Received: 29 April 2021 | Revised: 22 November 2021 | Accepted: 23 November 2021

DOI: 10.1002/fsn3.2755

ORIGINAL RESEARCH

Lactic acid bacteria isolated from Kazakh traditional fermented

milk products affect the fermentation characteristics and
sensory qualities of yogurt

Hui Li1 | Jiaxing Gao1 | Wenbo Chen1 | Chengjing Qian1 | Yong Wang2 |
Jing Wang3 | Lishui Chen1

1
China-­Australia Joint Research Center
for Dairy Future Technology, Beijing Key Abstract
Laboratory of Nutrition, Health & Food
Lactic acid bacteria (LAB) play a crucial role in the development of the taste, texture,
Safety, Beijing Engineering Laboratory
for Geriatric Nutrition Food Research, and aroma of traditional fermented milk products. Five LABs from Kazakh tradition-
COFCO Nutrition & Health Research
ally prepared dairy products showed continuous subculture stability, as well as proper
Institute, Beijing, China
2
Department of Chemical Engineering,
acidification and coagulation ability. They were identified as Pediococcus pentosaceus
Monash University, Clayton, Victoria, (1–­5, 1–­7), Enterococcus faecium (1–­19), and Lactobacillus plantarum (1–­12, 1–­15). Their
Australia
3
coagulation time and acidity values ranged from 5.97 to 12.78 h and 76.47 to 89.39°T.
Beijing Advanced Innovation Center for
Food Nutrition and Human Health, Beijing Yogurts prepared with L. plantarum were more condensed and textural integrity than
Technology & Business University, Beijing, those with P. pentosaceus and E. faecium. Determination of the volatile compound pro-
China
files suggested a higher diversity of volatile compounds than the control. The sensory
Correspondence evaluation presented positive overall sensory quality scores for the yogurts prepared
Hui Li, China-­Australia Joint Research
Center for Dairy Future Technology, with 1–­12 and 1–­15. The results provide additional information regarding the con-
Beijing Key Laboratory of Nutrition, tributions of native LABs to the unique flavor and sensory qualities of traditionally
Health & Food Safety, Beijing Engineering
Laboratory for Geriatric Nutrition Food prepared milk products. They may help to select starters or adjunct starters for de-
Research, COFCO Nutrition & Health veloping distinctive, traditional nomadic fermented milk to satisfy consumer demand
Research Institute, Beiqijia, Changping,
Beijing, China. and increase market acceptability.
Email: lihui_2006@hotmail.com
KEYWORDS
Funding information bacterial identification, lactic acid bacteria, starter culture, yogurt sensory qualities
Ministry of Science and Technology of
the People's Republic of China, Grant/
Award Number: 2016YFE0101200;
Beijing Municipal Science and Technology
Commission, Grant/Award Number:
Z191100008619003

1 | I NTRO D U C TI O N food is derived from (Guarcello et al., 2016). The Kazakh herdsmen
living in the Xinjiang Tianshan and Altai Mountain regions of north-
The food in each region or area has particular characteristics because west China have developed a variety of distinctive traditional, local
of the uniqueness of local ingredients and production techniques, fermented dairy products, such as milk curd, milk knots, yogurt, and
which are deeply rooted in tradition and linked to the territory the vrum (the coagulated layer of fat on the surface of boiled milk which

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,
provided the original work is properly cited.
© 2022 The Authors. Food Science & Nutrition published by Wiley Periodicals LLC.

Food Sci Nutr. 2022;10:1451–1460.  wileyonlinelibrary.com/journal/fsn3 | 1451

1452 | LI et al.

is dried by hanging it in a well-­ventilated place). These products are at −20°C in MRS broth supplemented with 20% (v/v) glycerol and
rich in nutrients, typically recognizable by their unique taste, and are only activated prior to testing by two sequential transfers in the
popular in local markets (Zuo et al., 2014). same broth used in the experiments.
Lactic acid bacteria (LAB) play a crucial role in the development of
the taste, texture, and aroma of traditional fermented milk products.
They are widely distributed in nature and are well-­known as starters 2.3 | Primary selection of LAB with milk
for traditional fermented products (Bao et al., 2011). Multiple re- fermentative characteristics
searchers have examined the LAB species composition of traditional
fermented products and expanded their potential industrial applica- The microbial culture was inoculated at a level of 2 ml/100 ml in ster-
tion. Ren et al. (2017) demonstrated that Lactobacillus was the most ile reconstituted skim milk (120 g/L) (BD Biosciences Pharmingen)
commonly isolated LAB species from fermented yak milk in central fortified with 6% sucrose and incubated at 42°C. Each strain was
Tibet. Enterococcus faecium was detected in all wooden vats used subcultured 10 times in sterile reconstituted skim milk. After 10
for the production of traditional stretched cheeses in Italy (Scatassa generations, the strains with continuous subculture stability, suit-
et al., 2015). Recently, several studies have also explored the con- able acidification, and coagulation ability were selected for further
tribution of relevant strains of native LAB to the sensory proper- analysis.
ties of traditional dairy products (Choobari et al., 2021; Guarrasi
et al., 2017; Madhubasani et al., 2020; Medeiros et al., 2016; Tian,
Shi, et al., 2019; Tian, Xu et al., 2019). 2.4 | LAB identification
To the best of our knowledge, limited work has focused on the
contribution of LAB to the organoleptic properties of traditional Strains were harvested for DNA extraction and purification after
Kazakh dairy products. Therefore, LAB strains in Kazakh dairy prod- cultivation in MRS broth at 37°C for 18 h. Total DNA was extracted
ucts were isolated, identified, and their fermentation characteristics using a Bacterial Genomic DNA Extraction Kit (Tiangen). DNA con-
were evaluated with the aim of providing additional information to centration and quality were determined with a NanoDrop 2000
select starters or adjunct starters for developing distinctive tradi- ultra-­microspectrophotometer (Thermo Fisher Scientific). The 16S
tional fermented milk products such as yogurt or cheese. rRNA gene sequence primers used for PCR amplification were
27F (5′-­ AGA GTT TGA TCC TGG CTC AG -­3′) and 1492R (5′-­TAC
GYT ACC TTG TTA CGA CTT -­3′) (Rashid & Hassanshahian, 2014).
2 | M ATE R I A L S A N D M E TH O DS Amplification was performed in a Veriti 96-­well thermal cycler (Life
Technologies). The PCR temperature profile was as follows: 95°C
2.1 | Sampling pre-­denaturation for 10 min, 35 cycles of 95°C denaturation for 30 s,
58°C annealing for 30 s, and 72°C extension for 30 s, then a 72°C
Five traditionally prepared samples (one cheese, one milk knot, two extension for 10 min. PCR products were sequenced by Invitrogen
yogurts, and one vrum) were obtained from local Kazakh herding Shanghai Trading Co. Ltd.
families living in Tianshan Mountain, Xinjiang, China. The samples The 16S rRNA gene nucleotide sequences of the five isolates
were collected aseptically in sterile bags, kept in an ice-­box container were analyzed and identified using the BLAST program on the NCBI
during transit to the laboratory, and stored at −20°C until analysis. website (http//:www.ncbi.nlm.nih.gov/blast). Alignments were per-
formed to construct a phylogenetic tree and compare similarities
among the sequences using the neighbor-­joining method in MEGA
2.2 | LAB isolation software version 6.0 (http://www.megas​oftwa​re.net) and boot-
strapped with 1000 replicates (Tamura et al., 2013).
Samples were isolated using two selective media, MRS and M17
(Beijing Land Bridge Technology), under both aerobic and anaerobic
conditions at 37 and 30°C for 72 h to obtain as many LAB strains as 2.5 | Fermentation characterization of selected
possible. Isolates were selected and purified according to previously strains in yogurt manufacturing
published methods (Medeiros et al., 2016).
All isolates were verified as LAB using a combination of Gram Yogurt was made from whole milk (Sanyuan Dairy). The milk was
reaction, catalase activity, and morphological analysis. Gram reac- homogenized, pasteurized at 95°C for 15 min, and cooled to 40°C
tions were visualized using an optical microscope (Scope A1; Carl before the selected microbial culture was inoculated (2 ml/100 ml).
Zeiss Microscopy) under oil immersion at 100-­fold magnification. Samples were incubated at 42°C until coagulation (until pH reached
Colony morphology was recorded using a colony counter Scan1200 4.6). The samples were immediately cooled in an ice-­water bath
(Interscience International). Cocci, bacilli, or coccobacilli colonies and stored at 4°C for 12 h to mature and the acidity was recorded.
that were gram-­positive with a negative catalase result were in- Texture, flavor compounds, and sensory qualities were analyzed.
cluded in the LAB group (Abosereh et al., 2016). Isolates were stored Lactobacillus delbrueckii subsp. bulgaricus CGMCC 1.1480, (China
LI et al. | 1453

General Microbiological Culture Collection Center, Beijing, China) and a scan range from 29 to 400 m/z at 3.33 scans/s. Identification
preserved at the China-­Australia Joint Research Center for Dairy of the volatile compounds was based on comparison of their mass
Future Technology was used as the control. spectra with those from previously analyzed authentic compounds
(spectra from the 14.0 NIST spectrum library). The retention index
used a homologous series of alkanes (C6-­C28; Sigma-­Aldrich). To
2.6 | Acidification and coagulation analysis quantify the volatile compounds, 2-­octanol was used as an internal
standard. Results were represented as the percentage of each com-
The pH was measured using a PB-­10 pH meter (Sartorius Scientific pound, thus allowing strain comparisons based on the relative con-
Instruments). Acidity was determined by titration with 0.1 N tents of each compound, but not of their concentrations in sample
NaOH using phenolphthalein as an indicator, and the results cultures (Cheng et al., 2017; Guarrasi et al., 2017).
were expressed in Thorner degrees (°T). Coagulation activity was
defined as clotting after incubation at 42°C within 4–­6 h (fast),
6–­12 h (medium), or >12 h (slow) time frames (Ayad et al., 2004; 2.9 | Sensory evaluation
Yi et al., 2011).
According to Ao et al. (2012) and Hashim et al. (2009), 12 panelists
aged 23–­46 y (staff from the China-­Australia Joint Research Center
2.7 | Textural analysis for Dairy Future Technology selected based on interest and experi-
ence in the sensory evaluation of yogurt and fermented milk) per-
Yogurt texture was evaluated by the backward-­extrusion test using formed the sensory evaluation of the yogurts. Panelists assessed
a TA-­X T Plus texture analyzer (Stable Micro Systems) equipped with the fermented yogurt characteristics using a 10-­point scale based on
a 5 kg loading cell (Mousavi et al., 2019; Wang et al., 2019). The the following attributes: appearance (1 = no curd, 10 = smooth yo-
parameters were modified from the Exponent template: 35 mm di- gurt gel without whey), flavor (1 = yogurt with unsatisfactory odor,
ameter cylinder probe, 1.0 mm/s test speed, 30 mm penetration dis- 10 = yogurt with satisfactory, fermented flavor), taste (1 = weak
tance, and 10 g surface trigger force. The tests were carried out on acidity or over acidity, 10 = suitable acidity), and overall quality
samples prepared in 125 ml containers (64 mm diameter and 70 mm (1 = poor quality, 10 = excellent quality).
height). Hardness (g), springness (%), consistency (g × s), and gummi-
ness index (g) were calculated using the Exponent program, based
on the instruction manual supplied by the manufacturer. 2.10 | Statistical analysis

All experiments were performed in triplicate and results were ex-

2.8 | Volatile compound analysis pressed as the mean ± SD. Statistical analyses were performed by
IBM SPSS Statistics 20 (IBM Inc.). One-­way analysis of variance was
Twenty grams of yogurt was weighed and mixed with 5 g NaCl in used to compare the means. Mean separations were performed by
a 60 ml glass vial. Five grams of the mixture was placed in a 20 ml S-­N-­K (Newman–­Keuls). Differences at p < .05 were considered sig-
head-­
space vial sealed with a polytetrafluoroethylene-­
faced sili- nificant. Additionally, principal component analysis (PCA) was per-
cone septum (VWR). Extractions were carried out with a solid formed with JMP Pro 16 software.
phase microextraction device (Supelco) containing a fused-­
silica
fiber coated with a 50/30 μm layer of Divinylbenzene/Carboxen/
Polydimethylsiloxane. The vial was equilibrated at 60°C, centrifuged 3 | R E S U LT S A N D D I S CU S S I O N
at 300 rpm for 2 min, and the fiber was exposed to the headspace
over the sample for 30 min. The fiber was conditioned before adsorp- 3.1 | Isolation and selection of the fermentative
tion by heating it in the gas chromatograph injection port at 250°C. LAB
After adsorption, the fiber was removed from the vial and immedi-
ately inserted into the GC-­MS injection port (Bezerra et al., 2017). Nineteen LAB strains were selected from MRS and M17 plates ac-
Separation was performed on an Agilent 7890A gas chromato- cording to morphological analysis of colony color, shape, and gloss.
graph (Agilent Technologies) coupled to a mass detector (Agilent Of these, 14 strains were identified as gram-­positive and catalase-­
5975C) and DB-­WAX capillary column (60 m × 250 μm × 0.25 μm). negative cocci; the other five were gram-­
positive and catalase-­
The temperature program was as follows: 3 min at 32°C, 12°C/min negative bacilli presumed to be LAB.
ramp to 48°C, hold for 10 min, 6°C/min ramp to 130°C, 10°C/min Five strains with continuous subculture stability, proper acidifi-
ramp to 200°C, 20°C/min ramp to 230°C, and hold for 2 min. The cation, and coagulation ability were selected for further analysis as
injection port temperature was 250°C, and the flow rate was 1.0 ml/ these qualities are essential for producing high-­quality constant fer-
min. The mass spectrometer was operated in electron impact mode mentation dairy products (Ayad et al., 2004). The five strains were
with a source temperature of 230°C, an ionization voltage of 70 eV, coded as 1–­5, 1–­7, 1–­12, 1–­15, and 1–­19.
1454 | LI et al.

Gram stain results showed that strains 1–­5, 1–­7, and 1–­19 were medium acidification and coagulation activity (Yi et al., 2011). Ayad
gram-­positive cocci, while strains 1–­12 and 1–­15 were gram-­positive et al. (2004) demonstrated that most L. plantarum strains are active
and rod-­shaped (Figure 1a). Additionally, molecular identification acid-­forming agents and coagulate milk in 3–­9 h. Despite having me-
suggested that the five strains were Pediococcus pentosaceus (1–­5, dium coagulation activity (clotting time from 8.44–­12.78 h), strains
1–­7), Enterococcus faecium (1–­19), and Lactobacillus plantarum (1–­12, 1–­7 (Pediococcus pentosaceus), 1–­
5 (Pediococcus pentosaceus), and
1–­15). Phylogenetic tree evaluation confirmed the molecular identi- 1–­19 (Enterococcus faecium) generated high acidity levels (>70.00°T),
fication results (Figure 1b). indicating that these strains were good candidate starters for the
dairy fermentation process.

3.2 | Fermentation time and acidity assays of the

five selected strains 3.3 | Yogurt textural properties

Strain 1–­12 (Lactobacillus plantarum) produced higher levels of acid Textural characteristics of the yogurts are listed in Table 1. Strain
(89.39°T) and had a lower clotting time (5.97 h) than the other strains, 1–­12 showed the highest hardness (187.48 g), springness (69.10%),
followed by strain 1–­15 (Lactobacillus plantarum) (86.87 °T; 7.66 hr) and consistency (3804.66 g × s) values, which were similar to those
(Figure 2). Previous reports showed that Lactobacilli have fast or of the control, followed by those of strain 1–­15, which also showed

F I G U R E 1 (a) Strain evaluation by Gram staining of the five selected strains (×1000), and (b) phylogenetic tree analysis based on 16S
rRNA gene sequences of the five selected isolates
LI et al. | 1455

F I G U R E 2 Fermentation time and

acidification of the selected LAB isolates.
Values are expressed as the mean ± SD.
a–­f: Different letters above the same
column indicate significant differences
(*p < .05)

TA B L E 1 Textural parameters of yogurts fermented by the five selected strains

Gumminess
Species Strain Hardness (g) Springness (%) Consistency (g × s) (g)

Pediococcus pentosaceus 1–­5 122.71 ± 2.01b 51.17 ± 1.44b 3396.43 ± 2.18b 16.33 ± 0.44b
Pediococcus pentosaceus 1–­7 153.44 ± 1.00 c 57.59 ± 1.51c 3657.48 ± 3.40 c 20.30 ± 0.97c
a a a
Enterococcus faecium 1–­19 107.81 ± 2.28 45.76 ± 1.23 2083.33 ± 3.22 13.41 ± 1.28a
Lactobacillus plantarum 1–­12 187.48 ± 2.25e 69.10 ± 1.67d 3804.66 ± 2.08e 19.45 ± 0.69c
d d d
Lactobacillus plantarum 1–­15 181.47 ± 0.93 68.80 ± 2.59 3772.41 ± 1.18 20.01 ± 0.99c
Lactobacillus delbrueckii subsp. Bulgaricus Control strain 186.59 ± 1.26e 67.78 ± 1.23d 3801.39 ± 1.84e 19.38 ± 0.83c

Note: Values are expressed as mean ± SD. Different letters within the same column indicate significant differences (*p < .05).

high hardness, springness, and consistency values. Strains 1–­12, 1–­ activities of indigenous and/or added microorganisms (Routray &
15, and 1–­7 showed similar gumminess values to that of the control Mishra, 2011; Tian, Shi, et al., 2019; Tian, Xu, et al., 2019). As shown
(19.38 g). According to Cheng et al. (2017), texture is an essential as- in Table 2, ketones and acids were the most abundant compounds
pect of yogurt quality and plays an important role in sensory evalua- in the yogurt aromatic profiles, followed by alcohols and aldehydes.
tion and consumer acceptability. Yogurts fermented by strains 1–­12 Ketones are generated from the β-­oxidation of acyl lipids and are a
and 1–­15 were condensed and had integrity, suggesting their poten- rich constituent of many dairy products (Matera et al., 2018; Tian,
tial in fermented dairy product preparation. Shi, et al., 2019; Tian, Xu et al., 2019). These compounds have char-
acteristic odors and low perception thresholds (Frank et al., 2004).
The highest ketone production occurred in Pediococcus pentosaceus
3.4 | Volatile compound profile (1–­5) and Enterococcus faecium (1–­19) (52.56% and 53.19%, respec-
tively), consistent with previous findings (Guarrasi et al., 2017).
Determination of the volatile compound profiles suggested that the Among the identified ketones, 2,3-­butanedione and acetoin were
five strains produced a higher diversity of volatile compounds than detected at high levels for 1–­5 and 1–­19 (41.05% and 44.85%, re-
the control (Table 2). Thirty-­six volatile compounds produced by the spectively), and at low levels for 1–­12 and 1–­15, (26.26% and 20.83%,
five sample strains were detected and grouped into seven chemical respectively) compared to 1–­7 and the control (30.42% and 30.58%,
categories: ketones, alcohols, aldehydes, acids, ethers, esters, and respectively). The 2,3-­butanedione and acetoin are generated from
polyaromatic hydrocarbons. pyruvate, which comes from lactose in milk and citrate metabolism
The characteristic aroma of fermented milk products is ex- (Dan et al., 2017; Pan et al., 2014). Both of these compounds im-
tremely variable because it is strongly influenced by the enzymatic part a “buttery, creamy, vanilla” flavor to yogurt (Dan et al., 2017;
TA B L E 2 Volatile compounds of each yogurt sample analyzed by SPME-­GC-­MS
1456
|

RI RI L Relative contents (%)

Volatile components WAX WAX 1–­5 1–­7 1–­19 1–­12 1–­15 Control

Ketones 2,3-­butanedione 1043 1056 8.50 ± 0.03 7.86 ± 0.06 11.37 ± 0.35 6.52 ± 0.08 5.32 ± 0.07 10.71 ± 0.19
2-­ heptanone 1180 1189 8.86 ± 0.03 4.41 ± 0.11 6.97 ± 0.06 8.57 ± 0.09 9.29 ± 0.03 1.29 ± 0.02
2-­ethyl-­c yclopentanone 0.5 ± 0.03
acetoin 1298 1299 32.55 ± 0.86 22.56 ± 0.88 33.48 ± 0.48 19.74 ± 0.05 15.51 ± 0.41 19.87 ± 0.11
Hydroxyacetone 0.25 ± 0.01
2-­nonanone 1.81 ± 0.09 0.9 ± 0.03 0.94 ± 0.01 2.04 ± 0.07 2.48 ± 0.04 5.14 ± 0.03
2-­methyloxolan-­3-­one 0.56 ± 0.02
4-­c yclopentene-­1, 3-­butanedione 0.36 ± 0.02 0.24 ± 0.03 0.39 ± 0.01
Undecan-­2-­one 0.48 ± 0.02 0.33 ± 0.01 0.43 ± 0.02 0.5 ± 0.01 0.68 ± 0.01 1.83 ± 0.06
52.56 ± 0.18 37.05 ± 0.15 53.19 ± 0.18 37.76 ± 0.52 33.84 ± 0.97 38.84 ± 0.08
Alcohols Ethanol 1.87 ± 0.02
3-­methylbut-­3-­en-­1-­ol 0.71 ± 0.02
Hexan-­1-­ol 0.32 ± 0.02 0.54 ± 0.01
D-­su-­O-­threonine ethyl ester 1.29 ± 0.02
2,3-­butanediol 4.85 ± 0.06
Furfuryl alcohol 10.38 ± 0.02 7.1 ± 0.2 8.97 ± 0.05 8.46 ± 0.03 7.98 ± 0.03
11.99 ± 0.34 7.1 ± 0.2 8.97 ± 0.05 8.46 ± 0.54 15.95 ± 0.52
Aldehydes Furfural 1.65 ± 0.04 1.01 ± 0.05 1.36 ± 0.04 1.22 ± 0.04 1.56 ± 0.03
Acids Acetic acid 1485 1425 15.78 ± 0.04 25.40 ± 0.05 15.03 ± 0.04 13.54 ± 0.05 25.32 ± 0.32 2.02 ± 0.06
Isobutyric acid 0.33 ± 0.01 0.7 ± 0.03 1.39 ± 0.02
Butyric acid 1667 1652 4.00 ± 0.01 4.98 ± 0.03 4.01 ± 0.01 5.64 ± 0.05 4.00 ± 0.01 7.24 ± 0.05
DL-­3-­methylvaleric acid 0.75 ± 0.04
Isovaleric acid 1.94 ± 0.05 4.77 ± 0.11 2.42 ± 0.55 2.69 ± 0.12
Hexanoic acid 1928 1925 6.2 ± 0.03 13.15 ± 0.13 8.98 ± 0.07 17.43 ± 0.10 10.23 ± 0.07 34.40 ± 1.87
2-­ethylhexanoic acid 0.28 ± 0.11 0.42 ± 0.02
Isovaleric acid 0.43 ± 0.02
Butanecarboxylic acid 0.24 ± 0.01 0.43 ± 0.02 0.43 ± 0.02 0.80 ± 0.01
Octanoic acid 2056 2084 4.29 ± 0.02 6.07 ± 0.30 4.70 ± 0.05 9.83 ± 0.04 4.70 ± 0.01 21.37 ± 0.09
Nonanoic acid 2187 2192 0.46 ± 0.01 0.6 ± 0.02 0.09 ± 0.01
32.73 ± 0.04 55.13 ± 0.09 37.15 ± 0.09 51.55 ± 0.05 45.09 ± 0.07 65.84 ± 0.42
LI et al.
 LI et al. | 1457

Settachaimongkon et al., 2014). Many studies have demonstrated

that 2,3-­butanedione, a diketone, can readily be converted to ac-

Control
etoin by the enzyme diacetyl reductase (Carballo et al., 1991; Dan
et al., 2017; Rattray et al., 2003). The flavor characteristics of
2,3-­butanedione and acetoin are similar, and in combination, they
contribute to the “buttery” flavor of fermented milk products; they
0.43 ± 0.04

0.59 ± 0.08
0.26 ± 0.03

0.29 ± 0.06

0.32 ± 0.01
0.75 ± 0.01
significantly contribute to their buttery and creamy aromas (Dan
et al., 2017; Nieto-­Arribas et al., 2011). Other ketones, such as
1–­15

Note: Values are expressed as mean ± SD. 1–­5: P. pentosaceus, 1–­7: P. pentosaceus, 1–­19: E. faecium, 1–­12: L. plantarum,1–­15: L. plantarum, control: L. delbrueckii subsp. Bulgaricus.
2-­heptanone, 2-­nonanone, and 2-­methyl-­n-­nonone were also de-
0.77 ± 0.01 tected. The relative abundance (%) of 2-­heptanone for the five iso-

0.31 ± 0.01
lates was higher than that of the control. 2-­heptanone, 2-­nonanone,
and 2-­methyl-­n-­nonone contribute to herbaceous, fruity, floral, and
1–­12

creamy yogurt odors and are considered favorable for flavoring milk
products (Bezerra et al., 2017).
Acids are aromatic compounds that have important effects on
yogurt flavor (Bezerra et al., 2017; Routray & Mishra, 2011). Acids
accounted for 32.73% (1–­5), 55.13% (1–­7), 37.15% (1–­19), 51.55% (1–­
1–­19

12), 45.09% (1–­15), and 65.84% (control) of the total compounds.

Acids can be generated through either lipolysis or glycolysis, but
they are mainly generated through lactose metabolism (Hayaloglu
et al., 2013). Acids are also used in the formation of other aromatic
compounds such as ketones and alcohols (Delgado et al., 2010).
Acetic acid was the major acid detected in all five isolates (>13%).
1–­7
Relative contents (%)

Acetic acid produces a vinegar flavor and is associated with the

slight tart taste of dairy products (Tian, Shi, et al., 2019; Tian, Xu,
0.48 ± 0.01

0.77 ± 0.01

et al., 2019). However, the control generated a higher relative abun-

dance (%) of hexanoic (34.40%) and octanoic (21.37%) acid than the
1–­5

isolates, with the exception of 1–­12 (Lactobacillus plantarum); the

relative abundances (%) of hexanoic and octanoic acid were 17.42%
and 9.83%, respectively, which were slightly higher than those of the
WAX
RI L

other four isolates. Butyric acid was also detected in all the yogurt
samples. Hexanoic acid, which originates from lipolysis, produces a
light cream flavor. Butyric and octanoic acids contribute to the char-
WAX

acteristic flavor of cheeses (Delgado et al., 2010).

RI

The classes of alcohols and aldehydes followed those of ke-

tones and acids in terms of the relative abundance (%) of volatile
Ethyl 2-­hydroxy-­2-­(4-­hydroxyphenyl)-­acetate
1-­methoxy-­2-­[2-­[2-­[2-­[2-­(2-­methoxyeth-­oxy)

compounds for the five isolates, and were not detected in the con-
ethoxy] ethoxy]ethoxy] ethoxy]ethane

trol strain yogurt. Furfuryl alcohol was the major alcohol present.
Although alcohols have a limited influence on flavor due to their
4-­phenyl-­3H-­1,3-­thiazole-­2-­thione

1-­benzyl-­3-­ methyl-­ ethylene urea

high sensory thresholds, they represent an index of the fermenta-

o-­methoxycarbonylphenolate

tion process (Langler et al., 1967). Furfural was the only aldehyde
detected in the yogurts.
Volatile components

1-­methyl octyl ether

The abundant and unique volatile compounds present in the fer-

Anthracen-­1-­amine

mented yogurts may be one reason for the singular flavor of tradi-
2-­acetylfuran

tional Kazakh fermented dairy products.

Styrene
TA B L E 2 (Continued)

3.5 | PCA of volatile compound profiles of yogurts

prepared with different strains
hydroarbons
Polyaromatic

PCA of volatile compounds of yogurts prepared with different

Ethers

Esters

starter cultures based on the content was conducted. In Figure 3a,

the scatter plot for the two first principal components (PC1 and PC2)
1458 | LI et al.

F I G U R E 3 Principal component analysis (PCA) on flavor compounds. (a) PCA plots from separation of yogurts prepared with the six
strains, and (b) Distribution of volatile compounds in yogurt prepared with the six strains based on content

TA B L E 3 Sensory evaluation of yogurt samples

Overall
Species Strain Appearance Flavor Taste quality

Pediococcus pentosaceus 1–­5 8.03 ± 0.48b 5.44 ± 0.31b 6.30 ± 0.41b 6.59 ± 0.22b
Pediococcus pentosaceus 1–­7 8.64 ± 0.27c 5.66 ± 0.32b 7.48 ± 0.24c 7.26 ± 0.18c
a a a
Enterococcus faecium 1–­19 6.16 ± 0.39 4.71 ± 0.32 5.51 ± 0.41 5.46 ± 0.17a
Lactobacillus plantarum 1–­12 9.45 ± 0.25e 7.73 ± 0.33d 8.51 ± 0.23e 8.56 ± 0.16e
d c d
Lactobacillus plantarum 1–­15 9.04 ± 0.24 6.94 ± 0.26 7.85 ± 0.37 7.94 ± 0.20 d
Lactobacillus delbrueckii subsp. Bulgaricus Control strain 9.16 ± 0.19d 8.91 ± 0.38e 9.10 ± 0.39f 9.06 ± 0.22f

Note: Values are expressed as mean ± SD. Different letters within the same column indicate significant differences (*p < .05).

represents the differences among the samples fermented by the six (Cheng, 2010). Although 1–­5, 1–­7, and 1–­19 were clustered into
strains. The corresponding loading plot was illustrated in Figure 3b, same group, they were located in the positive region of PC1 and
which represent the relative importance of each volatile compound the negative region of PC2, as the control was. There were total
and the relationships between volatile compounds and samples of four flavor compounds in this region, including three acids and
(Zhang et al., 2020). PC1 and PC2 explained 30% and 24% of the one ketone. Those compounds were also found in yogurt products
total variation, respectively. (Cheng, 2010).
Samples prepared with the six strains could be clustered into
three groups. The first group (1–­12 and 1–­15) was located in the
negative region of PC1 and the positive region of PC2. 1–­5, 1–­7, and 3.6 | Sensory qualities
1–­19 were clustered into the second group. The third group only
contained the control sample. This distribution of PCA also sug- Sensory evaluation results for the five yogurt samples are shown in
gested the flavor uniqueness of the traditional Kazakh fermented Table 3. Yogurts made with strains 1–­12 and 1–­15 received positive
dairy products. overall sensory quality scores for a smooth yogurt gel appearance
As shown in Figure 3b, five flavor compounds (2-­h eptanone, and relatively satisfactory fermented flavor. This evaluation may be
furfural, furfural alcohol, and other two polyaromatic hydrocar- a result of their good acidification and coagulation ability, condensed
bons) were located in the negative region of PC1 and the posi- and textural integrity, and diverse and unique flavors.
tive region of PC2. 2-­H eptanone, furfural, and furfural alcohol These results suggest that strains 1–­12 and 1–­15 can be used as
are typical compounds detected in fermented milk products adjunct cultures with a control strain, or other commercial strains,
LI et al. | 1459

for manufacturing fermented dairy products with a unique flavor REFERENCES

(Crow et al., 2001; Leroy & De Vuyst, 2004). Abosereh, N. A., El Ghani, S. A., Gomaa, R. S., & Fouad, M. T. (2016).
Molecular identification of potential probiotic lactic acid bacteria
strains isolated from Egyptian traditional fermented dairy prod-
ucts. Biotechnology, 15(1–­2), 35–­43. https://doi.org/10.3923/biote​
4 | CO N C LU S I O N ch.2016.35.43
Ao, X., Zhang, X., Zhang, X., Shi, L., Zhao, K., Yu, J., Dong, L., Cao, Y.,
Our results showed that the yogurts fermented by the five LAB that & Cai, Y. (2012). Identification of lactic acid bacteria in traditional
fermented yak milk and evaluation of their application in fermented
were isolated from fermented milk products in Xinjiang, China had
milk products. Journal of Dairy Science, 95(3), 1073–­1084. https://
the unique flavor and sensory qualities. Two isolates, strains 1–­12 doi.org/10.3168/jds.2011-­4224
(Lactobacillus plantarum) and 1–­15 (Lactobacillus plantarum), might Ayad, E. H. E., Nashat, S., El-­Sadek, N., Metwaly, H., & El-­Soda, M.
be practically applied as adjunct cultures for the production of fer- (2004). Selection of wild lactic acid bacteria isolated from tradi-
tional Egyptian dairy products according to production and tech-
mented milk products because of their desirable acidifying capacity,
nological criteria. Food Microbiology, 21(6), 715–­725. https://doi.
condensed texture, and pleasing flavor qualities. Creating start- org/10.1016/j.fm.2004.02.009
ers by mixing these native strains with commercial LAB should be Bao, Q., Chen, X., Liu, H., Zhang, W., Liu, W., Yu, J., Wang, F., & Zhang,
considered to develop distinctive, traditional nomadic fermented H. (2011). Isolation and identification of cultivable lactic acid bac-
milk products to satisfy consumer demand and increase market teria from traditional goat milk cake in Yunnan province of China.
African Journal of Microbiology Research, 5(29), 5284–­5291. https://
acceptability.
doi.org/10.5897/AJMR11.508
Bezerra, T. K. A., Arcanjo, N. M. D. O., Araújo, A. R. R. D., Queiroz, A.
AC K N OW L E D G M E N T S L. M. D., Oliveira, M. E. G. D., Gomes, A. M. P., & Madruga, M. S.
This work was financed by the National Key R & D Program of the (2017). Volatile profile in goat coalho cheese supplemented with
probiotic lactic acid bacteria. LWT -­Food Science and Technology, 76,
China-­Australia Joint Research Center for Dairy Future Technology
209–­215. https://doi.org/10.1016/j.lwt.2016.03.041
(No. 2016YFE0101200) supported by the Ministry of Science and Carballo, J., Martin, R., Bernardo, A., & Gonzalez, J. (1991). Purification,
Technology of the People's Republic of China; it was also financed characterization and some properties of diacetyl (acetoin) reductase
by Beijing Municipal Science and Technology Commission Project from Enterobacter aerogenes. European Journal of Biochemistry, 198,
327–­332. https://doi.org/10.1111/j.1432-­1033.1991.tb160​19.x
(Z191100008619003) of China. Additionally, the Beijing Advanced
Cheng, H. (2010). Volatile flavor compounds in yogurt: A review. Critical
Innovation Center for Food Nutrition and Human Health, Beijing Reviews in Food Science and Nutrition, 50(10), 938–­950. https://doi.
Technology & Business University was gratefully acknowledged. org/10.1080/10408​39090​3 044081
Finally, the authors were very grateful to all the panelists who par- Cheng, J., Xie, S., Yin, Y., Feng, X., Wang, S., Guo, M., & Ni, C. (2017).
Physiochemical, texture properties, and the microstructure of set
ticipated in the study.
yogurt using whey protein-­ sodium tripolyphosphate aggregates
as thickening agents. Journal of the Science of Food and Agriculture,
C O N FL I C T O F I N T E R E S T 97(9), 2819–­2825. https://doi.org/10.1002/jsfa.8110
The authors declare no conflict of interest. Choobari, S. Z. M., Sari, A. A., & Garmakhany, A. D. (2021). Effect of
Plantago ovata Forsk seed mucilage on survivability of Lactobacillus
acidophilus, physicochemical and sensory attributes of produced
AU T H O R C O N T R I B U T I O N low-­fat set yoghurt. Food Science & Nutrition, 9(2), 1040–­1049.
Hui Li: Data curation (lead); Project administration (lead); Writing https://doi.org/10.1002/fsn3.2074
–­ review & editing (equal). Jiaxing Gao: Formal analysis (equal); Crow, V., Curry, B., & Hayes, M. (2001). The ecology of non-­starter lac-
Validation (equal). Wenbo Chen: Data curation (equal); Writing –­ tic acid bacteria (NSLAB) and their use as adjuncts in New Zealand
Cheddar. International Dairy Journal, 11(4–­7 ), 275–­283. https://doi.
original draft (equal); Writing –­ review & editing (lead). Chengjing
org/10.1016/S0958​-­6946(01)00057​-­7
Qiang: Data curation (supporting). Yong Wang: Writing –­ review & Dan, T., Wang, D., Jin, R. L., Zhang, H. P., Zhou, T. T., & Sun, T. S.
editing (supporting). Jing Wang: Methodology (equal); Supervision (2017). Characterization of volatile compounds in fermented
(equal). Lishui Chen: Supervision (supporting). milk using solid-­ phase microextraction methods coupled with
gas chromatography-­mass spectrometry. Journal of Dairy Science,
100(4), 2488–­2500. https://doi.org/10.3168/jds.2016-­11528
E T H I C A L A P P R OVA L Delgado, F. J., González-­Crespo, J., Cava, R., García-­Parra, J., & Ramírez,
This study does not involve any human or animal testing. R. (2010). Characterisation by SPME–­GC–­MS of the volatile pro-
file of a Spanish soft cheese PDO Torta del Casar during ripening.
Food Chemistry, 118(1), 182–­189. https://doi.org/10.1016/j.foodc​
DATA AVA I L A B I L I T Y S TAT E M E N T
hem.2009.04.081
The data that support the findings of this study are available from Frank, D. C., Owen, C. M., & Patterson, J. (2004). Solid phase mi-
the corresponding author upon reasonable request. croextraction (SPME) combined with gas-­ chromatography and
olfactometry-­ mass spectrometry for characterization of cheese
aroma compounds. LWT -­Food Science and Technology, 37(2), 139–­
ORCID
154. https://doi.org/10.1016/s0023​-­6 438(03)00144​-­0
Hui Li https://orcid.org/0000-0003-0303-0611 Guarcello, R., Carpino, S., Gaglio, R., Pino, A., Rapisarda, T., Caggia,
Jing Wang https://orcid.org/0000-0003-4896-0142 C., Marino, G., Randazzo, C. L., Settanni, L., & Todaro, M. (2016).
1460 | LI et al.

A large factory-­ s cale application of selected autochthonous Ren, Y., Yang, Y., Zhang, D., Wang, D., Zhang, H., & Liu, W. (2017). Diversity
lactic acid bacteria for PDO Pecorino Siciliano cheese produc- analysis and quantification of lactic acid bacteria in traditionally
tion. Food Microbiology, 59, 66–­75. https://doi.org/10.1016/j. fermented yaks’ milk products from Tibet. Food Biotechnology,
fm.2016.05.011 31(1), 1–­19. https://doi.org/10.1080/08905​436.2016.1269290
Guarrasi, V., Sannino, C., Moschetti, M., Bonanno, A., Grigoli, A. D., & Routray, W., & Mishra, H. N. (2011). Scientific and technical aspects
Settanni, L. (2017). The individual contribution of starter and non-­ of yogurt aroma and taste: A review. Comprehensive Reviews
starter lactic acid bacteria to the volatile organic compound com- in Food Science and Food Safety, 10(4), 208–­220. https://doi.
position of Caciocavallo Palermitano cheese. International Journal org/10.1111/j.1541-­4337.2011.00151.x
of Food Microbiology, 259, 35–­42. https://doi.org/10.1016/j.ijfoo​ Scatassa, M. L., Gaglio, R., Macaluso, G., Francesca, N., Randazzo, W.,
dmicro.2017.07.022 Cardamone, C., Grigoli, A. D., Moschetti, G., & Settanni, L. (2015).
Hashim, I. B., Khalil, A. H., & Habib, H. (2009). Quality and acceptability Transfer, composition and technological characterization of the lac-
of a set-­t ype yogurt made from camel milk. Journal of Dairy Science, tic acid bacterial populations of the wooden vats used to produce
92(3), 857–­862. https://doi.org/10.3168/jds.2008-­1408 traditional stretched cheeses. Food Microbiology, 52, 31–­41. https://
Hayaloglu, A., Yasar, K., Tolu, C., & Sahingil, D. (2013). Characterizing doi.org/10.1016/j.fm.2015.06.008
volatile compounds and proteolysis in Gokceada artisanal goat Settachaimongkon, S., Nout, M. J. R., Antunes Fernandes, E. C., Hettinga,
cheese. Small Ruminant Research, 113(1), 187–­194. https://doi. K. A., Vervoort, J. M., van Hooijdonk, T. C. M., Zwietering, M. H.,
org/10.1016/j.small​rumres.2013.01.001 Smid, E. J., & van Valenberg, H. J. F. (2014). Influence of different
Langler, J. E., Libbey, L. M., & Day, E. A. (1967). Identification and eval- proteolytic strains of Streptococcus thermophilus in co-­culture with
uation of selected compounds in Swiss cheese flavor. Journal Lactobacillus delbrueckii subsp. bulgaricus on the metabolite profile
of Agricultural and Food Chemistry, 15(3), 386–­391. https://doi. of set-­yoghurt. International Journal of Food Microbiology, 177, 29–­
org/10.1021/jf601​51a024 36. https://doi.org/10.1016/j.ijfoo​dmicro.2014.02.008
Leroy, F., & De Vuyst, L. (2004). Lactic acid bacteria as functional Tamura, K., Stecher, G., Peterson, D., Filipski, A., & Kumar, S. (2013).
starter cultures for the food fermentation industry. Trends in Food MEGA6: Molecular evolutionary genetics analysis version 6.0.
Science & Technology, 15(2), 67–­78. https://doi.org/10.1016/j. Molecular Biology and Evolution, 30(12), 2725–­2729. https://doi.
tifs.2003.09.004 org/10.1093/molbe​v/mst197
Madhubasani, G. B. L., Prasanna, P. H. P., Chandrasekara, A., Gunasekara, Tian, H., Shi, Y., Zhang, Y., Yu, H., Mu, H., & Chen, C. (2019). Screening
D. C. S., Senadeera, P., Chandramali, D. V. P., & Vidanarachchi, J. of aroma-­producing lactic acid bacteria and their application in im-
K. (2020). Exopolysaccharide producing starter cultures positively proving the aromatic profile of yogurt. Journal of Food Biochemistry,
influence on microbiological, physicochemical, and sensory proper- 43, e12837. https://doi.org/10.1111/jfbc.12837
ties of probiotic goats' milk set-­yoghurt. Journal of Food Processing Tian, H., Xu, X., Chen, C., & Yu, H. (2019). Flavoromics approach to
and Preservation, 44(3), e14361. https://doi.org/10.1111/jfpp.14361 identifying the key aroma compounds in traditional Chinese milk
Matera, J., Luna, A. S., Batista, D. B., Pimentel, T. C., Moraes, J., Kamimura, fan. Journal of Dairy Science, 102(11), 9639–­9650. https://doi.
B. A., Ferreira, M. V. S., Silva, H. L. A., Mathias, S. P., Esmerino, E. A., org/10.3168/jds.2019-­16796
Freitas, M. Q., Raices, R. S. L., Quitério, S. L., Sant'Ana, A. S., Silva, Wang, X., Kristo, E., & LaPointe, G. (2019). The effect of apple pomace
M. C., & Cruz, A. G. (2018). Brazilian cheeses: A survey covering on the texture, rheology and microstructure of set type yogurt.
physicochemical characteristics, mineral content, fatty acid profile Food Hydrocolloids, 91, 83–­91. https://doi.org/10.1016/j.foodh​
and volatile compounds. Food Research International, 108, 18–­26. yd.2019.01.004
https://doi.org/10.1016/j.foodr​es.2018.03.014 Yi, H., Zhang, L., Han, X., Du, M., Zhang, Y., Li, J., Sun, K., & Hou, Y.
Medeiros, R. S., Araújo, L. M., Neto, V. Q., Andrade, P. P., Melo, M. A., & (2011). Isolation and applied potential of lactic acid bacteria from
Gonçalves, M. M. B. P. (2016). Identification of lactic acid bacte- Chinese traditional fermented food in specific ecological locali-
ria isolated from artisanal Coalho cheese produced in the Brazilian ties. Food Science and Biotechnology, 20(6), 1685–­1690. https://doi.
Northeast. CyTA -­Journal of Food, 14(4), 613–­620. https://doi. org/10.1007/s1006​8-­011-­0232-­1
org/10.1080/19476​337.2016.1185468 Zhang, L., Mi, S., Liu, R., Sang, Y., & Wang, X. (2020). Evaluation of vola-
Mousavi, M., Heshmati, A., Garmakhany, A. D., Vahidinia, A., & Taheri, M. tile compounds in milks fermented using traditional starter cultures
(2019). Texture and sensory characterization of functional yogurt and probiotics based on odor activity value and chemometric tech-
supplemented with flaxseed during cold storage. Food Science & niques. Molecules, 25(5), 1129. https://doi.org/10.3390/molec​ules2​
Nutrition, 7(3), 907–­917. https://doi.org/10.1002/fsn3.805 5051129
Nieto-­Arribas, P., Seseña, S., Poveda, J. M., Chicón, R., Cabezas, L., & Palop, Zuo, F. L., Feng, X. J., Chen, L. L., & Chen, S. (2014). Identification and
L. (2011). Enterococcus populations in artisanal Manchego cheese: partial characterization of lactic acid bacteria isolated from tradi-
Biodiversity, technological, and safety aspects. Food Microbiology, tional dairy products produced by herders in the western Tianshan
28(5), 891–­899. https://doi.org/10.1016/j.fm.2010.12.005 Mountains of China. Letters in Applied Microbiology, 59(5), 549–­556.
Pan, D. D., Wu, Z., Peng, T., Zeng, X. Q., & Li, H. (2014). Volatile organic https://doi.org/10.1111/lam.12313
compounds profile during milk fermentation by Lactobacillus pen-
tosus and correlations between volatiles flavor and carbohydrate
metabolism. Journal of Dairy Science, 97(2), 624–­631. https://doi.
How to cite this article: Li, H., Gao, J., Chen, W., Qian, C.,
org/10.3168/jds.2013-­7131
Wang, Y., Wang, J., & Chen, L. (2022). Lactic acid bacteria
Rashid, S., & Hassanshahian, M. (2014). Screening, Isolation and iden-
tification of lactic acid bacteria from a traditional dairy product isolated from Kazakh traditional fermented milk products
of Sabzevar, Iran. International Journal of Enteric Pathogens, 2(4), affect the fermentation characteristics and sensory qualities of
e18393. https://doi.org/10.17795/​ijep1​8393 yogurt. Food Science & Nutrition, 10, 1451–­1460. https://doi.
Rattray, F. P., Myling-­Petersen, D., Larsen, D., & Nilsson, D. (2003).
org/10.1002/fsn3.2755
Plasmid-­encoded diacetyl (acetoin) reductase in Leuconostoc pseu-
domesenteroides. Applied and Environmental Microbiology, 69(1),
304–­311. https://doi.org/10.1128/AEM.69.1.304-­311.2003