Advertisement

Metabolic footprint analysis of volatile metabolites to discriminate between different key time points in the fermentation and storage of starter cultures and probiotic Lactobacillus casei Zhang milk

  • Author Footnotes
    * These authors contributed equally to this work.
    Yaru Sun
    Footnotes
    * These authors contributed equally to this work.
    Affiliations
    Key Laboratory of Dairy Biotechnology and Engineering, Ministry of Education, Inner Mongolia Agricultural University, Hohhot 010018, China

    Key Laboratory of Dairy Products Processing, Ministry of Agriculture and Rural Affairs, Inner Mongolia Agricultural University, Hohhot 010018, China
    Search for articles by this author
  • Author Footnotes
    * These authors contributed equally to this work.
    Chuantao Peng
    Footnotes
    * These authors contributed equally to this work.
    Affiliations
    Key Laboratory of Dairy Biotechnology and Engineering, Ministry of Education, Inner Mongolia Agricultural University, Hohhot 010018, China

    Key Laboratory of Dairy Products Processing, Ministry of Agriculture and Rural Affairs, Inner Mongolia Agricultural University, Hohhot 010018, China
    Search for articles by this author
  • Jicheng Wang
    Affiliations
    Key Laboratory of Dairy Biotechnology and Engineering, Ministry of Education, Inner Mongolia Agricultural University, Hohhot 010018, China

    Key Laboratory of Dairy Products Processing, Ministry of Agriculture and Rural Affairs, Inner Mongolia Agricultural University, Hohhot 010018, China
    Search for articles by this author
  • Haotian Sun
    Affiliations
    Key Laboratory of Dairy Biotechnology and Engineering, Ministry of Education, Inner Mongolia Agricultural University, Hohhot 010018, China

    Key Laboratory of Dairy Products Processing, Ministry of Agriculture and Rural Affairs, Inner Mongolia Agricultural University, Hohhot 010018, China
    Search for articles by this author
  • Shuai Guo
    Affiliations
    Key Laboratory of Dairy Biotechnology and Engineering, Ministry of Education, Inner Mongolia Agricultural University, Hohhot 010018, China

    Key Laboratory of Dairy Products Processing, Ministry of Agriculture and Rural Affairs, Inner Mongolia Agricultural University, Hohhot 010018, China
    Search for articles by this author
  • Heping Zhang
    Correspondence
    Corresponding author
    Affiliations
    Key Laboratory of Dairy Biotechnology and Engineering, Ministry of Education, Inner Mongolia Agricultural University, Hohhot 010018, China

    Key Laboratory of Dairy Products Processing, Ministry of Agriculture and Rural Affairs, Inner Mongolia Agricultural University, Hohhot 010018, China
    Search for articles by this author
  • Author Footnotes
    * These authors contributed equally to this work.
Open ArchivePublished:January 14, 2021DOI:https://doi.org/10.3168/jds.2020-18968

      ABSTRACT

      Interest has been growing in the co-fermentation of starter cultures with probiotic bacteria in milk. However, the representative metabolites and metabolic changes at different key time points during milk fermentation and storage in starter cultures and probiotic bacteria are still unclear. In this study, we used gas chromatography/mass spectrometry–based metabolomics to identify volatile metabolites and discriminate between 6 different time points [fermentation initiation (FI), fermentation curd (FC), fermentation termination (FT), storage 1 d (S1d), storage 7 d (S7d), and storage 14 d (S14d)] during the fermentation and storage of starter cultures and Lactobacillus casei Zhang milk. Of the 52 volatile metabolites identified, 15 contributed to discrimination of the 6 time points. Then, using the profile from the different time points, we analyzed pairwise comparisons (FI vs. FC; FC vs. FT; FT vs. S1d; S1d vs. S7d; S7d vs. S14d); these time-lapse comparisons showed metabolic progressions from one fermentation stage to the next. We found representative and exclusive metabolites at specific fermentation and storage time points. The greatest difference in metabolites occurred between FC and FT, and the metabolic profiles between S7d and S14d were most similar. Interestingly, decanoic acid, octanoic acid, and hexanoic acid reached their highest level at storage 14 d, indicating that the post-fermentation storage of fermented milk with L. casei Zhang may add more probiotic functions. This work provides detailed insight into the time-specific profiles of volatile metabolites and their dynamic changes; these data may be used for understanding and eventually predicting metabolic changes in milk fermentation and storage, where probiotic strains may be used.

      Key words

      INTRODUCTION

      The term “probiotics” first appeared in 1974, and it has evolved to its present common definition as “live microorganisms that confer a health benefit when consumed in adequate amounts” (
      • Hill C.
      • Guarner F.
      • Reid G.
      • Gibson R.
      • Merenstein D.
      • Pot B.
      • Morelli L.
      • Canani R.
      • Flint H.
      • Salminen S.
      • Calder P.
      • Sanders M.
      The international scientific association for probiotics and prebiotics consensus statement on the scope and appropriate use of the term probiotic.
      ). Over the last few decades, the effects of probiotics have been widely reported, including the digestion of complex carbohydrates, the production of vitamins and amino acids, the alleviation of gastrointestinal disease, immune-modulating effects, and the prevention of bacterial infections (
      • Hor Y.Y.
      • Lew L.C.
      • Lau A.S.Y.
      • Ong J.S.
      • Chuah L.O.
      • Lee Y.Y.
      • Choi S.B.
      • Rashid F.
      • Wahid N.
      • Sun Z.H.
      • Kwok L.Y.
      • Zhang H.P.
      • Liong M.T.
      Probiotic Lactobacillus casei Zhang (LCZ) alleviates respiratory, gastrointestinal & RBC abnormality via immuno-modulatory, anti-inflammatory & anti-oxidative actions.
      ;
      • Zhang Y.
      Probiotic effects of Lactobacillus casei Zhang: From single strain omics to metagenomics.
      ). Fermented milk is a good carrier for probiotic bacteria, and the combination of probiotics and fermented milk has better probiotic function (
      • Saxelin M.
      • Lassig A.
      • Karjalainen H.
      • Tynkkynen S.
      • Surakka A.
      • Vapaatalo H.
      • Jarvenpaa S.
      • Korpela R.
      • Mutanen M.
      • Hatakka K.
      Persistence of probiotic strains in the gastrointestinal tract when administered as capsules, yoghurt, or cheese.
      ). In the present study, we used Lactobacillus casei Zhang, a novel probiotic strain isolated from koumiss samples collected in Inner Mongolia, China; it exhibits good fermentation properties and gives yogurt a unique flavor and good texture (
      • Saarela M.
      • Mogensen G.
      • Fonden R.
      • Matto J.
      • Mattila-Sandholm T.
      Probiotic bacteria: Safety, functional and technological properties.
      ;
      • Guo Z.
      • Wang J.C.
      • Yan L.Y.
      • Chen W.
      • Liu X.M.
      • Zhang H.P.
      In vitro comparation of probiotic properties of Lactobacillus casei Zhang, a potential new probiotic, with selected probiotic strains.
      ;
      • Zhang W.
      • Yu D.
      • Sun Z.
      • Wu R.
      • Chen X.
      • Chen W.
      • Meng H.
      • Hu S.
      • Zhang H.
      Complete genome sequence of Lactobacillus casei Zhang, a new probiotic strain isolated from traditional homemade koumiss in inner Mongolia, China.
      ;
      • Wang J.C.
      • Guo Z.
      • Zhang Q.
      • Yan L.
      • Chen Y.
      • Chen X.
      • Liu X.-M.
      • Chen W.
      • Zhang H.-P.
      Effect of probiotic Lactobacillus casei Zhang on fermentation characteristics of set yogurt.
      ).
      More than 100 kinds of volatile metabolites have been identified in yogurt; the main aroma sources are acetaldehyde, diacetyl, 2,3-pentadione, acetone, 2-butanone, and acetic acid (
      • Routray W.
      • Mishra H.N.
      Scientific and technical aspects of yogurt aroma and taste: A review.
      ;
      • Settachaimongkon S.
      • Robert Nout M.J.
      • Antunes Fernandes E.C.
      • Hettinga K.A.
      • Vervoort J.M.
      • van Hooijdonk T.C.M.
      • Zwietering M.H.
      • Smid E.J.
      • van Valenberg H.J.F.
      Influence of different proteolytic strains of Streptococcus thermophilus in co-culture with Lactobacillus delbrueckii ssp. bulgaricus on the metabolite profile of set-yoghurt.
      ). Some research has reported that co-fermentation of starter cultures with probiotic bacteria contributes to the production of desirable volatile metabolites such as acetaldehyde and diacetyl, and it improves the flavor quality of yogurt (
      • Østlie H.M.
      • Helland M.H.
      • Narvhus J.A.
      Growth and metabolism of selected strains of probiotic bacteria in milk.
      ;
      • Li C.
      • Song J.
      • Kwok L.
      • Wang J.
      • Dong Y.
      • Yu H.
      • Hou Q.
      • Zhang H.
      • Chen Y.
      Influence of Lactobacillus plantarum on yogurt fermentation properties and subsequent changes during post-fermentation storage.
      ;
      • Dan T.
      • Chen H.Y.
      • Li T.
      • Tian J.L.
      • Ren W.Y.
      • Zhang H.P.
      • Sun T.S.
      Influence of lactobacillus plantarum p-8 on fermented milk flavor and storage stability.
      ). Probiotic effects and the unique flavor of yogurt fermented with probiotic bacteria is one competitive factor in marketing.
      Metabolomics can be used to qualitatively and quantitatively analyze low molecular weight metabolites (
      • Jewett M.C.
      • Hofmann G.
      • Nielsen J.
      Fungal metabolite analysis in genomics and phenomics.
      ;
      • Wishart D.S.
      Metabolomics: Applications to food science and nutrition research.
      ) and has been applied in food science to monitor metabolic profiles (
      • Mozzi F.
      • Ortiz M.E.
      • Bleckwedel J.
      • De Vuyst L.
      • Pescuma M.
      Metabolomics as a tool for the comprehensive understanding of fermented and functional foods with lactic acid bacteria.
      ). Solid-phase microextraction (SPME) coupled with GC/MS has been used widely to analyze flavor compounds, including those in fermented milk (
      • Pan D.D.
      • Wu Z.
      • Peng T.
      • Zeng X.Q.
      • Li H.
      Volatile organic compounds profile during milk fermentation by Lactobacillus pentosus and correlations between volatiles flavor and carbohydrate metabolism.
      ), fermented soymilk (
      • Yin X.U.
      • Huang Y.J.
      • Chen X.
      • Mao-Lin L.U.
      • Rui-Xia G.U.
      Influence of milk content on flavor compounds in fermented soymilk.
      ), and goat milk cheese (
      • Chiofalo B.
      • Zumbo A.R.
      • Liotta L.
      • Mondello L.
      • Dugo P.
      • Chiofalo V.
      Characterization of Maltese goat milk cheese flavour using SPMEGC/MS.
      ).
      Previous metabolomics studies have shown that the volatile metabolic profiles of yogurt fermented with probiotics differ from those in yogurt without probiotics (
      • Østlie H.M.
      • Helland M.H.
      • Narvhus J.A.
      Growth and metabolism of selected strains of probiotic bacteria in milk.
      ;
      • Li C.
      • Song J.
      • Kwok L.
      • Wang J.
      • Dong Y.
      • Yu H.
      • Hou Q.
      • Zhang H.
      • Chen Y.
      Influence of Lactobacillus plantarum on yogurt fermentation properties and subsequent changes during post-fermentation storage.
      ;
      • Dan T.
      • Chen H.Y.
      • Li T.
      • Tian J.L.
      • Ren W.Y.
      • Zhang H.P.
      • Sun T.S.
      Influence of lactobacillus plantarum p-8 on fermented milk flavor and storage stability.
      ). However, those studies focused on metabolic differences at the end of fermentation rather than during fermentation and storage. Less is known about when specific volatile metabolites are produced. Such knowledge would provide valuable insight into time-specific profiles of volatile metabolites. This information is important because the volatile metabolites formed may contribute directly to the organoleptic quality of the yogurt (
      • Østlie H.M.
      • Treimo J.
      • Narvhus J.A.
      Effect of temperature on growth and metabolism of probiotic bacteria in milk.
      ).
      In the present work, we used GC/MS metabolomics to conduct metabolic footprint analysis and identify volatile metabolites, enabling us to discriminate between 6 time points during the fermentation and storage of starter cultures and L. casei Zhang milk: fermentation initiation (FI), fermentation curd (FC), fermentation termination (FT), storage 1 d (S1d), storage 7 d (S7d), and storage 14 d (S14d).

      MATERIALS AND METHODS

      Strains, Growth Media, and Inoculation Cultures

      We obtained L. casei Zhang from the Lactic Acid Bacteria Collection Center at Inner Mongolia Agricultural University; it was first isolated from koumiss in Xilin Guole of Inner Mongolia in 2002. The L. casei Zhang strains were activated in de Man, Rogosa, and Sharpe (MRS) liquid medium (027312; Huankai Microbial, Guangdong, China) at 37°C for 24 h. Then, cells were collected by centrifugation at 4,000 × g for 15 min at 4°C and resuspended in PBS buffer (0.8% NaCl, 0.02% KH2PO4, 0.115% Na2HPO4, 1% tryptone, and 0.1% sodium glutamate inactivated at 121°C for 15 min).
      Starter cultures Lactobacillus delbrueckii ssp. bulgaricus ND02 (ND02) and Streptococcus thermophilus S10 (S10) were also obtained from the Lactic Acid Bacteria Collection Center of Inner Mongolia Agricultural University. They were activated in MRS and M17 liquid medium (CM0817B; Oxoid, Basingstoke, UK) at 37°C for 24 h. Then, cells were collected by centrifugation at 4,000 × g for 15 min at 4°C and resuspended in PBS buffer (0.8% NaCl, 0.02% KH2PO4, 0.115% Na2HPO4, 1% tryptone, and 0.1% sodium glutamate inactivated at 121°C for 15 min).

      Fermented Milk Manufacture

      Skim milk powder (11.5%) was purchased from NZMP (Wellington, New Zealand). It was stirred and dissolved in distilled water (82%) at 50°C. Then, 7% sucrose was added, mixed well, and hydrated for 30 min. The bottles were sealed with plastic wrap and maintained at 60°C for 15 min. Homogenization was performed twice (65°C and 35 MPa) using a high-pressure homogenizer (Samro Homogenizer, Shanghai, China), and the homogenized milk was pasteurized at 95°C for 5 min. Then, the samples were cooled to the incubation temperature (37°C). The initial viable cell density of L. casei Zhang and starter cultures ND02 and S10 were 1 × 107, 1 × 103, and 1 × 106 cfu/mL, respectively. Milk fermentations were performed in 250 mL blue cap bottles containing 200 mL of milk. Fermentations were carried out under anaerobic conditions at 37°C, without agitation, and all experiments were performed in triplicate. When the pH of the samples reached 4.5 after 8 h, the fermented milks were transferred to 4°C for post-fermentation storage. Fermented milks were sampled at 6 key time points: FI, FC, FT, S1d, S7d, and S14d. Definitions of the time points are as follows: FI was defined as pasteurized milk incubated with starter cultures and L. casei Zhang, but before fermentation; FC was defined as fermented milk coagulation; FT occurred when the pH reached 4.5; and S1d, S7d, and S14d were defined as post-fermentation storage at 4°C for 1 d, 7 d and 14 d, respectively). Samples were stored at −20°C for subsequent analysis by GC/MS. A simplified scheme of the experimental approach is shown in Figure 1.
      Figure thumbnail gr1
      Figure 1Study design. Simplified scheme of the experimental approach used to study the representative metabolites and metabolic changes at 6 key time points [fermentation initiation (FI), fermentation curd (FC), fermentation termination (FT), storage 1 d (S1d), storage 7 d (S7d), and storage 14 d (S14d)] during fermentation and storage of starter cultures and probiotic Lactobacillus casei Zhang (LCZ) milk.

      Solid-Phase Microextraction Sampling Conditions

      Volatile compounds were collected in an SPME system. For each analysis, 20 mL of fermented milk was placed in a 100-mL gas washing flask with a purge head, and 1 μL of 1,2-dichlorobenzene solution (Sigma-Aldrich, St. Louis, MO) was added as internal standard solution. The final concentration of internal standard solution in each sample was 10 μg/L. Under magnetic stirring (200 rpm), the sample was extracted at 50°C for 1 h (
      • Dan T.
      • Wang D.
      • Wu S.
      • Jin R.
      • Ren W.
      • Sun T.
      Profiles of volatile flavor compounds in milk fermented with different proportional combinations of Lactobacillus delbrueckii ssp. bulgaricus and Streptococcus thermophilus..
      ), and an SPME fiber (50/30 μm DVB/Carboxen/PDMS; Supelco, Inc. Bellefonte, PA) was exposed in the headspace. The fiber was then inserted into the injection port of a 7890 B GC (Agilent Technologies Inc., Palo Alto, CA) for 3 min at 270°C to desorb volatile compounds into the GC column.

      Gas Chromatography/Mass Spectrometry

      Volatile metabolites in fermented milk were analyzed using GC/MS. Absorbed volatiles were analyzed using a 7890 B GC equipped with a 5977 A mass-selective detector (Agilent Technologies Inc.) and an HP-5MS column (30 m length, 0.25 mm i.d., 0.25 µm film thickness; Agilent Technologies Inc.;
      • Dan T.
      • Jin R.
      • Ren W.
      • Li T.
      • Chen H.
      • Sun T.
      Characteristics of milk fermented by streptococcus thermophilus mga45–4 and the profiles of associated volatile compounds during fermentation and storage.
      ). Helium was used as the carrier gas at 1 mL/min. The analytical method and parameters were based mainly on those of
      • Dan T.
      • Jin R.
      • Ren W.
      • Li T.
      • Chen H.
      • Sun T.
      Characteristics of milk fermented by streptococcus thermophilus mga45–4 and the profiles of associated volatile compounds during fermentation and storage.
      ,
      • Dan T.
      • Chen H.Y.
      • Li T.
      • Tian J.L.
      • Ren W.Y.
      • Zhang H.P.
      • Sun T.S.
      Influence of lactobacillus plantarum p-8 on fermented milk flavor and storage stability.
      . Volatile compounds were identified by matching of their mass spectra with those of a commercial database (National Institute of Standards Technology Mass Spectral Database, Agilent Technologies Inc.) and confirmed by comparison with the retention indices of authentic reference compounds or retention indices recorded in the literature. All identified compounds were semi-quantified as peak areas in the total ion chromatogram, and relative abundances were standardized by dividing peak areas of volatiles by that of the internal standard from the same sample, and were given as means based on independent duplicates.

      Multivariate Statistical Analysis

      The GC/MS data were normalized by dividing peak areas of volatiles by that of the internal standard from the same sample and were given as means based on independent triplicates. We used Venny (https://bioinfogp.cnb.csic.es/tools/venny/) and principal component analysis and partial least squares discriminant analysis (PLS-DA) to visualize differences in metabolic profiles between samples in Metaboanalyst (www.metaboanalyst.ca); a variable importance on projection (VIP) score is a measure of a variable's importance in the PLS-DA model. For variable selections, volatile metabolites with a VIP greater than 1 were selected as significantly different volatile metabolites.

      RESULTS

      Comparing Metabolic Profiles from 6 Time Points in Fermentation and Post-Fermentation Storage

      We analyzed and compared the metabolic profiles from 6 time points during fermentation and post-fermentation storage using a GC/MS-based metabolomics approach. A total of 52 volatile metabolites were detected in the fermented milk samples, and their identified and quantified information are shown in Supplemental Table S1 (https://doi.org/10.3168/jds.2020-18968). These included 8 aldehydes, 12 ketones, 9 carboxylic acids, 11 alcohols, and 12 aromatic hydrocarbons. Analysis of the volatile metabolites at 6 time points was performed with PLS-DA. We used R2Y (goodness-of-fit) and Q2 (predictive ability parameter) as a means to verify the accuracy and predictability of the PLS-DA model. The R2Y and Q2 values were 0.99 and 0.83, respectively, which were relatively high (close to 1), indicating that the model had good accuracy and predictability. Apart from S7d and S14d, the PLS-DA score plot (Figure 2a) showed that most of the time points (FI, FC, FT, and S1d) were differentiated clearly. This finding indicates that during fermentation and storage the metabolic profiles changed dynamically, and metabolic profiles were similar at S7d and S14d. We next generated a heat map of metabolic profiles for the 6 group samples and analyzed them by hierarchal clustering to visualize time-point-specific metabolic profiles during fermentation and storage (Figure 2b). The cluster analysis also revealed that the 6 group samples presented 6 small clusters, corresponding to the PLS-DA results. The discriminatory volatile metabolites that could be used to discriminate between different time points were analyzed and identified by VIP. Fifteen discriminatory volatiles were identified: pentanoic acid, acetoin, 2-hydroxy-3-pentanone, 2-nonanol, octanoic acid, benzaldehyde, 3-methyl-2-buten-1-ol, 2-decenal, 2,3-butanedione, hexanoic acid, decanoic acid, hexanal, 1-nonanol, 5-methyl-1-hexanol, and 3,7-dimethyl-1,6-octadien-3-ol (Figure 2c).
      Figure thumbnail gr2
      Figure 2Metabolic profiles at 6 time points during fermentation and post-fermentation storage. (a) Analysis of volatile metabolites at 6 time points [fermentation initiation (FI), fermentation curd (FC), fermentation termination (FT), storage 1 d (S1d), storage 7 d (S7d), and storage 14 d (S14d)] was performed using partial least squares discriminant analysis. (b) Heat map and hierarchal clustering of metabolic profiles of the 6 samples. (c) Volatile metabolites that can be used to discriminate between different time points were analyzed and identified by variable importance in projection (VIP).

      Metabolic Footprint of the Differences Between FI and FC

      We compared the volatile metabolites from FI to those of FC using Venny and PLS-DA according to our criteria in the multivariate statistical analysis described above; the metabolic footprint of the differences is shown in Figure 3. Fourteen volatiles appeared only at FI: nonanal, hexanal, heptanal, 6-methyl-5-hepten-2-one, decanal, 2-pyridinecarboxylic acid, (1-butylpentyl)-benzene, dimethyl sulfone, 3-hydroxybutanal, methanesulfonylacetic acid, dodecane, 2,4-dimethyl-decane, decanoic acid, and 2,4,6-trimethyl-decane. Eight volatiles appeared only at FC: 2,3-butanedione, acetoin, 2,3-pentanedione, hexanoic acid, octanoic acid, 1-ethyl-butyl-hydroperoxide, 1-heptanol, and 1-hexanol. Sixteen volatiles were common to FI and FC, and 6 of those compounds were identified as significantly different volatiles using VIP. In the PLS-DA model, the R2Y and Q2 values were 0.99 and 0.53, respectively. These 6 compounds were 1-nonanol, 1-octanol, 2-undecanone, 2-nonanone, 3-methyl-2-butanone, and 2-heptanone.
      Figure thumbnail gr3
      Figure 3Metabolic footprint of the differences between fermentation initiation (FI) and fermentation curd (FC). Comparison of the volatile metabolites from FI and FC using Venny (https://bioinfogp.cnb.csic.es/tools/venny/) and partial least squares discriminant analysis. VIP = variable importance in projection.

      Metabolic Footprint of the Differences Between FC and FT

      We compared the volatile metabolites from FC to those of FT using Venny and PLS-DA according to our criteria in the multivariate statistical analysis described above; the metabolic footprint of the differences is shown in Figure 4. Eight volatiles appeared only at FC: 3-methyl-2-butanone, 1-nonanol, benzaldehyde, tetradecane, 1-ethyl-butyl hydroperoxide, d-limonene, 3,7-dimethyl-1,6-octadien-3-ol, and 1,3,5-cycloheptatriene. Nineteen volatiles appeared only at FT: ethylene oxide, acetic acid, acetaldehyde, pentanoic acid, 1-(ethenylthio)-butane, 2-hydroxy-3-pentanone, 2-oxopropanoic acid, decanoic acid, dodecane, heptadecane, 2-pyridinecarboxylic acid, nonanal, (E)-2-decenal, 3-hydroxybutanal, dimethyl sulfone, 1-octen-3-ol, 2,4-dimethyl-decane, 3-methyl-2-buten-1-ol, and decanal. Sixteen volatiles were common to FC and FT, and 5 of those compounds were identified as significantly different volatiles using VIP. In the PLS-DA model, the R2Y and Q2 values were 0.98 and 0.85, respectively. These 5 compounds were hexanoic acid, octanoic acid, butylated hydroxytoluene, 2,3-butanedione, and 3-decyn-2-ol.
      Figure thumbnail gr4
      Figure 4Metabolic footprint of the differences between fermentation curd (FC) and fermentation termination (FT). Comparison of the volatile metabolites from FC and FT using Venny (https://bioinfogp.cnb.csic.es/tools/venny/) and partial least squares discriminant analysis. VIP = variable importance in projection.

      Metabolic Footprints of the Differences Between FT and S1d

      We compared the volatile metabolites from FT to those of S1d using Venny and PLS-DA according to our criteria in the multivariate statistical analysis described above; the metabolic footprint of the differences is shown in Figure 5. Eleven volatiles appeared only at FT: 2-oxopropanoic acid, dodecane, 2-pyridinecarboxylic acid, nonanal, butylated hydroxytoluene, 3-hydroxybutanal, 1-octen-3-ol, 2,4-dimethyl-decane, 3-methyl-2-buten-1-ol, p-xylene, and decanal. Four volatiles appeared only at S1d: 3-methyl-2-butanone, benzaldehyde, 2-nonanol, and benzoic acid. Twenty-four volatiles were common to FT and S1d, and 8 of those compounds were identified as significantly different volatiles using VIP. In the PLS-DA model, the R2Y and Q2 values were 0.84 and 0.56, respectively. These 8 compounds were acetaldehyde, 1-(ethenylthio)-butane, 3-decyn-2-ol, octanoic acid, 1-heptanol, 2,3-butanedione, hexanoic acid, and decanoic acid.
      Figure thumbnail gr5
      Figure 5Metabolic footprint of the differences between fermentation termination (FT) and storage 1 d (S1d). Comparison of the volatile metabolites from FT and S1d using Venny (https://bioinfogp.cnb.csic.es/tools/venny/) and partial least squares discriminant analysis. VIP = variable importance in projection.

      Metabolic Footprints of the Differences Between S1d and S7d

      We compared the volatile metabolites from S1d to those of S7d using Venny and PLS-DA according to our criteria in the multivariate statistical analysis described above; the metabolic footprint of the differences is shown in Figure 6a. Four volatiles appeared only at S1d: ethylene oxide, benzaldehyde, dimethyl sulfone, and 1-(ethenylthio)-butane. Thirteen volatiles appeared only at S7d: dimethylamine, 2-oxopropanoic acid, 1-octen-3-ol, 2-butyl-1-octanol, 3-methyl-2-buten-1-ol, nonanal, p-xylene, butylated hydroxytoluene, 3-hydroxybutanal, decanal, methanesulfonylacetic acid, heptanal, and 1,3,5-cycloheptatriene. Twenty-four volatiles were common to S1d and S7d, but none of those compounds were identified as significantly different volatiles.
      Figure thumbnail gr6
      Figure 6Metabolic footprints of the differences between storage times. Comparison of the volatile metabolites (a) from storage 1 d (S1d) and storage 7 d (S7d); and (b) from S7d and storage 14 d (S14d) using Venny (https://bioinfogp.cnb.csic.es/tools/venny/) and partial least squares discriminant analysis.

      Metabolic Footprints of Differences Between Storage 7 d and Storage 14 d

      We compared the volatile metabolites from S7d to those of S14d using Venny and PLS-DA according to our criteria in the multivariate statistical analysis described above; the metabolic footprint of the differences is shown in Figure 6b. Five volatiles appeared only at S7d: 2-oxopropanoic acid, heptadecane, benzoic acid, 1-heptanol, and p-xylene. Five volatiles appeared only at S14d: ethylene oxide, 1-(ethenylthio)-butane, 5-methyl-1-hexanol, tetradecane, and d-limonene. Thirty-two volatiles were common to S7d and S14d, but none of those compounds were identified as significantly different volatiles.

      DISCUSSION

      To the best of our knowledge, this study is the first of its kind to investigate metabolic changes at different time points during the fermentation and storage of starter cultures and L. casei Zhang milk using GC/MS-based metabolomics. By comparing the volatile metabolites from different time points, we observed a different metabolic profile.
      Although co-fermentation with S. thermophilus and L. bulgaricus forms the basis for the metabolic profiles of fermented milk (
      • Sieuwerts S.
      • Molenaar D.
      • van Hijum S.A.F.T.
      • Beerthuyzen M.
      • Stevens M.J.A.
      • Janssen P.W.M.
      • Ingham C.J.
      • de Bok F.A.M.
      • de Vos W.M.
      • van Hylckama Vlieg J.E.T.
      Mixed-culture transcriptome analysis reveals the molecular basis of mixed-culture growth in streptococcus thermophilus and lactobacillus bulgaricus..
      ), findings indicate that probiotic strains also contribute to the distinct flavor of yogurt, as well as its probiotic properties, as described by many authors (
      • Pan D.D.
      • Wu Z.
      • Peng T.
      • Zeng X.Q.
      • Li H.
      Volatile organic compounds profile during milk fermentation by Lactobacillus pentosus and correlations between volatiles flavor and carbohydrate metabolism.
      ;
      • Li C.
      • Song J.
      • Kwok L.
      • Wang J.
      • Dong Y.
      • Yu H.
      • Hou Q.
      • Zhang H.
      • Chen Y.
      Influence of Lactobacillus plantarum on yogurt fermentation properties and subsequent changes during post-fermentation storage.
      ;
      • Dan T.
      • Chen H.Y.
      • Li T.
      • Tian J.L.
      • Ren W.Y.
      • Zhang H.P.
      • Sun T.S.
      Influence of lactobacillus plantarum p-8 on fermented milk flavor and storage stability.
      ). More importantly, probiotic strains influence the formation of volatile metabolites not only during fermentation, but also during storage (
      • Zareba D.
      • Ziarno M.
      • Obiedzinski M.
      Volatile profile of non-fermented milk and milk fermented by Bifidobacterium animalis subsp. lactis.
      ). The metabolic characteristics of milk fermented with specific probiotics strains at specific fermentation and storage time points are much more reflective of the flavor qualities of the fermented milk products. In the present work, we compared the metabolic profiles at different time points during fermentation and storage of starter cultures and L. casei Zhang milk, reflecting the metabolic status and organoleptic characteristics at each time point.
      In accordance with the time of milk fermentation and storage, the time-lapse series of pairwise comparisons (FI vs. FC; FC vs. FT; FT vs. S1d; S1d vs. S7d; S7d vs. S14d) showed different metabolic progressions from one fermentation stage to the next. The greatest difference in metabolites occurred between FC and FT, and the metabolic profiles between S7d and S14d were the most similar. We identified the representative and exclusive metabolites at each fermentation and storage time point, and this information provided detailed insights into time-specific volatile metabolites and dynamic changes in volatile metabolites. To the best of our knowledge, the known metabolic pathways and specific contributions to fermented milks have not been reported in previous researches for some discriminatory volatile metabolites we identified; these metabolites need to be further studied. Among the volatile metabolites used to discriminate between stages, 10 that commonly exist in fermented milk have known metabolic pathways in lactic acid bacteria and make great contributions to the quality and organoleptic characteristics of fermented milk. These 10 metabolites are 2,3-butanedione, acetoin, 2-pentanone, 2-heptanone, acetaldehyde, 1-heptanol, 1-octanol, hexanoic acid, decanoic acid, and octanoic acid. We have provided the known metabolic pathways of these metabolites (Figure 7) and discussed them more deeply below.
      Figure thumbnail gr7
      Figure 7Metabolic pathways. The known metabolic pathways of the 10 volatile metabolites in the present study. FC = fermentation curd; FI = fermentation initiation; FT = fermentation termination; S1d = storage 1 d; S7d = storage 7 d; S14d = storage 14 d.
      Aldehydes play an important role in the formation of yogurt flavor and have a lower threshold (
      • Dan T.
      • Chen H.Y.
      • Li T.
      • Tian J.L.
      • Ren W.Y.
      • Zhang H.P.
      • Sun T.S.
      Influence of lactobacillus plantarum p-8 on fermented milk flavor and storage stability.
      ). Among the discriminatory aldehydes, the metabolic pathway of acetaldehyde has been described in lactic acid bacteria (
      • Settachaimongkon S.
      • Nout M.J.R.
      • Antunes Fernandes E.C.
      • Van Hooijdonk T.C.M.
      • Zwietering M.H.
      • Smid E.J.
      • Van Valenberg H.J.F.
      The impact of selected strains of probiotic bacteria on metabolite formation in set yoghurt.
      ). Acetaldehyde is one of the most important flavor compounds and gives yogurt a green apple flavor (
      • Cheng H.
      Volatile flavor compounds in yogurt: A review.
      ). Acetaldehyde in lactic acid bacteria can be formed through several metabolic pathways (
      • Liu W.J.
      Characteristics of acid and flavor-producing Streptococcus thermophilus and Lactobacillus bulgaricus, as well as their functional gene typing and expression. PhD Thesis.
      ); for example, it can be produced by dehydrogenation of ethanol via the activity of alcohol dehydrogenase or by decarboxylation of pyruvate. During milk fermentation, the production of acetaldehyde is bound up with pH and acetaldehyde is generally produced when the pH of the fermented milk drops to 5.0 (
      • Baranowska M.
      Intensification of the synthesis of flavor compounds in yogurt by milk enrichment with their precursors.
      ). In the present study, acetaldehyde was produced at FT and reached a peak at S14d, possibly because of increased acidity, which indicates that the fermentation environment may influence acetaldehyde content. The yield of acetaldehyde is related to the starter cultures used in milk fermentation (
      • Bongers R.S.
      • Hoefnagel M.H.N.
      • Kleerebezem M.
      High-level acetaldehyde production in lactococcus lactis by metabolic engineering.
      ). Previous studies have reported that L. bulgaricus and S. thermophilus establish a good symbiotic relationship, which can increase the content of acetaldehyde (
      • Wang J.C.
      • Guo Z.
      • Qi-Mu G.S.D.
      • Zang H.P.
      • Meng-He B.L.G.
      Effects of probiotic Lactobacillus casei Zhang on sensory properties of set fermented milk.
      ;
      • Zhang L.
      Application of Streptococcus thermophilus with high-yield acetaldehyde in fermented milk. MS Thesis.
      ;
      • Wang D.
      Analysis of volatile flavor compounds in fermented milk and research for its chromatographic fingerprints. MS Thesis.
      ). Together, the increased acidity and the combination of L. casei Zhang and S. thermophilus may have contributed to higher concentrations of acetaldehyde at storage 14 d.
      Ketones are mainly produced by β-oxidation, amino acid degradation, or degradation of unsaturated fatty acids (
      • Cheng H.
      Volatile flavor compounds in yogurt: A review.
      ). Among the discriminatory ketones, the metabolic pathways of 2,3-butanedione, 2-pentanone, 2-heptanone, and acetoin are described in lactic acid bacteria (
      • Chu F.L.
      • Yaylayan V.A.
      Model studies on the oxygen-induced formation of benzaldehyde from phenylacetaldehyde using pyrolysis GC-MS and FTIR.
      ;
      • Maria K.
      • Dimitris K.
      • Ioulia S.
      Enzymatic total synthesis of banana volatile (S)-2-pentyl (R)-3-hydroxyhexanoate.
      ;
      • Settachaimongkon S.
      • Nout M.J.R.
      • Antunes Fernandes E.C.
      • Van Hooijdonk T.C.M.
      • Zwietering M.H.
      • Smid E.J.
      • Van Valenberg H.J.F.
      The impact of selected strains of probiotic bacteria on metabolite formation in set yoghurt.
      ;
      • Zhang Y.
      • Wang L.F.
      • Zhang J.C.
      • Li Y.X.
      • He Q.W.
      • Li H.
      • Guo X.
      • Guo J.L.
      • Zhang H.P.
      Probiotic Lactobacillus casei Zhang ameliorates high-fructose-induced impaired glucose tolerance in hyperinsulinemia rats.
      ). The ketone 2,3-butanedione, also known as diacetyl, can give yogurt a buttery aroma (
      • Wang J.C.
      • Guo Z.
      • Qi-Mu G.S.D.
      • Zang H.P.
      • Meng-He B.L.G.
      Effects of probiotic Lactobacillus casei Zhang on sensory properties of set fermented milk.
      ). There are many ways to produce 2,3-butanedione, and it is widely accepted that the active acetaldehyde and pyruvate can synthesize α-acetolactic acid under the action α-acetolactate synthase, and then produce 2,3-butanedione by chemical oxidation decarboxylation (
      • Ramos A.
      • Jordan K.N.
      • Cogan T.M.
      • Santos H.
      13C nuclear magnetic resonance studies of citrate and glucose cometabolism by Lactococcus lactis..
      ;
      • de Vos W.M.
      Metabolic engineering of sugar catabolism in lactic acid bacteria.
      ). In our study, 2,3-butanedione began to appear at the FC point; the content of 2,3-butanedione increased continuously during storage and reached a peak at S14d. Some research has reported that the concentration of 2,3-butanedione may be species-specific, and its levels increased during storage (
      • Wang H.Y.
      • Li L.H.
      • Lu C.
      • Kang Z.Y.
      • Zhu H.
      Study on the effect of aldehydes and diacetyls to the flavor in fermented milk.
      ;
      • Wang X.N.
      Selection of Streptococcus thermophilus with high production of diacetyl and expression of the related functional genes. MS Thesis.
      ). Acetoin is converted from 2,3-butanedione by the enzyme diacetyl reductase (
      • Comasio A.
      • Harth H.
      • Weckx S.
      • Vuyst L.D.
      The addition of citrate stimulates the production of acetoin and diacetyl by a citrate-positive lactobacillus crustorum strain during wheat sourdough fermentation.
      ), which gives yogurt sweet, cultured, and buttery aromas (
      • Cheng H.
      Volatile flavor compounds in yogurt: A review.
      ). In the present study, acetoin began to appear at FT and increased during storage. The accumulation of 2,3-butanedione after FC may have contributed to the increase in acetoin, so it makes senses that the changes in these 2 volatile compounds were consistent. Some research has reported that L. casei can contribute to the production of 2,3-butanedione and acetoin, and other studies have shown that the addition of L. casei Zhang can significantly increase the content of 2,3-butanedione and acetoin during storage (
      • Wang J.C.
      • Guo Z.
      • Qi-Mu G.S.D.
      • Zang H.P.
      • Meng-He B.L.G.
      Effects of probiotic Lactobacillus casei Zhang on sensory properties of set fermented milk.
      ;
      • Hao W.B.
      Metabolic regulation and fermentation optimization of diacetyl production in Bacillus sp. DL01. PhD Thesis.
      ). The results of the present study also confirm this fact.
      Important methyl ketones are 2-heptenone and 2-pentanone, which give yogurt a fruity flavor. In our study, 2-heptenone and 2-pentanone were produced at FC and increased continuously during storage, reaching a peak at S14d. Other studies found that adding Lactobacillus helveticus H9 to fermented milk significantly increased the content of 2-heptanone after storage for 21 d (
      • Zhou T.
      • Huo R.
      • Kwok L.
      • Li C.
      • Ma Y.
      • Mi Z.
      • Chen Y.
      Effects of applying Lactobacillus helveticus H9 as adjunct starter culture in yogurt fermentation and storage.
      ). Levels of 2-pentanone tend to be relatively low, and its formation is related to the probiotic strain (
      • Hannon J.A.
      • Kilcawley K.N.
      • Wilkinson M.G.
      • Delahunty C.M.
      • Beresford T.P.
      Flavour precursor development in cheddar cheese due to lactococcal starters and the presence and lysis of lactobacillus helveticus.
      ). The results of the present study indicate that the addition of L. casei Zhang to fermented milk may also increase the content of 2-heptenone and 2-pentanone, especially during the storage stages.
      During milk fermentation, bacteria can produce different acids by degrading proteins, fats, and sugars; for example, lactic acid is generated by metabolizing lactose and volatile fatty acids with chain lengths between C4 and C10 from lipolysis (
      • Güler Z.
      Changes in salted yoghurt during storage.
      ;
      • Liu W.J.
      Characteristics of acid and flavor-producing Streptococcus thermophilus and Lactobacillus bulgaricus, as well as their functional gene typing and expression. PhD Thesis.
      ). Some research has reported that probiotics can change the fatty acid profile of fermented milk by forming biologically active fatty acids during fermentation (
      • Ekinci F.Y.
      • Okur O.D.
      • Ertekin B.
      • Guzel-Seydim Z.
      Effects of probiotic bacteria and oils on fatty acid profiles of cultured cream.
      ). Some studies have shown that L. casei Zhang can increase the content of hexanoic acid in fermented milk during storage (
      • Wang J.C.
      • Guo Z.
      • Qi-Mu G.S.D.
      • Zang H.P.
      • Meng-He B.L.G.
      Effects of probiotic Lactobacillus casei Zhang on sensory properties of set fermented milk.
      ). In the present study, decanoic acid, octanoic acid, and hexanoic acid all reached their highest levels at S14d, indicating that, decanoic acid and octanoic acid can also be increased by L. casei Zhang during storage. This finding may be related to the good proteolytic and lipolytic ability of L. casei Zhang (
      • Zhong Z.
      • Zhang W.Y.
      • Du R.T.
      • Meng H.
      • Zhang H.P.
      Lactobacillus casei Zhang stimulates lipid metabolism in hypercholesterolemic rats by affecting gene expression in the live.
      ). An increasing number of studies have shown that fermented milk has some probiotic properties—for example helping alleviate nutrition malabsorption syndrome, small intestine dysfunction, and hereditary pancreatic diseases—and these probiotic functions are closely related to hexanoic acid, octanoic acid, decanoic acid, and other short-, medium-, long-chain fatty acids (
      • Liu C.
      • Xu X.
      • Shi Y.
      • Wang C.
      Nutritional value and current research status of goat milk.
      ). These results may indicate that the post-fermentation storage of fermented milk and the addition of probiotic strain L. casei Zhang may increase the probiotic functions of fermented dairy products.
      The formation of higher alcohols in fermented milk is associated with lactose metabolism, methyl ketone reduction, and amino acid metabolism (
      • Wang D.
      Analysis of volatile flavor compounds in fermented milk and research for its chromatographic fingerprints. MS Thesis.
      ). The threshold of higher alcohols is generally higher and thus has little effect on the flavor of fermented milk (
      • Dan T.
      • Chen H.Y.
      • Li T.
      • Tian J.L.
      • Ren W.Y.
      • Zhang H.P.
      • Sun T.S.
      Influence of lactobacillus plantarum p-8 on fermented milk flavor and storage stability.
      ). We know that 1-heptanol and 1-octanol commonly exist in fermented milks (
      • Dan T.
      • Chen H.Y.
      • Li T.
      • Tian J.L.
      • Ren W.Y.
      • Zhang H.P.
      • Sun T.S.
      Influence of lactobacillus plantarum p-8 on fermented milk flavor and storage stability.
      ;
      • Nissen L.
      • Demircan B.
      • Taneyo-Saa D.L.
      • Gianotti A.
      Shift of aromatic profile in probiotic hemp drink formulations: A metabolomic approach.
      ;
      • Sarhir S.T.
      • Amanpour A.
      • Bouseta A.
      • Selli S.
      Key odorants of a moroccan fermented milk product “lben” using aroma extract dilution analysis.
      ), and these 2 higher alcohols were significantly different volatile metabolites in our study. Indeed, 1-heptanol began to appear at FC and reached a peak at S7d; 1-octanol was detected at FI and reached a peak at S14d.
      Interestingly, these 10 metabolites all reached a peak at S14d, indicating that L. casei Zhang may contribute to the improvement of flavor quality as storage progresses.

      CONCLUSIONS

      The present work provides detailed insight into time-specific volatile metabolites and their dynamic changes at different time points during the fermentation and storage of starter cultures and L. casei Zhang milk using a GC/MS-based metabolomics approach. Of 52 volatile metabolites identified, 15 contributed to the discrimination between these 6 time points. Our time-lapse comparisons (FI vs. FC; FC vs. FT; FT vs. S1d; S1d vs. S7d; S7d vs. S14d) showed different metabolic progressions from one fermentation stage to the next. We found representative and exclusive metabolites at specific fermentation and storage time points. The biggest difference in metabolites occurred between FC and FT, and the metabolic profiles between S7d and S14d were the most similar. Interestingly, decanoic acid, octanoic acid, and hexanoic acid reached their highest levels at 14 d of storage, indicating that the post-fermentation storage of fermented milk with L. casei Zhang may increase probiotic functions.

      ACKNOWLEDGMENTS

      This study was funded by the China Agriculture Research System (CARS-36) and Inner Mongolia Science & Technology Major Projects (zdzx2018018). The authors declare no competing financial or other conflicts of interest.

      Supplementary Material

      REFERENCES

        • Comasio A.
        • Harth H.
        • Weckx S.
        • Vuyst L.D.
        The addition of citrate stimulates the production of acetoin and diacetyl by a citrate-positive lactobacillus crustorum strain during wheat sourdough fermentation.
        Int. J. Food Microbiol. 2019; 289 (30218873): 88-105
        • Baranowska M.
        Intensification of the synthesis of flavor compounds in yogurt by milk enrichment with their precursors.
        Pol. J. Food Nutr. Sci. 2006; 15: 5-11
        • Bongers R.S.
        • Hoefnagel M.H.N.
        • Kleerebezem M.
        High-level acetaldehyde production in lactococcus lactis by metabolic engineering.
        Appl. Environ. Microbiol. 2005; 71 (15691976): 1109-1113
        • Cheng H.
        Volatile flavor compounds in yogurt: A review.
        Crit. Rev. Food Sci. Nutr. 2010; 50 (21108074): 938-950
        • Chiofalo B.
        • Zumbo A.R.
        • Liotta L.
        • Mondello L.
        • Dugo P.
        • Chiofalo V.
        Characterization of Maltese goat milk cheese flavour using SPMEGC/MS.
        S. Afr. J. Anim. Sci. 2004; 34: 176-180
        • Chu F.L.
        • Yaylayan V.A.
        Model studies on the oxygen-induced formation of benzaldehyde from phenylacetaldehyde using pyrolysis GC-MS and FTIR.
        J. Agric. Food Chem. 2008; 56 (18954073): 10697-10704
        • Dan T.
        • Chen H.Y.
        • Li T.
        • Tian J.L.
        • Ren W.Y.
        • Zhang H.P.
        • Sun T.S.
        Influence of lactobacillus plantarum p-8 on fermented milk flavor and storage stability.
        Front. Microbiol. 2019; 9 (30687239)3133
        • Dan T.
        • Jin R.
        • Ren W.
        • Li T.
        • Chen H.
        • Sun T.
        Characteristics of milk fermented by streptococcus thermophilus mga45–4 and the profiles of associated volatile compounds during fermentation and storage.
        Molecules. 2018; 23 (29641460): 878
        • Dan T.
        • Wang D.
        • Wu S.
        • Jin R.
        • Ren W.
        • Sun T.
        Profiles of volatile flavor compounds in milk fermented with different proportional combinations of Lactobacillus delbrueckii ssp. bulgaricus and Streptococcus thermophilus..
        Molecules. 2017; 22 (28961194)1633
        • de Vos W.M.
        Metabolic engineering of sugar catabolism in lactic acid bacteria.
        Antonie van Leeuwenhoek. 1996; 70 (8879408): 223-242
        • Ekinci F.Y.
        • Okur O.D.
        • Ertekin B.
        • Guzel-Seydim Z.
        Effects of probiotic bacteria and oils on fatty acid profiles of cultured cream.
        Eur. J. Lipid Sci. Technol. 2008; 110: 216-224
        • Güler Z.
        Changes in salted yoghurt during storage.
        Int. J. Food Sci. Technol. 2007; 42: 235-245
        • Guo Z.
        • Wang J.C.
        • Yan L.Y.
        • Chen W.
        • Liu X.M.
        • Zhang H.P.
        In vitro comparation of probiotic properties of Lactobacillus casei Zhang, a potential new probiotic, with selected probiotic strains.
        Lebensm. Wiss. Technol. 2009; 42: 1640-1646
        • Hannon J.A.
        • Kilcawley K.N.
        • Wilkinson M.G.
        • Delahunty C.M.
        • Beresford T.P.
        Flavour precursor development in cheddar cheese due to lactococcal starters and the presence and lysis of lactobacillus helveticus.
        Int. Dairy J. 2007; 17: 316-327
        • Hao W.B.
        Metabolic regulation and fermentation optimization of diacetyl production in Bacillus sp. DL01. PhD Thesis.
        Inner Mongolia Agricultural University, Hohhot2016
        • Hill C.
        • Guarner F.
        • Reid G.
        • Gibson R.
        • Merenstein D.
        • Pot B.
        • Morelli L.
        • Canani R.
        • Flint H.
        • Salminen S.
        • Calder P.
        • Sanders M.
        The international scientific association for probiotics and prebiotics consensus statement on the scope and appropriate use of the term probiotic.
        Nat. Rev. Gastroenterol. Hepatol. 2014; 11 (24912386): 506-514
        • Hor Y.Y.
        • Lew L.C.
        • Lau A.S.Y.
        • Ong J.S.
        • Chuah L.O.
        • Lee Y.Y.
        • Choi S.B.
        • Rashid F.
        • Wahid N.
        • Sun Z.H.
        • Kwok L.Y.
        • Zhang H.P.
        • Liong M.T.
        Probiotic Lactobacillus casei Zhang (LCZ) alleviates respiratory, gastrointestinal & RBC abnormality via immuno-modulatory, anti-inflammatory & anti-oxidative actions.
        J. Funct. Foods. 2018; 44: 235-245
        • Jewett M.C.
        • Hofmann G.
        • Nielsen J.
        Fungal metabolite analysis in genomics and phenomics.
        Curr. Opin. Biotechnol. 2006; 17 (16488600): 191-197
        • Li C.
        • Song J.
        • Kwok L.
        • Wang J.
        • Dong Y.
        • Yu H.
        • Hou Q.
        • Zhang H.
        • Chen Y.
        Influence of Lactobacillus plantarum on yogurt fermentation properties and subsequent changes during post-fermentation storage.
        J. Dairy Sci. 2017; 100 (28215898): 2512-2525
        • Liu C.
        • Xu X.
        • Shi Y.
        • Wang C.
        Nutritional value and current research status of goat milk.
        J. Food Sci. Technol. 2013; 36: 25-28
        • Liu W.J.
        Characteristics of acid and flavor-producing Streptococcus thermophilus and Lactobacillus bulgaricus, as well as their functional gene typing and expression. PhD Thesis.
        Inner Mongolia Agricultural University, Hohhot2014
        • Maria K.
        • Dimitris K.
        • Ioulia S.
        Enzymatic total synthesis of banana volatile (S)-2-pentyl (R)-3-hydroxyhexanoate.
        Eur. J. Org. Chem. 2011; : 3946-3950
        • Mozzi F.
        • Ortiz M.E.
        • Bleckwedel J.
        • De Vuyst L.
        • Pescuma M.
        Metabolomics as a tool for the comprehensive understanding of fermented and functional foods with lactic acid bacteria.
        Food Res. Int. 2013; 54: 1152-1161
        • Nissen L.
        • Demircan B.
        • Taneyo-Saa D.L.
        • Gianotti A.
        Shift of aromatic profile in probiotic hemp drink formulations: A metabolomic approach.
        Microorganisms. 2019; 7 (31671881): 509
        • Østlie H.M.
        • Helland M.H.
        • Narvhus J.A.
        Growth and metabolism of selected strains of probiotic bacteria in milk.
        Int. J. Food Microbiol. 2003; 87 (12927703): 17-27
        • Østlie H.M.
        • Treimo J.
        • Narvhus J.A.
        Effect of temperature on growth and metabolism of probiotic bacteria in milk.
        Int. Dairy J. 2005; 15: 989-997
        • Pan D.D.
        • Wu Z.
        • Peng T.
        • Zeng X.Q.
        • Li H.
        Volatile organic compounds profile during milk fermentation by Lactobacillus pentosus and correlations between volatiles flavor and carbohydrate metabolism.
        J. Dairy Sci. 2014; 97 (24359834): 624-631
        • Ramos A.
        • Jordan K.N.
        • Cogan T.M.
        • Santos H.
        13C nuclear magnetic resonance studies of citrate and glucose cometabolism by Lactococcus lactis..
        Appl. Environ. Microbiol. 1994; 60 (16349269): 1739-1748
        • Routray W.
        • Mishra H.N.
        Scientific and technical aspects of yogurt aroma and taste: A review.
        Compr. Rev. Food Sci. Food Saf. 2011; 10: 208-220
        • Saarela M.
        • Mogensen G.
        • Fonden R.
        • Matto J.
        • Mattila-Sandholm T.
        Probiotic bacteria: Safety, functional and technological properties.
        J. Biotechnol. 2000; 84 (11164262): 197-215
        • Sarhir S.T.
        • Amanpour A.
        • Bouseta A.
        • Selli S.
        Key odorants of a moroccan fermented milk product “lben” using aroma extract dilution analysis.
        J. Food Sci. Technol. 2019; 56 (31413409): 3836-3845
        • Saxelin M.
        • Lassig A.
        • Karjalainen H.
        • Tynkkynen S.
        • Surakka A.
        • Vapaatalo H.
        • Jarvenpaa S.
        • Korpela R.
        • Mutanen M.
        • Hatakka K.
        Persistence of probiotic strains in the gastrointestinal tract when administered as capsules, yoghurt, or cheese.
        Int. J. Food Microbiol. 2010; 144 (21074284): 293-300
        • Settachaimongkon S.
        • Robert Nout M.J.
        • Antunes Fernandes E.C.
        • Hettinga K.A.
        • Vervoort J.M.
        • van Hooijdonk T.C.M.
        • Zwietering M.H.
        • Smid E.J.
        • van Valenberg H.J.F.
        Influence of different proteolytic strains of Streptococcus thermophilus in co-culture with Lactobacillus delbrueckii ssp. bulgaricus on the metabolite profile of set-yoghurt.
        Int. J. Food Microbiol. 2014; 177 (24598513): 29-36
        • Settachaimongkon S.
        • Nout M.J.R.
        • Antunes Fernandes E.C.
        • Van Hooijdonk T.C.M.
        • Zwietering M.H.
        • Smid E.J.
        • Van Valenberg H.J.F.
        The impact of selected strains of probiotic bacteria on metabolite formation in set yoghurt.
        Int. Dairy J. 2014; 38: 1-10
        • Sieuwerts S.
        • Molenaar D.
        • van Hijum S.A.F.T.
        • Beerthuyzen M.
        • Stevens M.J.A.
        • Janssen P.W.M.
        • Ingham C.J.
        • de Bok F.A.M.
        • de Vos W.M.
        • van Hylckama Vlieg J.E.T.
        Mixed-culture transcriptome analysis reveals the molecular basis of mixed-culture growth in streptococcus thermophilus and lactobacillus bulgaricus..
        Appl. Environ. Microbiol. 2010; 76 (20889781): 7775-7784
        • Wang D.
        Analysis of volatile flavor compounds in fermented milk and research for its chromatographic fingerprints. MS Thesis.
        Inner Mongolia Agricultural University, Hohhot2017
        • Wang H.Y.
        • Li L.H.
        • Lu C.
        • Kang Z.Y.
        • Zhu H.
        Study on the effect of aldehydes and diacetyls to the flavor in fermented milk.
        Zhongguo Rupin Gongye. 2010; 10: 32-34
        • Wang J.C.
        • Guo Z.
        • Qi-Mu G.S.D.
        • Zang H.P.
        • Meng-He B.L.G.
        Effects of probiotic Lactobacillus casei Zhang on sensory properties of set fermented milk.
        Zhongguo Nong Ye Ke Xue. 2013; 46: 575-585
        • Wang J.C.
        • Guo Z.
        • Zhang Q.
        • Yan L.
        • Chen Y.
        • Chen X.
        • Liu X.-M.
        • Chen W.
        • Zhang H.-P.
        Effect of probiotic Lactobacillus casei Zhang on fermentation characteristics of set yogurt.
        Int. J. Dairy Technol. 2010; 63: 105-112
        • Wang X.N.
        Selection of Streptococcus thermophilus with high production of diacetyl and expression of the related functional genes. MS Thesis.
        Inner Mongolia Agricultural University, Hohhot2014
        • Wishart D.S.
        Metabolomics: Applications to food science and nutrition research.
        Trends Food Sci. Technol. 2008; 19: 482-493
        • Yin X.U.
        • Huang Y.J.
        • Chen X.
        • Mao-Lin L.U.
        • Rui-Xia G.U.
        Influence of milk content on flavor compounds in fermented soymilk.
        Shipin Kexue. 2013; 34: 1-5
        • Zareba D.
        • Ziarno M.
        • Obiedzinski M.
        Volatile profile of non-fermented milk and milk fermented by Bifidobacterium animalis subsp. lactis.
        Int. J. Food Prop. 2012; 15: 1010-1021
        • Zhang L.
        Application of Streptococcus thermophilus with high-yield acetaldehyde in fermented milk. MS Thesis.
        Inner Mongolia Agricultural University, Hohhot2014
        • Zhang W.
        • Yu D.
        • Sun Z.
        • Wu R.
        • Chen X.
        • Chen W.
        • Meng H.
        • Hu S.
        • Zhang H.
        Complete genome sequence of Lactobacillus casei Zhang, a new probiotic strain isolated from traditional homemade koumiss in inner Mongolia, China.
        J. Bacteriol. 2010; 192 (20675486): 5268-5269
        • Zhang Y.
        Probiotic effects of Lactobacillus casei Zhang: From single strain omics to metagenomics.
        Chin. Sci. Bull. 2019; 64: 307-314
        • Zhang Y.
        • Wang L.F.
        • Zhang J.C.
        • Li Y.X.
        • He Q.W.
        • Li H.
        • Guo X.
        • Guo J.L.
        • Zhang H.P.
        Probiotic Lactobacillus casei Zhang ameliorates high-fructose-induced impaired glucose tolerance in hyperinsulinemia rats.
        Eur. J. Nutr. 2014; 53 (23797890): 221-232
        • Zhong Z.
        • Zhang W.Y.
        • Du R.T.
        • Meng H.
        • Zhang H.P.
        Lactobacillus casei Zhang stimulates lipid metabolism in hypercholesterolemic rats by affecting gene expression in the live.
        Eur. J. Lipid Sci. Technol. 2012; 114: 244-252
        • Zhou T.
        • Huo R.
        • Kwok L.
        • Li C.
        • Ma Y.
        • Mi Z.
        • Chen Y.
        Effects of applying Lactobacillus helveticus H9 as adjunct starter culture in yogurt fermentation and storage.
        J. Dairy Sci. 2019; 102 (30343912): 223-235