M:Fo

odMi

crobio

logy

&Sa

fety

Investigation on Culturable Microflora in TibetanKefir Grains from Different Areas of ChinaJie Gao, Fengying Gu, Nesredin H Abdella, Hui Ruan, and Guoqing He

Abstract: Four samples of Tibetan kefir grains (TK-ZJUJ 01–04) from Tibet and surrounding areas were investigatedvia phenotypic and genotypic methods to compare and analyze the diversity of culturable microflora among differentorigins. As a result, 4 genera of microorganisms from TK-ZJUJ01: Bacillus subtilis (2.9 × 107 cfu/mL), Lactococcus lactis(8.2 × 107 cfu/mL), Kluyveromyces marxianus (3.0 × 106 cfu/mL), Saccharomyces cerevisiae (9.0 × 106 cfu/mL); 4 generafrom TK-ZJUJ02: Lactobacillus kefiri (1.0 × 108 cfu/mL), Pichia kudriavzevii (5.0 × 106 cfu/mL), K. marxianus (1.9 × 107

cfu/mL), Kazachstania unispora (6.2 × 107 cfu/mL); 6 genera from TK-ZJUJ03: Leuconostoc lactis (4.6 × 107 cfu/mL),L. lactis (3.0 × 107 cfu/mL), Lactobacillus plantarum (3.0 × 107 cfu/mL), K. unispora (3.0 × 106 cfu/mL), K. marxianus(2.0 × 106 cfu/mL), (1.7 × 107 cfu/mL); and 4 genera from TK-ZJUJ04: L. plantarum (1.8 × 107 cfu/mL), Acetobacterfabarum (5.0 × 106 cfu/mL), K. unispora (6.2 × 107 cfu/mL), Pichia guilliermondii (6.2 × 107 cfu/mL) were identified.Yeasts like P. kudriavzevii and P. guilliermondii isolated in this study were the first time reported in Tibetan kefir grains.For TK-ZJUJ 01–03, lactic acid bacteria were the major microorganisms, which accounted for more than 50% of all themicrobial population, while for TK-ZJUJ04, the largest microbial group was yeasts which accounted for more than 50%.In a word, study of diversity and composition of microflora provided us theoretical foundation for further investigationand application of Tibetan kefir grains.

Keywords: culturable microflora, lactic acid bacteria, phenotypic and genotypic methods, Tibetan kefir grains, yeast

Practical Application: This is the basic research in order to develop and industrialize a new kind of yogurt starter whichis naturally formed microbiota with both lactic acid bacteria and yeasts in it.

IntroductionTibetan kefir grain (TK) is a kind of natural starter for

fermented milk in Tibet, China, which is composed of variousmicroorganisms. The milk fermented by TK is a kind of self-carbonated yogurt, called Tibetan kefir, which is the traditionalfood of Tibetan people. As a kind of traditional food with along history, it appears not only in Tibet, but also in Taiwan,Russia, Turkey, Brazil, Argentina, and so on (Chen and others2008; De Antoni and others 2010; Magalhaes and others 2010;Kesmen and Kacmaz 2011; Plessas and others 2011). Statisticaldata show that people who have kefir as a diet are longevous.They find that probiotics in their gut are rich in type andquantity, while microbial components in gut are closely relatedto health (Eckburg and others 2005; Qin and others 2010). TKcan be considered as probiotic resource because it can enjoy avariety of health claims besides its nutritional status (Urdanetaand others 2007). A lot of studies about kefir’s biological activitieshave been done (Diniz and others 2003; Vinderola and others2005Rizk and others 2009; Silva and others 2009), which showthat kefir has antimicrobial activity, immunomodulating activity,antiinflammatory activity, antiproliferative activity, and so on, andhas the potential of becoming a kind of functional food.

MS 20111292 Submitted 10/25/2012, Accepted 5/14/2012. Authors are withthe College of Biosystems Engineering and Food Science, Zhejiang Univ., Hangzhou,310058, China. Direct inquiries to author He (E-mail: gqhe@zju.edu.cn).

While TK, as the starter of kefir in Tibet is drawing attentionfrom both domestic and overseas on its microbial components andapplication, but it has not been industrialized. Zhou and others(2009) have investigated that TK is a mixed system of yeasts andbacteria, which consists of Leuconostoc mesenteroides, Lactococcuslactis, Lactobacillus kefiri, Lactobacillus casei, Kluyveromyces marxianus,Saccharomyces unisporus, Saccharomyces cerevisiae, Candida humilis,and some uncultured microorganisms. The result also shows thatit is a complex system without fermentation stability. In orderto realize its industrialization and commercialization, the mostimportant thing is to know its microbial components.

Therefore, the objective of this paper is to investigate theculturable microflora of TK from different areas, find regulationsof grain’s formation and provide theoretical basis for furtherinvestigation and application of TK.

Materials and Methods

Samples and inoculation of TKTK samples were obtained from private households separately

in the city of Tibet and the surroundings, located in the westof China. They were TK-ZJUJ01 from Tibet, TK-ZJUJ02from Sichuan, TK-ZJUJ03 from Xinjiang and TK-ZJUJ04 fromQinghai. The grains were preserved in sterilized milk fromChina Mengniu Dairy at 4 ◦C, and activated at 28 ◦C, for 24 h(2 g grains with 100 mL milk), 3 times (De Antoni and others2010).

C© 2012 Institute of Food Technologists R©doi: 10.1111/j.1750-3841.2012.02805.x Vol. 77, Nr. 8, 2012 � Journal of Food Science M425Further reproduction without permission is prohibited

M:FoodMicrobiology&

Safety

Microflora in Tibetan kefir grains . . .

Observation of TK using scanning electronmicroscope(SEM)

TK samples were collected for microscopy (Guzel-Seydimand others 2005). For each sampling area, the TK was fixed in30 g/L glutaraldehyde in 0.1 M phosphate buffer, pH 7.0, for4 h at 25 ◦C. Samples were washed with phosphate buffer for15 min three times. Then grains were postfixed in 10 g/L osmiumtetroxide in phosphate buffer for 1 h at 25 ◦C. After washingwith phosphate buffer, samples were dehydrated in ethanol: 15%,30%, 50%, and 70% ethanol for 10 min each, 85% and 95% for15 min each, and 99.5% for 1 h. After dehydrating, samples werecritical-point dried and coated with gold. The preparations wereobserved using a scanning electron microscope (XL30ESEM;Philips Amsterdam, Netherlands).

Isolation and purification of bacteria and yeastFor each sample, 1 g of the grains was removed aseptically,

pestled and then diluted to 10−5, 10−6, 10−7 with 0.9% NaCl(w/v). Bacteria and yeasts were enumerated by the surface spreadtechnique (Magalhaes and others 2010). 100 µL of each dilutedsample and 3 different culture media were used for spread plate.Bacteria were enumerated in MRS (DeMan Rogosa-Sharpemedium) and AC (acetobacter medium) agar OXOID and.yeasts were enumerated on YPD (yeast extract peptone dextrosemedium) agar OXOID. After spreading, the AC plates wereincubated at 30 ◦C for 48 h, the MRS plates were incubated inanaerobic incubator at 37 ◦C for 48 h, and the YPD plates wereincubated at 28 ◦C for 48 h (Zhou 2002; Miguel and others2010). Chose the best dilution to get the mean colony formingunits (30 to 300 cfu on every plate was the best dilution) (Liu2006) and the single colonies for further tests.

Identification of bacteria and yeasts by conventionalmethods

The physiological and biochemical characteristics of differentmicroorganisms were investigated for identification of theirphenotypic characteristics using HANGWEI trace biochemicalreactions tube kits (Hangzhou Microbial Reagent Co., Ltd.Hangzhou, China) according to the manufacturer’s instructions.Phenotype characteristics of all bacterial isolates were determinedby the Bergey’s manual of systematic bacteriology (Garrity andothers 2004). Phenotype characteristics of all yeast isolates weredetermined by their morphology, spore formation, assimilation,and fermentation of different carbon sources according to Barnettand others (2000). Reference strains were used as active controland they were Lactobacillus plantarum lp15–2-1 CGMCC NO.3782, S. cerevisiae DG115 CGMCC NO.3262, Bacillus licheniformisZJUEL31410 CGMCC NO. 1397, and Acetobacter pasteurianusCGMCC NO. 1.41, which had been preserved in China GeneralMicrobiological Culture Collection Center (CGMCC) and alsopreserved in our laboratory.

Identification of bacteria and yeasts by molecular methodsThe bacterial and yeast DNA from pure cultures was extracted

according to the methods of X. F. Wang and others (2006) andMakimura and others (1999). Sequencing of portions of the16/26S rDNA gene was used for identification of genotypiccharacteristics of representative bacteria and yeasts. The forwardprimer was 338f (5′> ACT CCT ACG GGA GGC AGC AG<

3′), and the reverse primer was 518r (5′> ATT ACC GCG GCTGCT GG < 3′), for bacteria, and NL1 (5′>GCC ATA TCA

ATA AGC GGA GGA AAA g <3′), LS2 (5′> ATT CCC AAACAA CTC GAC TC <3′) for yeasts ( Zhou and others 2009),which were provided by Sangon company. A total of 50 µL PCRreaction system contained: 5 µL buffer solution, 4 µL MgCl2,4 µL dNTPs, 1 µL of each primer, 0.25 µL Taq polymerase,32.75 µL ddH2O, and 2 µL of the extracted DNA.

The PCR amplification was carried out as follows: for bacteria,template DNA was denatured for 5 min at 95 ◦C followed by30 cycles of denaturing at 94 ◦C for 30 s, annealing at 58 ◦Cfor 30 s and primer extension at 72 ◦C for 30 s. The tubes werethen incubated for 10 min at 72 ◦C for the final extension; foryeasts: template DNA was denatured for 5 min at 95 ◦C followedby 30 cycles of denaturing at 94 ◦C for 30 s, annealing at 55 ◦Cfor 30 s and primer extension at 72 ◦C for 30 s. The tubes werethen incubated for 10 min at 72 ◦C for the final extension.The amplification products were analyzed by electrophoresison 1.0% agarose gels before they were sequenced by InvitrogenCorporation. Get on line of www.ncbi.nlm.nih.gov/blast/ forsequence alignment. Each sequence datum was used as a querysequence to search for similar sequences from GenBank by meansof blast program. The multiple alignments and phylogenetic treeswere made with BioEdit 7.0.9 and MEGA 4.1.

Statistical analysisStatistical analysis was performed using SPSS 16.0. Data were

expressed as mean ± SD. Comparisons among replications inevery group were performed with ANOVA (analysis of variance)tests and differences were considered nonsignificant at P > 0.05.

Results and Discussion

Morphological characteristics analysis of TKMacroscopic morphological characteristics. No signifi-

cant difference was observed in TK samples from different areas inmorphology. They were all lobed, irregularly shaped, and whiteto yellow-white in color and had a slimy but firm texture, whichlooked like steamed rice accumulated together (Figure 1.). Theyvaried in size, however, always 2 to 3 cm in dia. TK samples fromdifferent areas could not be distinguished by macroscopic view.

Microscopic morphological characteristics. Scanningelectron microscopy of TK (Figure 2) indicated that, TK samplesfrom different areas were diverse in distribution, amount of

Figure 1–TK from different areas (1, TK-ZJUJ01; 2, TK-ZJUJ02; 3, TK-ZJUJ03; 4, TK-ZJUJ04).

M426 Journal of Food Science � Vol. 77, Nr. 8, 2012

M:Fo

odMi

crobio

logy

&Sa

fety

Microflora in Tibetan kefir grains . . .

microflora, and matrix. Each of them was composed of bacteriaand yeasts, which were sticking to the matrix. The schizogamy ofyeasts could be observed in all of the figures (A, B, C, D), whilein Figure 2A, there were more bacteria than yeasts, which lookedlike that the yeasts were inlaid in the bacteria; in Figure 2B,more yeasts were observed than in Figure 2A, and they were notcombined to bacteria so closely like Figure 2A, instead of beinginlaid, they were on the surface of the bacteria; in Figure 2C, a lot

of matrix was observed, and the yeasts were on the surface whilethe bacteria inside, in Figure 2D, it looked the same as Figure 2A,but more yeasts. So it could be deduced that yeasts were respon-sible to deplete oxygen and give a microaerophilic environment,and more yeasts could give bigger space, like in Figure 2B. Inaddition, TK samples from different areas were discrepant inmatrix production, as in Figure 2C. Obviously, more bacilli wereseen than cocci that could be due to the bad attachment of cocci,

Figure 2–Scanning electron micrographs of TK(A, TK-ZJUJ01; B, TK-ZJUJ02; C, TK-ZJUJ03; D,TK-ZJUJ04).

Table 1–Physiological and biochemical characteristics of 6 groups of LAB.

Strain Blank Active TK-ZJUJ01 TK-ZJUJ02 TK-ZJUJ03 TK-ZJUJ03 TK-ZJUJ03 TK-ZJUJ04characteristics control control∗ L3 L1 L1 L2 L3 L1

Colony color × Milky white Milky white Transparent Transparent Milky white Transparent TransparentSurface appearance × Smooth moist Smooth moist Smooth moist Smooth moist Smooth moist Smooth moist Smooth moistCell morphology × Bacilli Coccus Bacilli Coccus Coccus Bacilli BacilliGram faerbung test − + + + + + + +Catalase test − − − − − − − −Gelatin liquefaction test − − − − − − − −Amylohydrolysis test − − − − − − − −Hydrothion test − − − − − − − −Nitrate reduction test − − − − − − − −Aerogenesis test − − − − − − − −Milk curd test − + + + − + + +Indole test − − − − − − − −6% NaCl growth test − − − − + − − −8% NaCl growth test − − − − − − − −10 ◦C growth test − + + + + + + +45 ◦C growth test − + − − − − + +Glucose as carbon source − + + + + + + +Sucrose as carbon source − + − + + − + +Lactose as carbon source − + + − + + − −Maltose as carbon source − + + + − + − −Galactose as carbon source − + + + + + + +Mannose as carbon source − + − + − − − −Rhamnose as carbon source − − − − − − − −Melezitose as carbon source − + + − − + − −Esculin as carbon source − + + + + + − +Sorbierite as carbon source − − − − − − + −Xylose as carbon source − + − − − + − −D-fructose as carbon source − + + + + + + +Raffinose as carbon source − − − − − − − −Preliminary result × × Lactococcus Lactobacillus Leuconostoc Lactococcus Lactobacillus Lactobacillus∗ = Lactobacillus plantarum lp15–2-1 CGMCC NO. 3782; + = positive;−= negative; × = test or result not existing.

Vol. 77, Nr. 8, 2012 � Journal of Food Science M427

M:FoodMicrobiology&

Safety

Microflora in Tibetan kefir grains . . .

which was in agreement with the results obtained on Turkeyand Tibetan kefir grains using scanning electron microscope(Guzel-Seydim and others 2005; Zhou and others 2009).

Identification of phenotypic characteristics of differentmicroorganisms

By using the surface spread technique, 79 strains of isolationsof bacteria were obtained from AC media which were incubatedin incubator at 30 ◦C for 48 h; 318 strains of isolations of bacteriawere obtained from MRS media which were incubated in anaer-obic incubator at 37 ◦C for 48 h; 176 of yeasts were obtained fromYPD media which were incubated in incubator at 28 ◦C for 48 h.

According to phenotype characteristics determined by theBergey’s manual of systematic bacteriology (Barnett and others2000; Garrity 2004), all of the bacteria isolates could be dividedinto 9 groups [3 groups in TK-ZJUJ01: TK-ZJUJ01 L1 was agroup of white or cream-colored colonies with plication surface,gram positive and catalase positive, the cells were bacilliform, fur-ther physiological and biochemical tests showed that it was Bacillus(Table 2); TK-ZJUJ01 L2 was a group of off-white colonies withplication surface, gram positive and catalase positive, the cells werebacilliform, further physiological and biochemical tests showedthat it was Bacillus (Table 2); TK-ZJUJ01 L3 was a group of milkywhite colonies with smooth and moist surface, gram positive andcatalase negative, the cells were coccoid, further physiological andbiochemical tests showed that it was Lactococcus (Table 1). One

Table 2– Physiological and biochemical characteristics of2 groups of Bacillus.

Strain Blank Active TK-ZJUJ01 TK-ZJUJ01characteristics control control∗ L1 L2

Colony color × Off-white White or creamcolored

Off-white

Surfaceappearance

× Plication Plication Plication

Cellmorphology

× Bacilli Bacilli Bacilli

Gram faerbungtest

− + + +

Catalase test − + + +Gelatin

liquefactiontest

− + + +

V-P test − + + +Indole test − + + +50 ◦C growth

test− + + +

Caseinhydrolysis test

− + + +

Citric aciddeoxidizationtest

− + + +

Amylohydrolysistest

− + + +

Glucose ascarbon source

− + + +

Xylose as carbonsource

− + − −

Mannose ascarbon source

− + + −

Ammonium saltas nitrogensource

− + + +

Preliminaryresults

× × Bacillus Bacillus

∗ = Bacillus licheniformis ZJUEL31410 CGMCC NO. 1397; + = positive; − = nega-tive; × = test or result not existing.

group in TK-ZJUJ02: TK-ZJUJ02 L1 was a group of transparentcolonies with smooth and moist surface, gram positive and cata-lase negative, the cells were bacilliform, further physiological andbiochemical tests showed that it was Lactobacillus (Table 1). Threegroups in TK-ZJUJ03: TK-ZJUJ03 L1 was a group of transpar-ent colonies with smooth and moist surface, gram positive andcatalase negative, the cells were coccoid, further physiological andbiochemical tests showed that it was Leuconostoc (Table 1); TK-ZJUJ03 L2 was a group of milky white colonies with smooth andmoist surface, gram positive and catalase negative, the cells werecoccoid, further physiological and biochemical tests showed that itwas Lactococcus (Table 1); TK-ZJUJ03 L3 was a group of transparentcolonies with smooth and moist surface, gram positive and cata-lase negative, the cells were bacilliform, further physiological andbiochemical tests showed that it was Lactobacillus (Table 1). Twogroups in TK-ZJUJ04: TK-ZJUJ04 L1 was a group of transparentcolonies with smooth and moist surface, gram positive and cata-lase negative, the cells were bacilliform, further physiological andbiochemical tests showed that it was Lactobacillus (Table 1); TK-ZJUJ04 L2 was a group of white and opaque colonies with smoothsurface, gram negative and catalase positive, the cells were bacil-liform, further physiological and biochemical tests showed that itwas Acetobacter (Table 3).], and all of the yeast isolates were dividedinto 21 groups [7 groups in TK-ZJUJ01 (Table 4): TK-ZJUJ01 Y1was a group of milky white colonies with smooth surface, the cellswere oblong, budding reproduction, further physiological and bio-chemical tests showed that it was Kluyveromyces; TK-ZJUJ01 Y2,Y3, and Y4 were similar to Y1 and also identified as Kluyveromyces;TK-ZJUJ01 Y5 was a group of white colonies with smooth sur-face, the cells were roundness, budding reproduction, further phys-iological and biochemical tests showed that it was Saccharomyces;TK-ZJUJ01 Y6 and Y7 were similar to Y5 and also identified asSaccharomyces. Eight groups in TK-ZJUJ02 (Table 5): TK-ZJUJ02Y1 was a group of milky white colonies with rough surface, thecells were oblong, budding reproduction, further physiologicaland biochemical tests showed that it was Pichia; TK-ZJUJ02 Y2

Table 3–Physiological and biochemical characteristics of 1 groupof Acetobacter.

Strain Blank Active TK-ZJUJ04characteristics control control∗ A1

Colony color × Cream color andopaque

White and opaque

Surface appearance × Smooth SmoothCell morphology × Oblong Bacilli or oblongGram faerbung test − − −Catalase test − + +Gelatin liquefaction

test− − −

Oxidation of ethanol − − +Oxidation of acetic

acid− − +

Methyl red test − − +V-P test − − −Fibrinolysis test − − −30% glucose growth

test− − −

Citric aciddeoxidization test

− − −

Acid materialproduction

− + +

Amylohydrolysis test − − −Preliminary result × × Acetobacter∗ = Acetobacter pasteurianus CGMCC NO. 1.41; + = positive; − = negative; × = testor result not existing.

M428 Journal of Food Science � Vol. 77, Nr. 8, 2012

M:Fo

odMi

crobio

logy

&Sa

fety

Microflora in Tibetan kefir grains . . .

was a group of off-white colonies with smooth surface, the cellswere oval, budding reproduction, further physiological and bio-chemical tests showed that it was Kluyveromyces; TK-ZJUJ02 Y3and Y4 were similar to Y2 and also identified as Kluyveromyces;TK-ZJUJ02 Y5 was a group of milky white colonies with smoothand upward surface, the cells were roundness, budding reproduc-tion, further physiological and biochemical tests showed that it wasKazachstania; TK-ZJUJ02 Y6, Y7, and Y8 were similar to Y5 andalso identified as Kazachstania. Four groups in TK-ZJUJ03 (Ta-ble 6): TK-ZJUJ03 Y1 was a group of milky white colonies withsmooth and upward surface, the cells were roundness, buddingreproduction, further physiological and biochemical tests showedthat it was Kazachstania; TK-ZJUJ03 Y2 was a group of off-whitecolonies with smooth surface, the cells were oval, budding repro-duction, further physiological and biochemical tests showed thatit was Kluyveromyces; TK-ZJUJ03 Y3 was similar to Y2 and alsoidentified as Kluyveromyces; TK-ZJUJ03 Y4 was similar to Y1 and

also identified as Kazachstania. Two groups in TK-ZJUJ04 (Ta-ble 6): TK-ZJUJ04 Y1 was a group of milky white colonies withsmooth and upward surface, the cells were oval, budding reproduc-tion, further physiological and biochemical tests showed that it wasKazachstania; TK-ZJUJ04 Y2 was a group of off-white colonieswith rough and upward surface, the cells were oval, budding repro-duction, further physiological and biochemical tests showed thatit was Pichia.]. All of the microorganisms were identified to genusvia phenotypic and all of the groups mentioned above would beinvestigated by genotypic methods for the species identification.

Identification of genotypic characteristics of differentmicroorganisms

The PCR products were used as query sequences to searchfor similar sequences from GenBank by means of blast program.The multiple alignments and phylogenetic trees were made withBioEdit 7.0.9 and MEGA 4.1, as shown in Figure 3 to 6.

Table 4–Physiological and biochemical characteristics of 7 groups of yeasts in TK-ZJUJ01.

Strain Blank Active TK-ZJUJ01 TK-ZJUJ01 TK-ZJUJ01 TK-ZJUJ01 TK-ZJUJ01 TK-ZJUJ01 TK-ZJUJ01characteristics control control∗ Y1 Y2 Y3 Y4 Y5 Y6 Y7

Colony color × Milky white Milky white Off-white Milky white Milky white White White WhiteSurface appearance × Smooth Smooth Smooth Smooth Smooth Smooth Smooth SmoothCell morphology × Oval Oblong Oval Oblong Oblong Roundness Oval RoundnessVegetative propagation × Budding Budding Budding Budding Budding Budding Budding BuddingBallistospore test × − − − − − − − −Mycelium growth test × Pseudohypha Pseudohypha Pseudohypha Pseudohypha Pseudohypha Pseudohypha Pseudohypha PseudohyphaUrease test − − − − − − − − −Nitrate reduction test − − − − − − − − −Glucose as carbon source − + + + + + + + +Sucrose as carbon source − + + + + − + + −Lactose as carbon source − + − + + + − − −Maltose as carbon source − + − − − − + + +Xylose as carbon source − + − − − − − − −Galactose as carbon source − + + − + + + − +Melibiose as carbon source − − − − − − − − −Rhamnose as carbon source − − − − − − − − −Preliminary result × × Kluyveromyces Kluyveromyces Kluyveromyces Kluyveromyces Saccharomyces Saccharomyces Saccharomyces∗ = Saccharomyces cerevisiae DG115 CGMCC NO.3262; + = positive; − = negative; × = test or result not existing.

Table 5–Physiological and biochemical characteristics of 8 groups of yeasts in TK-ZJUJ02.

Strain Blank Active TK-ZJUJ02 TK-ZJUJ02 TK-ZJUJ02 TK-ZJUJ02 TK-ZJUJ02 TK-ZJUJ02 TK-ZJUJ02 TK-ZJUJ02characteristics control control∗ Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8

Colony color × Milky white Milky white Off-white Milky white Milky white Milky white Milky white Milky white Milky whiteSurface appearance × Smooth Rough Smooth Smooth Smooth Smooth and

upwardSmooth Smooth and

upwardSmooth

Cell morphology × Oval Oblong Oval Oblong Oblong Roundness Roundness Roundness RoundnessVegetative

propagation× Budding Budding Budding Budding Budding Budding Budding Budding Budding

Ballistospore test × − − − − − − − − −Mycelium growth

test× Pseudohypha Pseudohypha Pseudohypha pseudohypha Pseudohypha Pseudohypha Pseudohypha Pseudohypha Pseudohypha

Urease test − − − − − − − − − −Nitrate reduction test − − − − − − − − − −Glucose as carbon

source− + + + + + + + + +

Sucrose as carbonsource

− + − + + − − − − −

Lactose as carbonsource

− + − + + + − − − −

Maltose as carbonsource

− + − − − − − + − +

Xylose as carbonsource

− + − − − − − − − −

Galactose as carbonsource

− + − − + + + − + +

Melibiose as carbonsource

− − − − − − − − − −

Rhamnose as carbonsource

− − − − − − − − − −

Preliminary result × × Pichia Kluyveromyces Kluyveromyces Kluyveromyces Kazachstania Kazachstania Kazachstania Kazachstania∗ = Saccharomyces cerevisiae DG115 CGMCC NO.3262; + = positive; − = negative; × = test or result not existing.

Vol. 77, Nr. 8, 2012 � Journal of Food Science M429

M:FoodMicrobiology&

Safety

Microflora in Tibetan kefir grains . . .

Physiological, biochemical characteristics, and molecular meth-ods revealed that, 4 genera of microorganisms were isolated fromTK-ZJUJ01, including Bacillus subtilis, Lactococcus lactis, K. marx-ianus, S. cerevisiae; 4 genera of microorganisms were isolated fromTK-ZJUJ02, L. kefiri, Pichia kudriavzevii, K. marxianus, Kazach-stania unispora; 6 genera of microorganisms were isolated fromTK-ZJUJ03, Leuconostoc lactis, L. lactis, L. plantarum, K. unispora,K. marxianus, Kazachstania exigua; 4 genera of microorganismswere isolated from TK-ZJUJ04, L. plantarum, Acetobacter fabarum,K. unispora, Pichia guilliermondii. P. kudriavzevii and P. guilliermondiiisolated in this study were the first time reported in TK.

For every of the grains, lactic acid bacteria were responsible forlactic acid fermentation and milk curd, as L. lactis detected fromTK-ZJUJ01, L. kefiri detected from TK-ZJUJ02, Leu. lactis, Lac.lactis, and L. plantarum detected from TK-ZJUJ03, L. plantarum,detected from TK-ZJUJ04. Although most of them were anaer-obic or facultative anaerobe, the grains could be cultured in air.So, we could figure out that there were some aerobes coexisting

in the grains to deplete oxygen and give a microaerophilicenvironment, like yeasts and bacillus we had found in the grains,in addition, acid resistance should also be taken into consideration.Besides the microenvironment established by the microorganisms,the formation of the grains aroused our concern, too. Studieshad shown that the grains matrix was comprised of proteinsand polysaccharides (Wang and others 2008), and accordantly,B. subtilis, L. kefiri, L. lactis, L. plantarum, found in every grain,respectively, were typically species to produce EPS with extensiveattention (Hemme and Foucaud-Scheunemann 2004; Wang andothers 2008; Wang and others 2010; Fang and others 2011)

Composition and proportion of microbialWe could get the mean colony count of every species from the

number of mean colony forming units and the identification results(Table 7). Both the species and amount of yeasts in TK-ZJUJ02and TK-ZJUJ03 were more than in the other 2 TK samples, whichwas in accordance with the electromicrography results.

Table 6–Physiological and biochemical characteristics of 6 groups of yeasts in TK-ZJUJ03 and TK-ZJUJ04.

Strain Blank Active TK-ZJUJ03 TK-ZJUJ03 TK-ZJUJ03 TK-ZJUJ03 TK-ZJUJ04 TK-ZJUJ04characteristics control control∗ Y1 Y2 Y3 Y4 Y1 Y2

Colony color × Milky white Milky white Off-white Milky white Milky white Milky white Off-whiteSurface appearance × Smooth Smooth and

upwardSmooth Smooth Smooth Smooth and

upwardRough and upward

Cell morphology × Oval Roundness oroval

Oval Oblong Roundness oroval

Roundness oroval

Oval

Vegetative propagation × Budding Budding Budding Budding Budding Budding BuddingBallistospore test × − − − − − − −Mycelium growth test × Pseudohypha Pseudohypha Pseudohypha Pseudohypha Pseudohypha Pseudohypha PseudohyphaUrease test − − − − − − − −Nitrate reduction test − − − − − − − −Glucose as carbon source − + + + + + + +Sucrose as carbon source − + − + + + + −Lactose as carbon source − + − − − + − −Maltose as carbon source − + − − − − − −Xylose as carbon source − + − − − − − −Galactose as carbon source − + + + + + + −Melibiose as carbon source − − − − − − − −Rhamnose as carbon source − − − − − − − −Preliminary result × × Kazachstania Kluyveromyces Kluyveromyces Kazachstania Kazachstania Pichia∗ = Saccharomyces cerevisiae DG115 CGMCC NO.3262; + = positive; − = negative; × = test or result not existing.

TK-ZJUJ01 Y1TK-ZJUJ01 Y3TK-ZJUJ01 Y2TK-ZJUJ01 Y4

Kluyveromyces marxianus strain DX3-4

Kluyveromyces lactis isolate G113

Saccharomyces cerevisiae strain D53

TK-ZJUJ01 Y5TK-ZJUJ01 Y6TK-ZJUJ01 Y7

Pichia kluyveri strain SL2-1-1

94

71

80

0.05 B

TK-ZJUJ01 L1Bacillus subtilis strain VSG-4

TK-ZJUJ01 L2Bacillus pumilus isolate 4.4.2

TK-ZJUJ01 L3Lactococcus lactis strain 36-1

96

100

0.02A

TK- Y1TK-ZJUJ01 Y3TK-ZJUJ01 Y2TK-ZJUJ01 Y4

Kluyveromyces marxianus strain DX3-4

Kluyveromyces lactis isolate G113

Saccharomyces cerevisiae strain D53

TK-ZJUJ01 Y5TK-ZJUJ01 Y6TK-ZJUJ01 Y7

Pichia kluyveri strain SL2-1-1

94

71

80

B

TK- Y1TK-ZJUJ01 Y3TK-ZJUJ01 Y2TK-ZJUJ01 Y4

Kluyveromyces marxianus strain DX3-4

Kluyveromyces lactis isolate G113

Saccharomyces cerevisiae strain D53

TK-ZJUJ01 Y5TK-ZJUJ01 Y6TK-ZJUJ01 Y7

Pichia kluyveri strain SL2-1-1

94

71

80

- Y1TK-ZJUJ01 Y3TK-ZJUJ01 Y2TK-ZJUJ01 Y4

Kluyveromyces marxianus strain DX3-4

Kluyveromyces lactis isolate G113

Saccharomyces cerevisiae strain D53

TK-ZJUJ01 Y5TK-ZJUJ01 Y6TK-ZJUJ01 Y7

Pichia kluyveri strain SL2-1-1

94

71

80

B

--

TK-

A

--

TK-

--

TK-

A

Figure 3–Relationships of partial 16S rDNA forbacteria (A) and 26S rDNA for yeasts (B) seq-uences isolated from TK-ZJUJ01 to those ofreference organisms obtained from GenBank.

M430 Journal of Food Science � Vol. 77, Nr. 8, 2012

M:Fo

odMi

crobio

logy

&Sa

fety

Microflora in Tibetan kefir grains . . .

TK-ZJUJ02 L1Lactobacillus kefiri strain NM180-3Lactobacillus plantarum clone WWC C3ALM0

Lactobacillus acidophilus strain BCRC106

99

0.01 A

TK-ZJUJ02 Y6TK-ZJUJ02 Y7TK-ZJUJ02 Y8

Kazachstania unispora strain D190

TK-ZJUJ02 Y5Saccharomyces cerevisiae strain D53

TK-ZJUJ02 Y2TK-ZJUJ02 Y3TK-ZJUJ02 Y4

Kluyveromyces marxianus strain DX3-4

TK-ZJUJ02 Y1Pichia kudriavzevii strain SL1-7100

99

98

98

0.01B

TK-ZJUJ02 L1Lactobacillus kefiri strain NM180-3Lactobacillus plantarum clone WWC C3ALM0

Lactobacillus acidophilus strain BCRC106

99

0.01 A

TK-ZJUJ02 L1Lactobacillus kefiri strain NM180-3Lactobacillus plantarum clone WWC C3ALM0

Lactobacillus acidophilus strain BCRC106

99

0.01

TK-ZJUJ02 L1Lactobacillus kefiri strain NM180-3Lactobacillus plantarum clone WWC C3ALM0

Lactobacillus acidophilus strain BCRC106

99

0.01 A

TK-ZJUJ02 Y6TK-ZJUJ02 Y7TK-ZJUJ02 Y8

Kazachstania unispora strain D190

TK-ZJUJ02 Y5Saccharomyces cerevisiae strain D53

TK-ZJUJ02 Y2TK-ZJUJ02 Y3TK-ZJUJ02 Y4

Kluyveromyces marxianus strain DX3-4

TK-ZJUJ02 Y1Pichia kudriavzevii strain SL1-7100

99

98

98

0.01B

TK-ZJUJ02 Y6TK-ZJUJ02 Y7TK-ZJUJ02 Y8

Kazachstania unispora strain D190

TK-ZJUJ02 Y5Saccharomyces cerevisiae strain D53

TK-ZJUJ02 Y2TK-ZJUJ02 Y3TK-ZJUJ02 Y4

Kluyveromyces marxianus strain DX3-4

TK-ZJUJ02 Y1Pichia kudriavzevii strain SL1-7100

99

98

98

0.01

TK-ZJUJ02 Y6TK-ZJUJ02 Y7TK-ZJUJ02 Y8

Kazachstania unispora strain D190

TK-ZJUJ02 Y5Saccharomyces cerevisiae strain D53

TK-ZJUJ02 Y2TK-ZJUJ02 Y3TK-ZJUJ02 Y4

Kluyveromyces marxianus strain DX3-4

TK-ZJUJ02 Y1Pichia kudriavzevii strain SL1-7100

99

98

98

0.01B

Figure 4–Relationships of partial 16S rDNA forbacteria (A) and 26S rDNA for yeasts (B)sequences isolated from TK-ZJUJ02 to those ofreference organisms obtained from GenBank.

TK-ZJUJ03L1Leuconostoc lactis strain NM173-6

TK-ZJUJ03L2Lactococcus lactis strain 36-1

TK-ZJUJ03L3Lactobacillus plantarum strain IMAU70089

100

100

100

0.02A

TK-ZJUJ03Y1Kazachstania unispora strain D190

Saccharomyces cerevisiae strain D53

TK-ZJUJ03Y4Kazachstania exigua strain CBS 135

Kluyveromyces marxianus strain DX3-4

TK-ZJUJ03Y2

TK-ZJUJ03Y3100

100

99

78

0.01

B

TK-ZJUJ03L1Leuconostoc lactis strain NM173-6

TK-ZJUJ03L2Lactococcus lactis strain 36-1

TK-ZJUJ03L3Lactobacillus plantarum strain IMAU70089

100

100

100

0.02A

TK-ZJUJ03L1Leuconostoc lactis strain NM173-6

TK-ZJUJ03L2Lactococcus lactis strain 36-1

TK-ZJUJ03L3Lactobacillus plantarum strain IMAU70089

100

100

100

0.02

TK-ZJUJ03L1Leuconostoc lactis strain NM173-6

TK-ZJUJ03L2Lactococcus lactis strain 36-1

TK-ZJUJ03L3Lactobacillus plantarum strain IMAU70089

100

100

100

0.02A

TK-ZJUJ03Y1Kazachstania unispora strain D190

Saccharomyces cerevisiae strain D53

TK-ZJUJ03Y4Kazachstania exigua strain CBS 135

Kluyveromyces marxianus strain DX3-4

TK-ZJUJ03Y2

TK-ZJUJ03Y3100

100

99

78

0.01

B

TK-ZJUJ03Y1Kazachstania unispora strain D190

Saccharomyces cerevisiae strain D53

TK-ZJUJ03Y4Kazachstania exigua strain CBS 135

Kluyveromyces marxianus strain DX3-4

TK-ZJUJ03Y2

TK-ZJUJ03Y3100

100

99

78

0.01

TK-ZJUJ03Y1Kazachstania unispora strain D190

Saccharomyces cerevisiae strain D53

TK-ZJUJ03Y4Kazachstania exigua strain CBS 135

Kluyveromyces marxianus strain DX3-4

TK-ZJUJ03Y2

TK-ZJUJ03Y3100

100

99

78

0.01

B

Figure 5–Relationships of partial 16S rDNA forbacteria (A) and 26S rDNA for yeasts (B) seq-uences isolated from TK-ZJUJ03 to those ofreference organisms obtained from GenBank.

TK-ZJUJ04 L1Lactobacillus plantarum clone WWC C3ALM0Lactobacillus kefiri strain NM180-3Lactobacillus acidophilus strain BCRC106

A

TK-ZJUJ04 A1Acetobacter fabarum strain NM118-1Acetobacter pasteurianus strain NM73-1Acetobacter malorum strain NM156-4

71

B

TK-ZJUJ04 Y1Kazachstania unispora strain D190

Saccharomyces cerevisiae strain D53Pichia kluyveri strain SL2-1-1

TK-ZJUJ04 Y2Pichia guilliermondii strain CXF-1

C

TK-ZJUJ04 L1Lactobacillus plantarum clone WWC C3ALM0Lactobacillus kefiri strain NM180-3Lactobacillus acidophilus strain BCRC106

A

TK-ZJUJ04 L1Lactobacillus plantarum clone WWC C3ALM0Lactobacillus kefiri strain NM180-3Lactobacillus acidophilus strain BCRC106

TK-ZJUJ04 L1Lactobacillus plantarum clone WWC C3ALM0Lactobacillus kefiri strain NM180-3Lactobacillus acidophilus strain BCRC106

100

0.02A

TK-ZJUJ04 A1Acetobacter fabarum strain NM118-1Acetobacter pasteurianus strain NM73-1Acetobacter malorum strain NM156-4

71

B

TK-ZJUJ04 A1Acetobacter fabarum strain NM118-1Acetobacter pasteurianus strain NM73-1Acetobacter malorum strain NM156-4

71

TK-ZJUJ04 A1Acetobacter fabarum strain NM118-1Acetobacter pasteurianus strain NM73-1Acetobacter malorum strain NM156-4

71

TK-ZJUJ04 A1Acetobacter fabarum strain NM118-1Acetobacter pasteurianus strain NM73-1Acetobacter malorum strain NM156-4

71

TK-ZJUJ04 A1Acetobacter fabarum strain NM118-1Acetobacter pasteurianus strain NM73-1Acetobacter malorum strain NM156-4

71

0.002 B

TK-ZJUJ04 Y1Kazachstania unispora strain D190

Saccharomyces cerevisiae strain D53Pichia kluyveri strain SL2-1-1

TK-ZJUJ04 Y2Pichia guilliermondii strain CXF-1

C

TK-ZJUJ04 Y1Kazachstania unispora strain D190

Saccharomyces cerevisiae strain D53Pichia kluyveri strain SL2-1-1

TK-ZJUJ04 Y2Pichia guilliermondii strain CXF-1

TK-ZJUJ04 Y1Kazachstania unispora strain D190

Saccharomyces cerevisiae strain D53Pichia kluyveri strain SL2-1-1

TK-ZJUJ04 Y2Pichia guilliermondii strain CXF-199

7174

0.05C

Figure 6–Relationships of partial 16S rDNA forbacteria (A, B) and 26S rDNA for yeasts (C) seq-uences isolated from TK-ZJUJ04 to those ofreference organisms obtained from GenBank.

Vol. 77, Nr. 8, 2012 � Journal of Food Science M431

M:FoodMicrobiology&

Safety

Microflora in Tibetan kefir grains . . .

Table 7–Mean colony counts of isolated species in every sample.

Isolated TK-ZJUJ01 TK-ZJUJ02 TK-ZJUJ03 TK-ZJUJ04species (cfu/mL) (cfu/mL) (cfu/mL) (cfu/mL)

Bacillus subtilis (2.9 ± 0.01) × 107∗ – – –Leuconostoc lactis – – (4.6 ± 0.11) × 107∗ –Lactococcus lactis (8.2 ± 0.06) × 107∗ – (3.0 ± 0.01) × 107∗ –Lactobacillus kefiri – (1.0 ± 0.01) × 108∗ – –Lactobacillus plantarum – – (3.0 ± 0.02) × 107∗ (1.8 ± 0.01) × 107∗

Acetobacter fabarum – – – (5.0 ± 0.03) × 106∗

Kluyveromyces marxianus (3.0 ± 0.01) × 106∗ (1.9 ± 0.02) × 107∗ (2.0 ± 0.02) × 106∗ –Saccharomyces cerevisiae (9.2 ± 0.21) × 106∗ – – –Kazachstania unispora – (6.2 ± 0.09) × 107∗ (3.0 ± 0.09) × 106∗ (6.2 ± 0.12) × 107∗

Kazachstania exigua – – (1.7 ± 0.01) × 107∗ –Pichia kudriavzevii – (5.0 ± 0.14) × 106∗ – –Pichia guilliermondii – – – (2.0 ± 0.04) × 106∗

— = nonexistence; ± = SD; ∗ = differences were nonsignificant among replications in each group.

Figure 7–Distribution frequency of theculturable microbial population ofTK-ZJUJ01–04.

The distribution frequencies of the culturable bacteria andyeast strains from the TK belonged to 4 areas after 3 d activationwere shown in Figure 7. The results obtained in this study clearlydemonstrated that the culturable microbial species composition ofTK varied in different origins (Figure 7). There were both gener-ality and diversity in the culturable microbial species compositionof TK from different origins. The generality was that both yeastsand lactic acid bacteria were contained, sometimes acetobacteralso, while the diversity was that culturable microbial species com-position varied and no common microorganism could be foundin all TK samples. For TK-ZJUJ 01–03, lactic acid bacteria werethe major microorganisms, which accounted for more than 50%of all the microbial population, while for TK-ZJUJ04, the largestmicrobial group was yeasts which accounted for more than 50%.Besides lactic acid bacteria and yeasts, there were still some others,like B. subtilis and A. fabarum. In addition, the results confirmedthe microbial diversity in TK, in favor of microbial resourcesdevelopment and utilization. There was no mutual species in allof the grains which was in agreement with the findings aboutkefir grains from 10 different sources (Miguel and others 2010),but some regulations still could be found, which might be due to

factors such as origins, milk type, and culture methods in differenthouseholds.

ConclusionsThe results in this study demonstrated the microbial diversity of

TK. This was the 1st to report the presence of P. kudriavzevii and P.guilliermondii in TK. We could conclude that, in a grain, differentmicroorganism species took different part during its formation.As the discussion above, there should be microbe for lactic acidfermentation, oxygen depletion, and EPS production, which werethe basis factors for a grain’s formation. Knowing their regulationand process of formation made the recombination possible. Therecombined grain with known microbe and amount could providemore comprehensive functions and had fermentation stability, andonly in this way could it be industrialized. In a word, the researchand development of TK would give new energies to dairy industry.

AcknowledgmentsThe authors are thankful to those who sent the kefir samples

for this research and this work is supported by the High-TechResearch and Development Program of China (2006AA10Z316).

M432 Journal of Food Science � Vol. 77, Nr. 8, 2012

M:Fo

odMi

crobio

logy

&Sa

fety

Microflora in Tibetan kefir grains . . .

Conflict of InterestThe authors declare that they have no conflict of interest.

ReferencesBarnett JA, Payne RW, Yarrow D. 2000. Yeast – Characteristic and Identification. 3rd ed.

Cambridge: Cambridge University Press.Chen HC, Wang SY, Chen MJ. 2008. Microbiological study of lactic acid bacteria in kefir grains

by culture-dependent and culture-independent methods. Food Microbiol 25(3):492–501.DOI: 10.1016/J.Fm.2008.01.003.

De Antoni G, Zago M, Vasek O, Giraffa G, Carminati D, Marco MB, Suarez V. 2010. Lac-tobacillus plantarum bacteriophages isolated from Kefir grains: phenotypic and molecularcharacterization. J Dairy Res 77(1):7–12. DOI: 10.1017/S0022029909990203.

Diniz RO, Garla LK, Schneedorf JM, Carvalho JCT. 2003. Study of anti-inflammatory ac-tivity of Tibetan mushroom, a symbiotic culture of bacteria and fungi encapsulated into apolysaccharide matrix. Pharmacol Res 47(1):49–52.

Eckburg PB, Bik EM, Bernstein CN, Purdom E, Dethlefsen L, Sargent M, Relman DA.2005. Diversity of the human intestinal microbial flora. Science 308(5728):1635–8. DOI:10.1126/science.1110591.

Fang LC, Wei X, Cai P, Huang QY, Chen H, Liang W, Rong XM. 2011. Role of extracel-lular polymeric substances in Cu(II) adsorption on Bacillus subtilis and Pseudomonas putida.Bioresource Technology 102(2): 1137–41. DOI: 10.1016/j.biortech.2010.09.006.

Garrity GM, Bell JA, Lilburn TG. 2004. Bergey’s manual of systematic bacteriology. 2nd ed.New York, Berlin, Heidelberg: Springer.

Guzel-Seydim Z, Wyffels JT, Seydim AC, Greene AK. 2005. Turkish kefir and kefir grains:microbial enumeration and electron microscobic observation. Int J Dairy Technol 58(1):25–9.

Hemme D, Foucaud-Scheunemann C. 2004. Leuconostoc, characteristics, use in dairytechnology and prospects in functional foods. Int Dairy J 14(6):467–94. DOI:10.1016/j.idairyj.2003.10.005.

Kesmen Z, Kacmaz N (2011). Determination of lactic microflora of kefir grains and kefir beverageby using culture-dependent and culture-independent methods. J Food Sci 76(5):M276–83.DOI: 10.1111/j.1750–3841.2011.02191.x.

Hui L. 2006. The modern food microbiology experiment technology. Beijng, China: ChinaLight Industry Press.

Magalhaes KT, Pereira GVD, Dias DR, Schwan RF. 2010. Microbial communities and chem-ical changes during fermentation of sugary Brazilian kefir. World J Microbiol Biotechnol26(7):1241–50. DOI: 10.1007/s11274–009-0294-x.

Makimura K, Tamura Y, Mochizuki T, Hasegawa A, Tajiri Y, Hanazawa R, Yamaguchi H.1999. Phylogenetic classification and species identification of dermatophyte strains based onDNA sequences of nuclear ribosomal internal transcribed spacer 1 regions. J Clin Microbiol37(4):920–4.

Miguel MGDP, Cardoso PG, Lago LD, Schwan RF. 2010. Diversity of bacteria present inmilk kefir grains using culture-dependent and culture-independent methods. Food Res Int43(5):1523–8. DOI: 10.1016/j.foodres.2010.04.031.

Plessas S, Alexopoulos A, Bekatorou A, Mantzourani I, Koutinas AA, Bezirtzoglou E. 2011.Examination of freshness degradation of sourdough bread made with kefir through mon-itoring the aroma volatile composition during storage. Food Chem 124(2):627–33. DOI:10.1016/j.foodchem.2010.06.086.

Qin JJ, Li RQ, Raes J, Arumugam M, Burgdorf KS, Manichanh C, Consortium, MetaHIT.2010. A human gut microbial gene catalogue established by metagenomic sequencing. Nature464(7285):59-U70. DOI: 10.1038/Nature08821.

Rizk S, Maalouf K, Baydoun E. 2009. The antiproliferative effect of kefir cell-free frac-tion on HuT-102 malignant T lymphocytes. Clin Lymphoma Myeloma 9:S198–203. DOI:10.3816/Clm.2009.S.012.

Silva KR, Rodrigues SA, Xavier L, Lima AS. 2009. Antimicrobial activity ofbroth fermented with kefir grains. Appl Biochem Biotechnol 152(2):316–25. DOI:10.1007/s12010–008-8303–3.

Urdaneta E, Barrenetxe J, Aranguren P, Irigoyen A, Marzo F, Ibanez FC. 2007. Intesti-nal beneficial effects of kefir-supplemented diet in rats. Nutr Res 27(10):653–58. DOI:10.1016/j.nutres.2007.08.002.

Vinderola CG, Duarte J, Thangavel D, Perdigon G, Farnworth E, Matar C. 2005. Immunomod-ulating capacity of kefir. J Dairy Res 72(2):195–202. DOI: 10.1017/S0022029905000828.

Wang XF, Haruta S, Wang P, Ishii M, Igarashi Y, Cui ZJ. 2006. Diversity of a stable enrichmentculture which is useful for silage inoculant and its succession in alfalfa silage. Fems MicrobiolEcol 57(1):106–15. DOI: 10.1111/j.1574–6941.2006.00099.x.

Wang YP, Ahmed Z, Feng W, Li C, Song SY. 2008. Physicochemical properties of exopolysac-charide produced by Lactobacillus kefiranofaciens ZW3 isolated from Tibet kefir. Int J BiolMacromol 43(3):283–8. DOI: 10.1016/j.ijbiomac.2008.06.011.

Wang YP, Li C, Liu P, Ahmed Z, Xiao P, Bai XJ. 2010. Physical characterization of exopolysac-charide produced by Lactobacillus plantarum KF5 isolated from Tibet Kefir. Carbohydr.Polym. 82(3):895–903. DOI: 10.1016/j.carbpol.2010.06.013.

Zhou DQ. 2002. Microbiology Course. Beijing, China: Higher Education Press.Zhou JZ, Liu XL, Jiang HH, Dong MS. 2009. Analysis of the microflora in Tibetan kefir

grains using denaturing gradient gel electrophoresis. Food Microbiol 26(8):770–5. DOI:10.1016/J.Fm.2009.04.009.

Vol. 77, Nr. 8, 2012 � Journal of Food Science M433