China; GHW, from Hangzhou Guanhuawang Food Co., Ltd, Tonglu, China; JTA, from family workshop A, Hangzhou, China; JTB, from family workshop B, Hangzhou, China. These PDM samples were all made of fresh mustard (Brssica juncea, Coss.; Jiutoujie in Chinese) in the spring of that year, but the quality (for example, different colours of the five samples shown in Figure 1) of the PDM produced by different manufacturers varied.

Figure 1. Five commercially available brands of PDM samples analysed in the study.

Figura 1. Muestras de cinco marcas de PDM disponibles a nivel comercial analizadas en el estudio.

2.2. Chemicals

A mixture of n-alkanes (C8-C20; Sigma-Aldrich, St. Louis, MO, USA) was used for the retention index (RI) analysis, and the RI calculation was described by Tian et al. (2015). Cyclohexanone (378.8 mg/L cyclohexanone in absolute methanol; Sigma-Aldrich, St. Louis, MO, USA) was used as the internal standard. Sodium chloride was used for volatile extraction (Aprea et al., 2012), and the other reagents were all either of analytical grade or the highest purity commercially available.

Four different coating fibres for HS-SPME were purchased from Supelco, Inc. (Bellefonte, PA, USA): polydimethylsiloxane (PDMS-100 μm); carboxen/polydimethylsiloxane (CAR/PDMS-75 μm); polydimethylsiloxane/divinylbenzene (PDMS/DVB-65 μm); and DVB/CAR/PDMS-50/30 μm.

2.3. Analysis of proximate composition

The moisture content was determined using a halogen moisture detector (DHS-20A; Shanghai Eastsen Analytical Instrument Co., Ltd, China). Crude protein, crude fat, ash, sodium chloride, and nitrite content were determined according to the National Standard of PRC. Briefly, the crude protein content was determined using the Kjeldahl procedure with a nitrogen-to-protein conversion factor of 6.25 (GB5009.5-2010). The crude fat content was determined by weighing the PDM samples and extracting the crude fat with aether with a Soxhlet apparatus (GB/T5009.6-2003). The ash content was determined by heating 3-g samples in a furnace at 550°C for approximately 4 h until a constant weight was obtained (GB5009.4-2010). The sodium chloride content was determined according to the silver nitrate titration method (GB/T 12457-2008). The nitrite content was determined by the N(1-naphty1)-ethylenediamine dihydrochloride spectrophotometric method (GB5009.33-2016). The crude fibre content was determined using the method reported by Rowland and Roberts (1994).

2.4. HS-SPME-GC-MS analysis

HS-SPME-GC-MS analysis of the volatile compounds in PDM was performed as described by Cheng et al. (2015) with minor modifications. Briefly, the PDM sample (1 g) and the internal standard (10 μL) were immediately transferred to a 20-mL headspace bottle containing 10 mL of sodium chloride (NaCl) saturated solution. The mixture was then equilibrated with a hotplate/stirrer (model PC-420, Corning Inc. Life Science, Acton, MA, USA) at 60°C for 15 min. The volatile compounds were extracted in the following 50 min at 60°C by a manual SPME fibre coated with a divinylbenzene/Carboxen/polydimethylsiloxane (DVB/CAR/PDMS-50/30 μm, Supelco Inc. Bellefonte, PA, USA) film. The fibre was conditioned prior to use by heating in the injection port of a GC system at 270°C for 1 h. After the process of extraction, the fibre was removed from the headspace bottle and inserted into the injection port of the GC-MS apparatus for analysis of volatile compounds. Identification of volatile compounds was based on mass spectra matching in the standard Wiley and NIST11 library and RI reported in the literature (Cheng et al., 2016). The concentration of the identified volatile compounds (expressed as μg/g of PDM) was quantified from the relation of their areas with that of the internal standard (cyclohexanone, 3.788 μg/g of PDM). All of the HS-SPME-GC-MS measurements were conducted in triplicate for each PDM sample.

2.4.1. Optimal conditions of HS-SPME

The following experimental parameters were investigated to improve the absorption of volatile compounds of PDM: extraction temperature between 40°C and 80°C; extraction time between 20 and 60 min; incubation time between 5 and 25 min; and four coating fibres (DVB/CAR/PDMS, PDMS/DVB, CAR/PDMS, and PDMS). The PDM sample of GGX was randomly selected as the test object.

2.4.2. GC-MS conditions

Volatile compounds were analysed using an Agilent 7890A-5975C GC/MSD equipped with a DB-5 capillary column (30 m × 0.25 mm, 0.25 μm film thickness) (Agilent Technologies). The HS-SPME extract was desorbed from the fibre at 250°C for 3 min in the gas chromatography injection port. The injection port was operated in splitless mode, and 99.999% pure helium was used as the carrier gas at a constant flow rate of 1.4 mL/min. The oven temperature program was set for 5 min at 60°C, then raised to 180°C at a rate of 5°C/min, and held there for 1 min. Finally, the temperature was ramped up to 260°C at a rate of 10°C/min and isothermally maintained for 5 min. The mass spectra were obtained by electronic impact (EI) at 70 eV, and the electron ionization source temperature was maintained at 250°C. The data were collected at 1-s intervals over the m/z range of 29–350 am. They were compared with the spectra of the NIST11 library and validated by comparison to RI calculated with a homologous series of n-alkanes (C8-C20) ran under the same conditions as the samples.

2.5. Statistical analysis

All experiments were carried out in triplicate (n = 3). The data were analysed by one-way analysis of variance, using SPSS 20.0 (SPSS Inc., Chicago, IL, USA). The results were expressed as the means ± standard deviations.

3. Results and discussion

3.1. Proximate composition

In the study, to reduce the errors caused by sample selection, we collected five commercially available brands of PDM samples to identify the volatile compounds in PDM. Since the detailed production process of PDM by different manufacturers varied, the quality of the PDM products was different as shown by the colour differences in Figure 1. To better understand all these samples, we determined their proximate compositions which may be related to the differences of volatile compounds among the different PDM brands. The characterization of the proximate composition in five PDM brands is shown in Table 1. Moisture was the major component in PDM, and especially for the JTB brand, water was almost 50% of its weight. We assumed that moisture content is highly related to the drying process of PDM production, and the five PDM brands showed moderate differences. In addition, the crude protein content was significantly different among the five brands. Similar to fresh mustard (Li et al., 2012), PDM is rich in crude fibre and minerals (i.e. ash). There were no significant differences in the crude fat (with a low content of approximately 1 g/100 g) among all brands except for the JTB brand. Since PDM is a kind of preserved vegetable with a small amount of pure, granulated, non-iodized pickling salt treatment according to Huang et al. (2012), PDM presented a high content of sodium chloride, especially up to 21.33% in the GHW brand. In view of the toxicity of nitrite in foods (Chamandosst, Moradi, & Hosseini, 2016), its content in PDM is a common concern since PDM is a kind of pickled vegetable. The maximum nitrite content of pickled vegetables is set as 20 mg/kg according to the Food Safety Law in China. To our relief, all PDM samples in the study satisfied this requirement, but the GGX brand represented the highest nitrite content (10 mg/kg), which was almost three times that in other brands.

Table 1. Proximate compositions of five pickled and dried mustard brands (g/100 g).

Tabla1. Composiciones próximas de cinco marcas de mostaza encurtida y deshidratada (g/100 g).

Samples	Moisture	Crude protein	Crude fat	Crude fibre	Ash	Sodium chloride	Nitrite (mg/kg)
XH	28.91 ± 0.19^c	10.55 ± 0.13^c	1.19 ± 0.05^b	14.38 ± 0.63^b	27.81 ± 0.14^a	19.45 ± 0.56^b	2.81 ± 0.25^c
GGX	20.79 ± 0.47^e	13.38 ± 0.47^a	1.21 ± 0.12^b	16.68 ± 0.83^a	21.84 ± 0.14^d	13.13 ± 0.37^d	10.09 ± 0.62^a
GHW	40.21 ± 0.36^b	9.55 ± 0.04^d	1.08 ± 0.14^b	8.55 ± 0.97^d	26.96 ± 0.14^b	21.33 ± 0.17^a	3.96 ± 0.67^b
JTA	26.22 ± 0.58^d	12.51 ± 0.07^b	1.07 ± 0.10^b	14.38 ± 0.58^b	24.93 ± 0.14^c	17.79 ± 0.08^c	3.32 ± 0.67^bc
JTB	49.98 ± 0.14^a	8.17 ± 0.28^e	2.07 ± 0.08^a	10.68 ± 0.90^c	21.80 ± 0.14^d	13.59 ± 0.74^d	2.80 ± 0.26^c

Different letters within a column are significantly different (p < 0.05).

Las letras distintas en una misma columna indican que los valores son significativamente diferentes (p < 0.05).

In summary, the main approximate compositions in PDM were moisture, ash, sodium chloride, crude fibre, and crude protein, and the nitrite content in PDM was in the legal range. The five commercially available brands of PDM samples analysed in the study showed different contents of the approximate composition, suggesting the different quality of PDM products by various manufacturers.

3.2. Optimization of HS-SPME conditions

The volatile compounds in PDM were extracted using HS-SPME, and different extraction conditions as described previously were evaluated and optimized based on the total ion response in GC-MS. The peak area and number of peaks were selected as the evaluation index (Dong, Wang, Chen, Xia, & Jin, 2014).

Figure 2(a,b) shows the effects of extraction temperature and extraction time on the peak area and number of peaks for the volatile compounds in PDM by HS-SPME-GC-MS. The peak area of volatile compounds significantly rose with increasing extraction temperature or time. However, the number of peaks decreased when the extraction temperature was higher than 60°C or when the extraction time was more than 50 min. Therefore, the extraction temperature and time were selected as 60°C and 50 min, respectively.

Figure 2. Effects of extraction temperature (a), extraction time (b), incubation time (c), and coating fibres (d) on the peak area and number of peaks for volatile compounds in pickled and dried mustard by HS-SPME-GC-MS. Values are presented as the means ± SDs. Bars with different letters are significantly different (p < 0.05).

Figura 2. Efectos de la temperatura de extracción (a), duración de la extracción (b), duración de la incubación (c) y fibras de recubrimiento (d) en el área de picos y el número de picos de compuestos volátiles presentes en la mostaza encurtida y deshidratada analizados por HS-SPME-GC-MS. Los valores representan las medias ± DE. Las barras con letras diferentes son significativamente diferentes (p < 0.05).

Figure 2(c) shows the effects of incubation time on the extraction of volatile compounds from the PDM sample. With increasing incubation time, the peak area and number of peaks gradually increased and became constant after 15 min, indicating that the volatile compounds in the fibre, the vial headspace, and the sample reached equilibrium at 15 min. Therefore, 15 min was selected as the incubation time.

Figure 2(d) shows the effects of four different coating fibres (DVB/CAR/PDMS, PDMS/DVB, CAR/PDMS, and PDMS) on the extraction of volatile compounds from the PDM samples. The results showed that there were significant differences among the four coating fibres (p < 0.05), and the DVB/CAR/PDMS fibre obtained the largest peak area and the second largest number of peaks for the volatile compounds. Therefore, the DVB/CAR/PDMS fibre was selected for the experiments.

In summary, the optimal extraction conditions were as follows: extraction temperature, 60°C; extraction time, 50 min; incubation time, 15 min; and fibre coating, DVB/CAR/PDMS. These conditions were applied during the extraction of volatile compounds from the PDM samples.

3.3. HS-SPME-GC-MS analysis of volatile composition characteristics

With the optimal extraction condition, the volatile compounds of PDM were identified and quantified by GC-MS afterwards. The detected compounds and their concentrations (μg/g) in five different PDM brands from triplicate experiments are shown in Table 2. Approximately 32, 29, 33, 34, and 32 kinds of volatile compounds were identified in the XH, GGX, GHW, JTA, and JTB brands, respectively. There was no significant difference in the number of volatile compounds for the five PDM brands. The compounds were grouped into eight categories according to the chemical structure, i.e. hydrocarbons, alcohols, phenols, aldehydes, ketones, acids, esters, and others. Figure 3 shows the concentration of each category of volatile compounds in five pickled and dried mustard brands. Among the 41 total volatile compounds identified, aldehydes, acids, and esters were the predominant components in PDM according to their relatively high concentrations in almost all samples.

Table 2. Volatile compounds of pickled and dried mustard sampled by HS-SPME and identified by GC-MS and RI.

Tabla 2. Muestras de compuestos volátiles contenidos en la mostaza encurtida y deshidrata, tomadas por HS-SPME e identificadas mediante GC-MS y RI.

		LRI^2			Concentration (mean ± SD)/μg/g ^4,5,6
No.	Compounds^1	Cal.	Ref.	MI^3	XH	GGX	GHW	JTA	JTB
	Hydrocarbons
1	Tridecane	1312	1314	MS,RI	0.52 ± 0.17	0.62 ± 0.03	0.35 ± 0.09	0.28 ± 0.05	0.41 ± 0.18
2	Tetradecane	1412	1413	MS,RI	1.53 ± 0.42	0.64 ± 0.05	1.23 ± 0.19	0.79 ± 0.06	1.25 ± 0.28
3	Hexadecane	1613	1612	MS,RI	0.49 ± 0.14	0.28 ± 0.03	0.52 ± 0.14	0.27 ± 0.01	0.64 ± 0.08
	Total^7				2.54 ± 0.29^a	1.53 ± 0.12^bc	2.10 ± 0.35^ab	1.34 ± 0.00^c	2.30 ± 0.51^a
	Alcohols
4	2-phenylethanol	1141	1136	MS,RI	1.31 ± 0.30	1.49 ± 0.04	0.98 ± 0.15	0.82 ± 0.02	0.75 ± 0.35
5	hexadecan-1-ol	1405	–	MS	0.23 ± 0.04	0.06 ± 0.00	0.14 ± 0.02	nd	0.14 ± 0.02
	Total^7				1.54 ± 0.34^a	1.54 ± 0.04^a	1.13 ± 0.17^ab	0.82 ± 0.02^b	0.89 ± 0.34^b
	Phenols
6	2-methoxyphenol	1107	1072	MS,RI	nd	4.09 ± 0.34	nd	nd	nd
7	2,6-di(propan-2-yl)phenol	1424	–	MS	1.10 ± 0.03	1.33 ± 0.03	2.26 ± 0.35	0.88 ± 0.04	1.64 ± 0.08
8	2,6-ditert-butylphenol	1528	1520	MS,RI	0.27 ± 0.10	nd	0.24 ± 0.07	0.15 ± 0.02	0.32 ± 0.06
	Total^7				1.37 ± 0.11^d	5.43 ± 0.33^a	2.50 ± 0.39^b	1.03 ± 0.03^d	1.96 ± 0.08^c
	Aldehydes
9	Benzaldehyde	980	975	MS,RI	17.36 ± 2.09	nd	6.39 ± 0.66	5.43 ± 0.12	3.62 ± 0.34
10	2-phenylacetaldehyde	1062	1081	MS,RI	3.13 ± 0.51	3.39 ± 0.06	4.86 ± 0.37	5.63 ± 0.02	2.56 ± 0.21
11	Nonanal	1118	1110	MS,RI	nd	nd	0.69 ± 0.98	1.64 ± 1.16	3.06 ± 0.72
12	2-hydroxy-4-methylbenzaldehyde	1218	1218	MS,RI	13.73 ± 0.12	1.70 ± 0.15	nd	nd	nd
13	Decanal	1219	1210	MS,RI	nd	1.06 ± 0.40	2.27 ± 0.35	1.63 ± 0.07	1.72 ± 0.04
14	(E)-2-phenylbut-2-enal	1293	–	MS	3.22 ± 0.28	4.45 ± 0.64	3.84 ± 0.44	3.10 ± 0.09	2.42 ± 0.31
15	5-methyl-2-phenylhex-2-enal	1512	–	MS	nd	1.42 ± 0.14	1.27 ± 0.18	1.04 ± 0.12	nd
16	Tetradecanal	1630	1601	MS,RI	1.00 ± 0.49	nd	nd	nd	0.52 ± 0.17
17	(E)-3-(furan-2-yl)-2-phenylprop-2-enal	1719	–	MS	nd	0.38 ± 0.17	nd	nd	nd
18	pentadecanal	1734	1702	MS,RI	0.74 ± 0.39	nd	nd	nd	nd
	Total^7				39.18 ± 2.38^a	12.41 ± 0.43^c	19.32 ± 2.02^b	18.46 ± 1.04^b	13.91 ± 0.98^c
	Ketones
19	3,5-Octadien-2-one	1108	1097	MS,RI	3.66 ± 2.13	nd	0.68 ± 0.02	0.40 ± 0.02	0.32 ± 0.02
20	(5Z)-6,10-dimethylundeca-5,9-dien-2-one	1470	1453	MS,RI	0.79 ± 0.24	0.29 ± 0.09	0.51 ± 0.16	0.40 ± 0.04	0.61 ± 0.04
21	2,6-ditert-butylcyclohexa-2,5-diene-1,4-dione	1492	1462	MS,RI	1.44 ± 0.15	1.18 ± 0.21	2.08 ± 0.52	1.03 ± 0.05	1.88 ± 0.33
22	6,10,14-trimethylpentadecan-2-one	1866	1849	MS,RI	1.58 ± 0.97	0.39 ± 0.06	0.53 ± 0.17	0.41 ± 0.03	0.31 ± 0.02
	Total^7				7.47 ± 1.21^a	1.86 ± 0.12^c	3.79 ± 0.46^b	2.25 ± 0.14^c	3.13 ± 0.30^bc
	Acids
23	Heptanoic acid	1085	1073	MS,RI	3.40 ± 0.12	nd	4.81 ± 0.63	3.87 ± 0.08	3.41 ± 0.12
24	Octanoic acid	1183	1156	MS,RI	3.65 ± 0.71	5.45 ± 0.38	7.75 ± 0.97	7.23 ± 0.61	7.85 ± 0.94
25	Nonanoic acid	1277	1272	MS,RI	0.88 ± 0.43	0.97 ± 0.21	2.91 ± 0.60	2.49 ± 0.16	2.03 ± 0.31
	Total^7				7.93 ± 0.41^b	6.42 ± 0.59^b	15.47 ± 2.16^a	13.58 ± 0.44^a	13.29 ± 0.78^a
	Esters
26	methyl benzoate	1112	1113	MS,RI	0.16 ± 0.02	nd	0.37 ± 0.10	0.42 ± 0.02	0.41 ± 0.03
27	methyl 2-phenylacetate	1195	1177	MS,RI	0.61 ± 0.11	nd	0.47 ± 0.06	0.60 ± 0.05	0.56 ± 0.25
28	5-propyloxolan-2-one	1383	–	MS	0.13 ± 0.02	0.31 ± 0.01	0.46 ± 0.06	0.41 ± 0.04	0.56 ± 0.07
29	butyl butanoate	1395	–	MS	0.22 ± 0.09	nd	0.27 ± 0.04	0.19 ± 0.03	0.20 ± 0.01
30	4,4,7a-trimethyl-6,7-dihydro-5H-1-benzofuran-2-one	1567	1558	MS,RI	4.01 ± 0.74	2.30 ± 0.27	3.54 ± 0.56	2.27 ± 0.27	4.06 ± 0.44
31	bis(2-methylpropyl) benzene-1,2-dicarboxylate	1896	–	MS	0.45 ± 0.03	0.83 ± 0.21	0.44 ± 0.04	0.41 ± 0.07	0.25 ± 0.09
32	methyl hexadecanoate	1945	1934	MS,RI	0.73 ± 0.42	0.13 ± 0.02	nd	0.24 ± 0.03	0.14 ± 0.01
33	dibutyl benzene-1,2-dicarboxylate	1992	1975	MS,RI	nd	0.24 ± 0.04	0.16 ± 0.02	0.17 ± 0.05	nd
	Total^7				6.30 ± 1.24^a	3.80 ± 0.48^b	5.71 ± 0.68^a	4.70 ± 0.34^ab	6.20 ± 0.32^a
	Others
34	furan-2-carbaldehyde	857	843	MS,RI	7.30 ± 0.65	23.09 ± 0.23	5.61 ± 0.49	5.39 ± 0.39	2.52 ± 2.49
35	3-isothiocyanatoprop-1-ene	901	887	MS,RI	nd	nd	nd	0.13 ± 0.05	nd
36	2,5-dimethylpyrazine	928	918	MS,RI	2.43 ± 0.34	5.56 ± 0.31	1.46 ± 0.17	0.45 ± 0.03	0.85 ± 0.28
37	5-methylfuran-2-carbaldehyde	980	941	MS,RI	nd	43.36 ± 3.96	nd	nd	nd
38	1-(1H-pyrrol-2-yl)ethanone	1078	–	MS	nd	11.61 ± 0.69	0.58 ± 0.03	0.29 ± 0.06	nd
39	4,5-Dimethyl-1,3-thiazol-2-amine	1212	–	MS	0.35 ± 0.12	nd	0.59 ± 0.23	0.55 ± 0.13	0.50 ± 0.18
40	1,3-benzothiazole	1252	1208	MS,RI	0.46 ± 0.05	0.36 ± 0.07	0.31 ± 0.11	0.20 ± 0.04	0.28 ± 0.11
41	3-phenylpropanenitrile	1263	–	MS	0.68 ± 0.27	0.38 ± 0.07	0.97 ± 0.13	0.90 ± 0.03	0.97 ± 0.12
	Total^7				11.22 ± 0.52^b	84.35 ± 4.76^a	9.52 ± 0.55^bc	7.91 ± 0.29^bc	5.12 ± 2.74^c

^aCompounds with International Union of Pure and Applied Chemistry (IUPAC) name.

^bLRI cal., linear retention indices calculated from the RT of a series of n-alkanes standards (C8–C20). LRI Ref., linear retention referenced.

^cMethod of identification: MS, mass spectrum comparison using Wiley and NIST11 libraries; RI: retention index in agreement with literature value.

^dμg/g: concentration was expressed in microgram per gram of PDM, cyclohexanone was the internal standard, and the data listed were the mean of three assays ± SD (standard deviation).

^ePDM samples: XH, from Shaoxing Xianheng Food Co., Ltd., Shaoxing, China; GGX, from Jinyun County Quanyou Food Co., Ltd., Jinhua, China; GHW, from Hangzhou Guanhuawang Food Co., Ltd, Tonglu, China; JTA, from family workshop A, Hangzhou, China; JTB, from family workshop B, Hangzhou, China.

^fnd: not detected.

Different letters within a row are significantly different (p < 0.05).

¹Nombres de los compuestos determinados por la Unión International de Química Pura y Aplicada (iupac).

²LRI cal., índices de retención lineal calculados del RT de una serie de estándares de n-alcanos (C8–C20). LRI Ref., retención lineal referenciada.

³Método de identificación: MS, comparación de espectro de masas empleando las bibliotecas Wiley y NIST11; RI: índice de retención de acuerdo a los valores indicados en la literatura.

⁴μg/g: la concentración se expresa en microgramos por gramo de PDM, la ciclohexanona constituye el estándar interno y los datos enumerados corresponden a las medias de tres ensayos ± DE (desviación estándar).

⁵Muestras de PDM: XH, de Shaoxing Xianheng Food Co., Ltd., Shaoxing, China; GGX, de Jinyun County Quanyou Food Co., Ltd., Jinhua, China; GHW, de Hangzhou Guanhuawang Food Co., Ltd, Tonglu, China; JTA, del taller familiar A, Hangzhou, China; JTB, del taller familiar B, Hangzhou, China.

⁶nd: no detectado.

⁷Las letras distintas al interior de una misma fila indican valores significativamente diferentes (p < 0.05).

Figure 3. Concentration of each category of volatile compounds in five pickled and dried mustard brands (μg/g).

Figura 3. Concentración de cada categoría de compuestos volátiles en cinco marcas de mostaza encurtida y deshidratada (μg/g).

Aldehydes were listed as the largest volatile group with 10 compounds identified, in which benzaldehyde (No. 9), 2-phenylacetaldehyde (No. 10), decanal (No. 13), and (E)-2-phenylbut-2-enal (No. 14) were present in almost five PDM brands and considered to be the four main volatile compounds in PDM. These four components were also identified in some related research concerning mustard or other preserved vegetable. For example, benzaldehyde (No. 9), 2-phenylacetaldehyde (No. 10), nonanal (No. 11), decanal (No. 13), and tetradecanal (No. 16) were identified in fresh mustard and their pickles (Dayun Zhao et al., 2007), while 2-hydroxy-4-methylbenzaldehyde (No. 12) and (E)-2-phenylbut-2-enal (No. 14) were detected in Sichuan dongcai, a traditional salted vegetable (Yao et al., 2015). The total concentration of the aldehydes group, except for the GGX brand, was the highest value for each brand. The concentration of aldehydes was 39.18 ± 2.38 μg/g in the XH brand, which was almost two to three times the concentration in other brands.

Though only three acids were identified in PDM, the total concentration of acids was higher than that of eight esters in every PDM brand. The three acids included heptanoic acid (No. 23), octanoic acid (No. 24), and nonanoic acid (No. 25), and they richly existed in all PDM samples except that heptanoic acid (No. 23) was not detected in the GGX brand. The concentration of acids in GHW (15.47 ± 2.16 μg/g), JTA (13.58 ± 0.44 μg/g), and JTB (13.29 ± 0.78 μg/g) was almost twice that in the XH (7.93 ± 0.41 μg/g) and GHW (6.42 ± 0.59 μg/g) brands. For esters, the concentration in the GGX brand was significantly different from the other brands, showing the lowest content of 3.80 ± 0.48 μg/g. Among the eight esters, 4,4,7a-trimethyl-6,7-dihydro-5H-1-benzofuran-2-one (No. 30) turned out to be the most abundant volatile compound in the PDM samples with the concentration varying from 2.27 ± 0.27 to 4.06 ± 0.44 μg/g. Zhao et al. (2007) identified much more abundant esters in fresh mustard and their pickles compared to our study of PDM samples, and only methyl hexadecanoate (No. 32) was detected in both studies.

The contents of three hydrocarbons and two alcohols (<2 μg/g) were relatively lower than those of esters in the PDM volatile compounds. However, hydrocarbons are common volatile compounds; several hydrocarbons such as tridecane (No. 1), pentadecane, hexadecane (No. 3), and dodecane were identified as relevant odour compounds in seaweed from Shizuoka prefecture (Yamamoto et al., 2014). In this study, tetradecane (No. 2) and 2-phenylethanol (No. 4) were slightly more abundant than the others. Tridecane (No. 1) identified in PDM was also found in fresh mustards but not in their pickles (Dayun Zhao et al., 2007). Moreover, Xu et al. (2013) reported that alcohols were one of the most abundant flavour compounds in eight kinds of fresh mustards. This could be explained by the fact that the alcohols in mustards were used to form other flavour compounds such as esters after the mustards were pickled and dried.

The contents of ketones with four compounds varied from 1.86 ± 0.12 to 7.47 ± 1.21 μg/g in the different brands. In this group, 2,6-ditert-butylcyclohexa-2,5-diene-1,4-dione (No. 21) was the main volatile compound in the PDM samples, which appeared in all five brands with a relatively high concentration. Since the concentration of 3,5-octadien-2-one (No. 19) was significantly higher in the XH brand, which was also detected in fresh mustard (Zhao et al., 2007), the total concentration of ketones in the XH brand was the highest among all the PDM brands. Similarly, the content of phenols with three compounds also varied among the different brands (1.03 ± 0.03–5.43 ± 0.33 μg/g). 2-Methoxyphenol (No. 6) was only detected in the GGX brand at a concentration of 4.09 ± 0.34 μg/g, while 2,6-di(propan-2-yl)phenol (No. 7) was detected in all brands at a rather high concentration.

As for others, this group also presented high concentrations in PDM. 5-Methylfuran-2-carbaldehyde (No. 37; 43.36 ± 3.96 μg/g) and 1-(1H-pyrrol-2-yl)ethanone (No. 38; 11.61 ± 0.69 μg/g) were only detected with high values in the GGX brand, leading to the highest group content of the brand (84.35 ± 4.76 μg/g). The JTB brand presented the lowest concentration of others (5.12 ± 2.74 μg/g), while the concentrations in XH (11.22 ± 0.52 μg/g), GHW (9.52 ± 0.55 μg/g), and JTA (7.91 ± 0.29 μg/g) brands showed a non-significant difference. Zhao, Tang, and Ding (2001) reported that 1,3-benzothiazole (No. 40), which was identified in all five PDM brands in our study (< 0.5 μg/g), played an important role in the typical flavour components of potherb mustard pickles. Furan-2-carbaldehyde (No. 34) and 2,5-dimethylpyrazine (No. 36) were also presented in all five PDM brands and their concentrations were relatively higher than other heterocyclic compounds, which should be considered to be the two main volatile compounds in PDM.

It is worthy to note that PDM is made of Brassica juncea, Coss., a kind of vegetable that belongs to the order Brassicales. Plants that belong to this order characteristically contain glucosinolates, which are sulphur-containing secondary plant metabolites (Hanschen, Lamy, Schreiner, & Rohn, 2014). Isothiocyanate, thiocyanates, and nitrile are the three main breakdown products of glucosinalates, which are responsible for the pungent flavour of several Brassica vegetables (Hashimoto, Miyazawa, & Kameoka, 1982). Zhao et al. (2007) identified five isothiocyanates and four nitriles in fresh potherb mustard and their pickles, which accounted for the major proportion of the volatile compounds. In this study, however, we only found two breakdown products of glucosinolates, i.e. 3-isothiocyanatoprop-1-ene (No. 35; only detected in the JTA brand with a low concentration of 0.13 ± 0.05 μg/g) and 3-phenylpropanenitrile (No. 41). Though their concentrations were the lowest of the total identified compounds, their contributions to the typical flavour of PDM should never be ignored.

From a comparison of the five PDM brands (Figure 3), the overall volatile composition characteristics were found to be similar especially in the numbers and the kinds of compounds; however, the concentrations of volatile compounds varied in the PDM products by different manufacturers. It is clearly shown in Figure 3 that the GGX brand presented the highest concentration of volatile compounds whereas the JTB brand showed the lowest. Besides, aldehydes and acids were the two components with relative highest concentration in all the GHW, JTA, and JTB brands. But aldehydes and others were the major contents of volatile compounds in the XH and GGX brands, respectively. The differences appeared to be related to the dynamic changes of volatile compounds in the production process of PDM from fresh mustard. To maintain a better quality of PDM, especially the flavour quality, a better production process is needed in the future. Since Zhao et al. (2007) evaluated the volatile components of fresh mustard (Brassica juncea, Coss.) and their pickles with different pickling times using SPME-GC-MS, a further study could be focused on the dynamic changes of volatile compounds in the drying process of producing PDM from the pickles.

In summary, a total of 41 volatile compounds were identified and quantified in all PDM brands by HS-SPME-GC-MS measurement. Approximately 15 compounds were considered to be the predominant volatile compounds in PDM according to their relatively high concentrations among all of the identified compounds, including tetradecane (No. 2), 2-phenylethanol (No. 4), 2,6-di(propan-2-yl)phenol (No. 7), benzaldehyde (No. 9), 2-phenylacetaldehyde (No. 10), decanal (No. 13), (E)-2-phenylbut-2-enal (No. 14), 2,6-ditert-butylcyclohexa-2,5-diene-1,4-dione (No. 21), heptanoic acid (No. 23), octanoic acid (No. 24), nonanoic acid (No. 25), 4,4,7a-trimethyl-6,7-dihydro-5H-1-benzofuran-2-one (No. 30), furan-2-carbaldehyde (No. 34), 2,5-dimethylpyrazine (No. 36), and 3-phenylpropanenitrile (No. 41). Additionally, it is noteworthy that some compounds showed remarkable concentrations in some samples, such as 3,5-octadien-2-one (No. 19) in the XH brand and 5-methylfuran-2-carbaldehyde (No. 37) and 1-(1H-pyrrol-2-yl)ethanone (No. 38) in the GGX brand.

4. Conclusion

In this study, to identify and quantify the volatile compounds of PDM, five commercially available brands of PDM samples were analysed by optimized HS-SPME-GC-MS measurement. The optimal extraction conditions were selected based on the number of peaks and cumulative peak areas as follows: extraction temperature, 60°C; extraction time, 50 min; incubation time, 15 min; and fibre coating, DVB/CAR/PDMS. There are altogether 41 volatile compounds were identified in PDM, and aldehydes were the dominant components. Besides, the concentrations of volatile compounds varied in five PDM brands by different manufacturers. Further work will attempt to clarify the dynamic changes of volatile compounds in PDM during the drying process. Moreover, the effects of different domestic cooking methods on the volatile compounds in PDM will be investigated.

Acknowledgement

This work was supported by the National Science-Technology Support Plan Project of China under Grant No. 2014BAD04B01.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the National Science-Technology Support Plan Project of China under Grant No. 2014BAD04B01.