Quantitative analysis of predominant yeasts and volatile compounds in the process of pickled wax gourd

Abstract

Yeasts have been reported to be responsible for flavor compound formation in, and the quality of, pickled wax gourd. In this study, the diversity of yeasts and their predominant species in pickled wax gourd, as well as the capability to produce aroma compounds were detected by 5.8 S rDNA-ITS clone library combined with RT-PCR technology and GC–MS. It showed the predominant yeast species in the pickling processing were Issatchenkia orientalis, Pichia kudriavzevii and Candida xylopsoci, and clones numbers of each yeast species reached 10⁷ copies mL^–1. A total of 66 volatile constituents were identified and quantified, and the major volatile compounds of pure culture with yeast were alcohols, which were the critical flavor composition of pickled wax gourd. Together, these results could provide information about the quality control of pickled wax gourd including the flavor composition and the growth of predominant microbial species.

Se ha demostrado que las levaduras son responsables de la formación del compuesto sabor y la calidad de la calabaza blanca en escabeche. En este estudio, se detectaron la diversidad de levaduras y sus especies predominantes en la calabaza blanca en escabeche, como también la capacidad de producir compuestos aromáticos mediante la genoteca 5.8 S rDNA-ITS combinada con tecnología RT-PCR y GC-MS. Todo ello mostró que las especies de levadura predominantes en el proceso de escabeche son: Issatchenkia orientalis, Pichia kudriavzevii y Candida xylopsoci, además un número de clones de cada especie de levadura alcanzó las 10⁷ copias mL^–1. Un total de 66 componentes volátiles fueron identificados y cuantificados. Además los mayores compuestos volátiles de cultivo puro con levaduras fueron los alcoholes, los cuales fueron la composición de sabor fundamental de la calabaza blanca en escabeche. Todos estos resultados conjuntamente podrían ofrecer información acerca del control de calidad de la calabaza blanca en escabeche, la composición del sabor y el crecimiento de especies microbianas predominantes.

Keywords:

• wax gourd
• yeast
• clone library
• RT-PCR
• volatile compounds

Palabras clave:

• calabaza blanca
• levadura
• genoteca
• RT-PCR
• compuestos volátiles

Introduction

Pickled wax gourd is a very distinctive traditional fermented food in eastern area Zhejiang province, China, well known as an appetizer food because of its special flavor. The flavor and nutrition of pickle products are due to the microorganism of pickled food system, including yeasts. The flavoring compounds caused by yeast metabolism mainly include esters, alcohols and acids, depending on the species and amounts of yeasts in fermentation process (Hazelwood, Daran, Van Maris, Pronk, & Dickinson, 2008; Zhuang, Fu, & Huang, 2012). Alcohols and esters are the main group of compounds that form the fermentation flavor (Romano, Fiore, Paraggio, Caruso, & Capece, 2003). However, alcoholic fermentation of yeasts will affect the normal flavor. Therefore, it is interesting and important to study the yeast diversity of pickled wax gourd and the flavoring aroma compounds involved (Shi & Zhou, 2002). Interactions of microorganisms produce different volatile compounds, the quality attributes of which define the characteristics of pickled wax gourd and have great influence on its appreciation by consumers.

Previous studies (Lan, Chen, & Yanagida, 2009; Yuan, Lu, & Yu, 2009) on microbiota of fermented wax gourd was always conducted after the culture of different microorganisms in different media. It is a time-consuming method, and new techniques for microbial ecology have been reported such as clone library and the real-time quantitative PCR (RT-PCR), which avoid the shortcoming with traditional numerical populations associated with culture-dependent methods (Andorrà, Berradre, Mas, Esteve-Zarzosoa, & Guillamóna, 2012). The presence of viable but non-cultivable microorganisms in wax gourd has been previously reported (Zhao, Wu, Shen, Weng, & Chen, 2014). The use of culture-independent methods to identify and quantify yeasts is an ideal tool for studying yeast species interaction. Most of them rely on the direct amplification of yeast DNA from sample by PCR (Andorrà et al., 2012). PCR is specific and sensitive, especially the real-time quantitative PCR (QPCR). QPCR is a kind of creative technique that combines nucleic acid amplification, hybridization and spectral analysis with real-time detection, and is the most effective technique in microbial detection (Bleve, Rizzotti, Dellaglio, & Torriani, 2003).

The fermentation system of wax gourd harbors a large, active and complex community of microorganism. In previous studies (Lan et al., 2009; Shen, Zhao, Wu, & Weng, 2012; Yuan et al., 2009), the diversity of bacteria species in the process of pickled wax gourd, as well as quality characteristics associated with pickled wax gourd products were well documented by using the systematic study of molecular detection method (Zhao et al., 2014). The volatile flavor composition can be formed by yeasts, which affects the flavor of the picked wax gourd. The present study is the first to aim to identify the individual yeast species and dynamic changes of yeast population in pickled wax gourd. The aim of this study was to use RT-PCR with SYBR Green I for rapid and precise analysis of yeast species in the process of producing pickled wax gourd. The volatile components profile of the predominant yeasts was also detected by gas chromatography mass spectrometry (GC–MS). The result can elucidate the relation between the product quality and yeast species in pickled wax gourd, and it can provide the microbial base for the quality control of pickled vegetables processing as well.

Materials and methods

Preparation of fermented samples

Wax gourd was purchased from the farmers’ market of Ningbo city and was pickled in accordance with the following method: the wax gourd was washed, seeded, cut (about 7 × 7 cm²) and mixed with 2.5% salt in a bucket in face-to-face, back-to-back manner, and salt was cast on sides other than the skin side. After keeping it for 24 h at room temperature, the wax gourd was fished out, mixed with salt (NaCl) and layered in a jar and then the jar was sealed. Salt was added to a final concentration of approximately 7%. Fermentation was continued for about 90 days at room temperature. The fermented samples (about 50 mL brine) were collected from the center of the jar on days 15, 30, 60 and 90, respectively.

DNA extraction

After filtration (with four layers of sterilized gauze) and centrifugation (4°C, 4500g), the supernatants were removed. The genomic DNA was then extracted using Fast DNA SPIN Kit (MP Bio-medicals, Santa Ana, CA, USA) according to the manufacturer’s instruction.

Yeast 5.8S rDNA-ITS gene clone library construction and sequencing

PCR was performed in a final volume of 50 μL reaction mix containing 5.0 μL of PCR buffer with MgCl₂ (TaKaRa, Japan), 4.0 μL of PCR nucleotide mix (2.5 mM each dNTP, TaKaRa), 1.0 μL of each primer (ITS1 and ITS4), 0.25 μL Ex-Taq DNA polymerase (5 U μL^–1, TaKaRa), 1.0 μL template DNA (genomic DNA extracted from fermented samples as described above), and 36.5 μL nuclease-free water. This mixture was then loaded into 0.2 mL thin-wall microcentrifuge tubes and subjected to the following PCR programs: an initial 4 min denaturizing at 94°C, followed by 33 cycles of denaturizing at 94°C for 30 s, annealing at 55°C for 30 s, with an extension at 72°C for 45 s, and a final elongation step of 10 min at 72°C. After cycling, the PCR products were analyzed on a 1.5% agarose gels stained with ethidium bromide in 1x Tris–Borate–EDTA buffer, and purified using an EZ-10 Spin Column DNA Gel Extraction Kit (Sangon, China). The PCR reaction of each sample includes four repeats and a blank control in order to avoid errors and contamination.

The PCR purified products were connected into a pMD18-T vector (TaKaRa) and transformed into Escherichia coli JM109 competent cells (TaKaRa), following the manufacturer’s instructions. Clones were then randomly selected for sequencing with primers M13F/R targeting the flanking regions of the multi-cloning site of pMD18-T vector. The sequences of 5.8 S rDNA-ITS were edited by using DAMBE and DNA star program and identified using BLAST at National Center for Biotechnology Information (NCBI).

Quantitative analysis of predominant yeast species

From the result of bacterial 5.8 S rDNA-ITS gene clone library, the information about the predominant yeast could be obtained during the fermentation process. Primers (Table 1) for predominant yeast species were designed by using MEGA and primer 5.5, based on the sequences of 5.8 S rDNA-ITS gene clone library. Io39F/Io138R, Pi257F/Pi426R and Ca51F/Ca203R are suitable for differentiation of Issatchenkia orientalis, Pichia kudriavzevii and Candida xylopsoci with other yeasts. They have different PCR programs. Io39F/Io138R as follows: 95°C for 2 min, 40 cycles of 95°C for 15 s, 60.2°C for 20 s, 72°C for 20 s and followed by a melting curve (95°C for 15 s, 60°C for 15 s, heating for 20 min, 95°C for 15 s). Pi257F/Pi426R as follows, 95°C for 2 min, 40 cycles of 95°C for 15 s, 62.5°C for 20 s, 72°C for 15 s and a melting curve as described above. Ca51F/Ca203R as follows, 95°C for 2 min, 40 cycles of 95°C for 15 s, 61°C for 20 s, 72°C for 15 s and a melting curve as described above.

Table 1. Primers designed to target species-specific regions of the 5.8 S rDNA-ITS gene of predominant yeast species.

Tabla 1. Cebadores diseños para detectar regiones de especies específicas del gen 5.8 S rDNA-ITS de las especies de levadura predominantes.

Target yeast	Primer	F/R	Primer sequence	Amplified length (bp)
Issatchenkia orientalis	Io39F	F	CGGAACGAAAACAACAACACC	100
	Io138R	R	CCAAGAGATCCGTTGTTGAAAGT
Pichia kudriavzevii	Pi257F	F	TTTGAGCGTCGTTTCCATCTTGC	170
	Pi426R	R	TTTGAGCGTCGTTTCCATCTTGC
Candida	Ca51F	F	GAAATAAATTGTGGTGGCCACTAG	153
	Ca203R	R	CCTGTTTGAGCGTCATTTCTCC

The method used in this study is to generate a standard curve using plasmids containing target ribosomal DNA sequences (Perini et al., 2011). The standard curve for predominant yeast species was obtained from corresponding 10-fold scalar dilutions of purified products (three replicates) by specific primers. In all experiments, negative controls (NTC) containing sterilized water were tested in triplicate. Clones with plasmids containing target (such as Issatchenkia orientalis, Pichia kudriavzevii and Candida xylopsoci) ribosomal DNA sequences from 5.8 S rDNA-ITS gene clone library were amplified with M13F and M13R primers and then analyzed on a 1.5% agarose gel and purified using a Gel Extraction Kit. The concentration of purified product was measured with a Qubit fluorometer (from Life Technologies, Carlsbad, CA, USA) following the manufacturer’s instructions. The copy number of the product was calculated using the following formula: molecules (μL^–1) equals to (A × 6.022 × 10²³) × (660 × 10⁹ × B)^–1, where A is the purified DNA concentration (ng μL^–1), B is the length of the cloned sequence, 6.022 × 10²³ is the Avogadro’s number and 660 is the average molecular weight of one base pair.

Quantitative real-time PCR was performed in a final volume of 20 μL containing 10 μL SYBR Green mix (TaKaRa), 0.5 μL specific primers and 1 μL sub-samples of 10-fold scalar dilutions products and 8 μL nuclease-free water. All amplification reactions were carried out using fluorescence quantitative PCR instrument (Eppendorf, Germany). Clones with plasmids containing 5.8 S rDNA-ITS sequences such as Issatchenkia orientalis, Pichia kudriavzevii and Candida xylopsoci were also used to test primer specificity by RT-PCR.

Determination of the growth curve of predominant yeast species and pH curves

To 150 mL malt extract broth (MEB) was added 15 mL individual predominant yeast species that had been activated for 12 h in 500 mL Erlenmeyer flask, and centrifuged in shake flasks at 28°C and 1500g. The pH value of yeast culture was measured every 2 h during the fermentation. Growth curve and pH curve were drafted when time was the abscissa, OD₅₆₀ (optical density at 560 nm) and pH value were the vertical axis.

Volatile compounds analysis

Each culture solution was sampled 2 mL into a 15 mL headspace vial in the stable phase of cultivating in medium (MEB). The samples were pretreated by Headspace Solid-Phase Micro-extraction (HS-SPME), inserting the injector needle of the SPME manually through the septum bottle, launching fiber. The steps were as follows: 40 min isothermal adsorption at 50°C, then moved to 20°C water bath for 10 min balance, retracted fiber, and directly injected (1 μL) in GC system.

Chromatographic separation was achieved with the Vocol column (J & W, Agilent Technologies, CA, USA), which has dimensions of 1.8 μm × 60 m × 0.32 mm film thickness. Helium was used as the carrier gas at a constant flow rate of 0.8 mL min^–1. The column temperature was initially held at 35°C for 3 min, then, the temperature was increased until 40°C with a heating rate of 3°C min^–1, and the temperature was held for 1 min. The second ramp was programmed from 40°C to 210°C with a heating rate of 5°C min^–1. A post-run of 16 min at 210°C was established. The mass spectrometer was used in electron ionization mode; all spectra were acquired using a mass range of m/z 30 – 650. The transfer line temperature was set at 210°C, the ion source temperature at 200°C, detector voltage 0.8 kV. Retrieval library: Wiley and NIST library. For the samples processed by HS-GC–MS, the absolute areas were directly used for the data elaboration.

Statistical analyses

The results obtained were analyzed using SPSS version 16.0 (SPSS Inc., Chicago, IL, USA). Any significant difference was determined by one-way analysis of variance (ANOVA) followed by the Tukey test for multiple comparisons considering difference statistically at P < 0.05.

Results and discussion

Construction of yeast 5.8 S rDNA-ITS gene clone library

Sequencing of the positive clones from the samples of pickled system was identified by BLAST in the GenBank database. The results showed that the picked process involving six species, including uncultured eukaryote, which was not cultured in the experiment. All the sequences from the four different periods had a high similarity of more than 99% compared to the 5.8°S rDNA-ITS gene sequence published in GenBank (NCBI). The composition and distribution of yeast communities in different pickled periods are shown in Table 2.

Table 2. Yeast community composition of raw pickled wax gourd based on 5.8 S rDNA-ITS clone library.

Tabla 2. Composición de una comunidad de levaduras de calabaza blanca cruda en escabeche basada en la genoteca 5.8S rDNA-ITS.

Genus	Species/total amount of species	The amount of clones in different periods (d)
		15	30	60	90
Candida	Candida tropicalis	3		2	1
	Candida xylopsoci		12	12	10
Issatchenkia	Issatchenkia orientalis	18	11	11	15
Rhodotorula	Rhodotorula mucilaginosa	1
Uncultured eukaryote	Uncultured eukaryote			1	1
Pichia	Pichia kudriavzevii	18	17	14	13

It shows that the predominant yeast species on day 15 were Issatchenkia orientalis and Pichia kudriavzevii, the number of clones was the same, 18; they accounted for 45% (18/40) of the total number of clones in the sample, respectively. In addition, other yeast species of Candia tropicalis and Rhodotorula mucilaginosa, were also detected from the sample, whose number of clones was 3 and 1, respectively; they accounted for 7.5% (3/40) and 2.5% (1/40) of the total capacity, respectively. On day 30, Pichia kudriavzevii was still the predominant yeast species, the number of clones was 17, which accounted for 42.5% (17/40), followed by Issatchenkia orientalis’s, whose number of clones was 11, accounting for 27.5% (11/40) of the library. Candida xylopsoci did not appear by the 15th day and began to be detected after 30 days, with the number of clones being 12, accounting for 30% (12/40) of the library. The community structure on day 60 and 90 was similar. The predominant yeast species at this stage consisted of Issatchenkia orientalis, Pichia kudriavzevii and Candida xylopsoci. The proportion of Issatchenkia orientalis in total clones from the samples on day 60 and 90 were 27.5% (11/40) and 37.5% (15/40), respectively. Similarly, Pichia kudriavzevii was 35% (14/40) and 32.5% (13/40), while Candida xylopsoci was 30% (12/40) and 25% (10/40), respectively.

Quantitative analysis of predominant yeast in the pickling process

RT-PCR reactions were conducted with samples fermented for 15, 30, 60 and 90 days, and the results are shown in Figure 1. The regression coefficient (r²) values for standard curves of all RT-PCR assays in each run were always above 0.99, and the amplification efficiency is between 0.98 and 1.02.

Figure 1. Quantitative analysis of predominant yeast species including Issatchenkia orientalis (Io) (a), Pichia kudriavzevii (Pi) (b), and Candida xylopsoci (Ca) (c) in pickled processing of wax gourd.

Figura 1. Análisis cuantitativo de las especies predominantes de levadura incluidas Issatchenkia orientalis (Io) (a), Pichia kudriavzevii (Pi) (b) y Candida xylopsoci (Ca) (c) en el proceso de escabeche de la calabaza blanca.

The 5.8 S rDNA-ITS copy numbers of the predominant yeasts in the fermented samples were calculated by comparison of the Ct value (threshold cycle) and standard curves. Taking into account of the DES (diethyl sulfate) buffer volume and sample volume in the DNA extraction step, the copies of yeast ribosomal should be calculated as follows: copies of yeast ribosomal (copies mL^–1) equals to (copy number × 120) × 50^–1, where 120 is the volume of DES buffer (μL), 50 is the volume of sample used for DNA extraction. Copies of the predominant yeast ribosomal in the fermented process are shown in Table 3.

Table 3. Copies of the predominant yeast species ribosome in pickled wax gourd product.

Tabla 3. Copias de especies ribosoma en levaduras predominantes en el producto calabaza blanca en escabeche.

Predominant yeast species	Copies of the predominant yeast species ribosomal (copies mL^–1)
	15 d	30 d	60 d	90 d
Issatchenkia orientalis	4.55 ± 0.21 × 10^7 c	4.18 ± 0.18 × 10^7 b	9.99 ± 0.26 × 10^7 d	3.26 ± 0.24 × 10^7 a
Pichia kudriavzevii	2.90 ± 0.14 × 10^8 c	2.13 ± 0.11 × 10^8 b	7.62 ± 0.34 × 10^8 d	2.06 ± 0.13 × 10^8 a
Candida	6.57 ± 0.26 × 10^2 a	4.94 ± 0.15 × 10^7 c	4.88 ± 0.37 × 10^7 c	2.02 ± 0.12 × 10^7 b

Note: One-way ANOVA and Tukey tests were used to determine significant differences for copies of the predominant yeast species ribosome in pickled wax gourd product. Different lowercase letters in one column indicate significant differences (P < 0.05) for copies among the different time points.

Nota: Se utilizaron ANOVA de un sentido y los tests Tukey para determinar diferencias significativas para copias de especies predominantes ribosoma en levaduras en el producto calabaza blanca en escabeche. Las diferentes letras minúsculas indican diferencias significativas (P < 0,05) para las copias entre los diferentes tiempos.

Table 3 shows that there is little variation in the population of Pichia kudriavzevii throughout the pickling process on 15–90 days, the copies remained 2.90 × 10⁸copies mL^–1 on the 15th day. As Candida was growing, the copies of Pichia kudriavzevii ribosomal decreased, then the copies increased again till to 2.06 × 10⁸ copies mL^–1. Issatchenkia orientalis had the same trend as Pichia kudriavzevii, declining gradually during 15–30 days fermentation and then increased to a high level of 9.99 × 10⁷ copies mL^–1 at the 60th day, then declining to reach a final value of 3.26 × 10⁷ copies mL^–1. The population of Candida spp. in the pickled sample was of a very low level at the early stage, it reached only 6.57 × 10² copies mL^–1 in 15 days, the copies gradually increased during 30–60 days pickling fermentation, and then decreased, finally became stable at 2.02 × 10⁷ copies mL^–1.

Growth curve and pH curve of predominant yeast species

The growth of yeasts and pH value changes are important factors on the quality of the product. This information is critical for quality control in the processing of pickled wax gourd. The predominant yeast species of the uncooked wax gourd were Issatchenkia orientalis, Pichia kudriavzevii and Candida xylopsoci. The growth and pH curve for three individual yeast species in MEB medium were showed in Figure 2. The results showed that the yeast growth was type-S in all three yeast species, while pH value declined as the growth of various yeasts. Issatchenkia orientalis reached the stable growth phase after 18 h cultivation, the pH value decreased rapidly from the 6th hour to 12th hour, finally becoming stable at 3.9. Similarly, Pichia kudriavzevii reached the stable growth phase after 16 h cultivation, the pH value decreased rapidly from the 6th hour to the 10th hour, reaching a final value of 3.8 after 22 h cultivation. While Candida xylopsoci reached the stationary growth phase at the 14th hour, the pH value decreased gradually to a final value of 4.5 after 15 h cultivation.

Figure 2. The growth and pH curve of predominant yeast species: (a) Issatchenkia orientalis; (b) Pichia kudriavzevii; (c) Candida xylopsoci. Expression values are the means ± standard deviations of data from at least two independent analyses of samples.

Figura 2. El crecimiento y curva pH de las especies predominantes de levadura: (a) Issatchenkia orientalis; (b) Pichia kudriavzevii; (c) Candida xylopsoci. Los valores de expresión son promedios ± desviación estándar de los datos de al menos dos análisis independientes de las muestras.

Analysis of volatile components in pickled wax gourd with GC–MS

The volatile compounds of pickled wax gourd product, pure culture from individual Issatchenkia orientalis, Pichia kudriavzevii and Candida xylopsoci were detected by GC–MS, as shown in Table 4. The class and content in four different samples are shown in Table 5. As a consequence, a total of 66 volatile constituents were identified and quantified in four different samples, which included 6 acids, 11 alkanes, 10 alcohols, 4 ketones, 12 esters, 8 aldehydes, 4 alkenes and 11 others. The four samples have been proved to be strong producers of alcohols during fermentation. For ethanol, its relative content accounted for more than 15.72% in all volatile compounds detected. Meanwhile, three different yeasts investigated showed similar levels of 3-methyl-1-butanol, and its relative content of the 3-methyl-1-butanol accounted for more than 36.23%, but the pickled wax gourd brine showed a lower level of 10.51%. The 2-methyl-1-butanol was only detected in pickled wax gourd brine, but not detected in individual yeasts. The wax gourd brine also proved to be a strong producer of organic acids by pickling fermentation, while the three yeast species showed a minor contributor in the culture medium. This result showed that most organic acids are from bacterial fermentation in pickled wax gourd production, which is associated with our previous study (Zhao et al., 2014). High amounts of acetic acid is produced in wax gourd brine, but it was not detected in all individual yeast species culture. Propionic acid, acetic acid and 2-hydroxy propanoic acid were also detected in wax gourd brine sample.

Table 4. Comparison of volatile compound composition of pickled wax gourd product (S1), Issatchenkia orientalis (S2), Pichia kudriavzevii (S3) and Candida xylopsoci (S4) by GC–MS analysis.

Tabla 4. Comparación de la composición de compuestos volátiles del producto cabalaza blanca en escabeche (S1), Issatchenkia orientalis (S2), Pichia kudriavzevii (S3) y Candida xylopsoci (S4) mediante análisis GC-MS.

No.	Compound	Relative molecular mass	The relative content (%)
			S1	S2	S2	S4
Acids
1	Propionic acid	70	3.71
2	Acetic acid	60	21.43
3	Butanoic acid	46	2.73
4	3-Methyl butyrate	102	0.95	0.10		0.34
5	2-Methyl butyrate	102	1.48	0.04		0.28
6	2-Hydroxy-propanoic acid	90	0.66
Alkanes
7	Cyclopropane	42	0.08
8	2,2-Diethoxy propane	132	0.05
9	Cyclotetrasiloxane, octamethyl-	296	1.11	0.25	0.40	0.30
10	Dimethylsiloxane pentamer	370	3.26	0.63	1.14	1.06
11	Cyclohexasiloxane, dodecamethyl	444	6.20	0.73	0.62	1.16
12	1,1-Diethoxyethane	118		0.06	0.08
13	p-Tolyltrimethylsilane	164		0.39
14	2-Chloro-2-methyl-2-heptane	148		0.03	0.03
15	2-(2,2-Dichloroethyl)-2-methyl-1-oxa-cyclobutane	168			0.03
16	Tetranitromethane	196				0.08
17	Hexane	282				0.05
Alcohols
18	Ethanol	46	19.96	15.72	18.65	26.12
19	Silanol, trimethyl	90	3.01
20	2-Methyl propanol	74	2.95	7.50	7.43	2.96
21	1-Butanol	74	0.44
22	3-Methyl-1-butanol	88	10.51	49.14	44.89	36.23
23	2-Methyl-1-butanol	88	8.66
24	Benzeneethanol	122	0.71	5.80	7.62	8.06
25	2-Furanmethanol	98		0.02	0.05
26	3-Methylthio-1-propanol	106			0.08
27	α-Trimethyl-7-phenyl-bicyclo nonan-β-ol	258			0.10
Ketones
28	2,3-Pentanedione	100		0.08	0.08	0.46
29	3-(2-Phenylethyl)pentane-2,4-dione	204		0.04
30	2,3-Butanedione	86			0.09	0.46
31	4-Methyl-2-pentanedione	100			0.08	0.23
Esters
32	Methyl acetate	74	1.14
33	Ethanoic acid 2,2,2-trichloroethyl ester	190	0.19
34	Acetate ester	88	8.31	13.89	12.8	11.92
35	2-Hydroxy-butanoic acid methyl ester	118	0.25
36	Acetic acid ethenyl ester	86		0.11
37	Propanoic acid ethyl ester	102		0.78	0.65	0.20
38	3-Oxo-butanoic acid 2-propenyl ester	143		0.14
39	Ethyl isobutyrate	116		0.47	0.33
40	Acetic acid,2-methylpropyl ester	116		0.48	0.65	0.74
41	Ethyl butyrate	116		0.08	0.06
42	Isoamyl acetate	130		1.84	2.81	2.54
43	Acetic acid,2-phenylethyl ester	164				0.16
Aldehydes
44	Acetaldehyde	44	1.05	0.64	0.42	0.68
45	Propanal,2-methyl-	72		0.05	0.07	0.13
46	3-Methylbutanal	86		0.55	0.39	1.78
47	2-Methylbutanal	86		0.13	0.09	0.92
48	2-Furancarboxaldehyde	96				0.05
49	Propanal,3-methylthio-	104				0.08
50	Benzaldehyde	106				0.21
51	Benzeneacetaldehyde	120				2.56
Alkenes
52	1,1-Dideuterio-2-methyl-1-propene	56	0.05
53	N-n-Butyl-N′-n-pentyl diazene	156	0.03
54	1,3-Diphenyl-1-trimethylsilyloxy-1-pentene	310	0.54			0.07
55	6-[(2-Oxocyclohexyl)methyl]-4-oxa-5-azaspiro[2,4]hept-5-ene	207				0.05
Others
56	(1e)-1,3-Diisopropyl-3-methyltriazene	143	0.13
57	2,3-Epithiopropyl ally ether	130	0.12
58	Dodecahydro-pyrido[1,2-B]isoquinolin-6-one	207	0.19
59	2,3-(Methylenedioxy)-6-((trimethylsilyl)methyl)benzyl iodide	348	0.10
60	Ammonium carbamate	61		0.31
61	Trans-β-ionon-5,6-epoxide	208			0.05	0.06
62	2,3-Bis (methylamino) butanedinitrile	138			0.05
63	Trans-3-chloro-3,6-dimethyltetrahydropyran	148			0.26
64	Acetonitrile-d3	41				0.08
65	(1’s,3s)-3-Hydroxy-3-phenyl-n-(1-phenylethyl)propionamid	269				0.06
66	2-(4-Hydroxy-5-methylhexyl)-1,3-dioxolane	188				0.05

Note: S1 means the final product of 90 d pickling, S 2–4 means 24 h pure cultures of each strain.

Nota: S1 significa el producto final con 90 días en escabeche, S 2–4 significa 24 h de cultivos puros de cada cepa.

Table 5. Volatile composition and content of pickled wax gourd product (S1), individual yeast culture from Issatchenkia orientalis (S2), Pichia kudriavzevii (S3) and Candida xylopsoci (S4).

Tabla 5. Composición de volátiles y contenido del producto calabaza blanca en escabeche (S1), cultivo individual de levadura de Issatchenkia orientalis (S2), Pichia kudriavzevii (S3) y Candida xylopsoci (S4).

Compound	Class and total content (%)
	S1	S2	S3	S4
Acids	(6) 30.96	(2) 0.14	-	(2) 0.62
Alkanes	(5) 10.7	(6) 2.09	(6) 2.3	(5) 2.65
Alcohols	(7) 46.24	(5) 78.18	(7) 78.82	(4) 73.37
Ketones	-	(2) 0.12	(3) 0.25	(3) 1.02
Esters	(4) 9.89	(8) 17.79	(6) 17.3	(5) 15.56
Aldehydes	(1) 1.05	(4) 1.37	(4) 0.97	(8) 6.41
Alkenes	(3) 0.62	-	-	(2) 0.12
Others	(4) 0.54	(1) 0.31	(3) 0.36	(4) 0.25

The acetate ester was detected in all four samples, though they were different in concentration. While three yeast species showed the similar levels of acetate ester, the wax gourd brine showed lower levels. The class of esters were different in wax gourd brine, Issatchenkia orientalis, Pichia kudriavzevii and Candida xylopsoci, it was 4, 8, 6 and 5, respectively. The ester compounds such as methyl acetate, ethanoic acid 2,2,2-trichloroethyl ester and 2-hydroxy-butanoic acid methyl ester were only detected in the wax gourd brine. Propanoic acid, ethyl ester, acetic acid 2-methylpropyl ester and isoamyl acetate were produced by three kinds of yeast species. Ethyl isobutyrate and ethyl butyrate were produced by Issatchenkia orientalis and Pichia kudriavzevii. Acetic acid ethenyl ester and 3-oxo-butanoic acid 2-propenyl ester was only detected in Issatchenkia orientalis. Only acetic acid 2-phenylethyl ester can be informed in the strain of Candida xylopsoci.

To date, little information about the yeast community structure and its population of different species in each period of pickled wax gourd is available. It is our first systematic study of the yeast community variety and volatile composition produced by various predominant yeasts in pickled wax gourd. The results showed that the structure of yeast community was simple, and it contained four genera, which belonged to Issatchenkia, Pichia and Candida. It showed that the predominant yeast species on day 15 were Issatchenkia orientalis and Pichia kudriavzevii. On day 30, Pichia kudriavzevii was still the predominant species, while Issatchenkia orientalis began to decline, and Candida xylopsoci not detected on day 15 began to appear. The community structures on day 60 and 90 were similar, and these predominant yeast species composed of Issatchenkia orientalis, Pichia kudriavzevii and Candida xylopsoci. Issatchenkia orientalis can degrade extracellular malic acid in grape must, which significantly influence the sensory properties and qualities of wines (Gao & Fleet, 1995). Issatchenkia orientalis and Saccharomyces cerevisiae collaborate in grape must can reduce the malic acid content in wine, and the effect was better than that of Issatchenkia orientalis alone (Kim, Hong, & Park, 2008). Pichia kudriavzevii can be used to produce ethanol from 1% sodium hydroxide-treated rice straw (Oberoi et al., 2012), while Pichia kudriavzevii can also be used in alkali- and ozone-treated cotton stalks (Kaur, Oberoi, Bhargav, Sharma-Shivappac, & Dhaliwala, 2012). These characteristics of yeast species more or less affect the quality of the pickled wax gourd product.

RT-PCR has been used to quantify different yeast species in natural fermentative vegetables (Park et al., 2009). There was also someone who conducted RT-PCR for detection and quantification of spoilage yeasts in orange juice (Renard, Di Marco, Egea-Cortines, & Weiss, 2008). By RT-PCR assay, the experiment time can be cut to 4 or 5 h. The bacterial diversity in pickled wax gourd had been investigated in our previous work by RT-PCR method in our laboratory, and RT-PCR profile can provide an indication of microbial biomass of various species in pickled wax gourd reported in a previous study (Zhao et al., 2014). RT-PCR used to quantify the yeast in pickled wax gourd exhibited a complete possibility in this study. It can be seen that the populations of the Issatchenkia orientalis ribosomal reached the values of more than 10⁷ copies mL^–1 in the pickled process, and the Pichia kudriavzevii ribosomal’s number reached the values of more than 10⁷ copies mL^–1. They had the same trend in the pickle process, due to the rapid growth of Candida, the number of the 30th day was lower than that of the 15th day, and then the number increased, at last it was stable at 3.26 × 10⁷ and 2.06 × 10⁸ copies mL^–1, respectively. The number of Candida ribosomal was very little at the beginning of the pickle process, then increased rapidly to more than 10⁷ copies mL^–1, and finally stayed stable at 2.02 × 10⁷ copies mL^–1.

In this study, it has been found that the volatile compounds between yeast culture cultivated in MEB and the pickled wax gourd were different. It can be seen that the yeast species produced the highest amount of alcohols in MEB medium. While in the pickled wax gourd, the amount of acids was higher than that of alcohols. It possibly indicated that these metabolites including alcohols from these special yeast species to facilitate the formation of flavor compounds such as acids, esters, aldehydes, etc. in the pickling processing of wax gourd. It also provided the evidences of aroma profile of pickled wax gourd. In the pickled wax gourd, other microorganisms such as Lactobacillus spp. also existed (Zhao et al., 2014). It can be deduced that they may cooperate to produce different volatile compounds from different GC–MS test results. In raw pickled wax gourd, for the bacterial populations during the initial stage, the lactic acid bacteria was of the highest amount, while at last stage the predominant species were Pediococcus, Weissella and Virgibacillus, which can produce low molecular weight organic acids, such as acetic acid, butyric acid, etc. (Shen et al., 2012). It indicated that the results obtained in this study are consistent with previous reports.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

Authors gratefully acknowledge the financial support of National Natural Science Foundation, China [grant number 31171735], [grant number 31471709] and thank for the support “Sponsored by K. C. Wong Magna Fund in Ningbo University” and Critical Project in Agriculture Field in Ningbo [grant number 2012C10016], China.