Effect of ultrafiltered fractions from casein on lactic ...download.xuebalib.com/xuebalib.com.40037.pdf · Original article Effect of ultrafiltered fractions from casein on lactic

Original article

Effect of ultrafiltered fractions from casein on lactic acid

biosynthesis and enzyme activity in yoghurt starter cultures

Qingli Zhang,1,2 Mindy M. Brashears,2 Zhimin Yu,3 Jiaoyan Ren,1 Yinjuan Li1 & Mouming Zhao1*

1 College of Light Industry and Food Sciences, South China University of Technology, Guangzhou 510640, China

2 Department of Animal and Food Sciences, Texas Tech University, Lubbock, TX 79409, USA

3 School of Biological Engineering, Dalian Polytechnic University, Dalian 116038, China

(Received 5 September 2012; Accepted in revised form 19 January 2013)

Summary In this work, five ultrafiltered fractions (UFs) with molecular mass less than 3 kDa (kilo-daltons) from

casein hydrolysates treated with alcalase, flavourzyme, neutrase, papain and trypsin, respectively, were

obtained. The effect of five UFs on the fermentation for lactic acid (LA) production by mixed cultures of

Streptococcus thermophilus (St) and Lactobacillus delbrueckii subsp. bulgaricus (Lb) during 72 h of cultiva-

tion was investigated. Compared with the control, LA production was, respectively, enhanced by 23.66%,

39.01%, 29.74%, 49.64% and 47.40% with the supplement of UF-A, UF-F, UF-N, UF-P and UF-T at

24 h. The possible mechanism of LA production enhanced was elucidated by the time course analysis of

the specific activity of glucokinase, phosphoglucose isomerase, 6-phosphofructokinase, pyruvate kinase

and lactate dehydrogenase during fermentation process. In addition, the results obtained showed the

diverse influence of five UFs on the bacterial fermentation was attributed to their different amino acid

distribution.

Keywords Casein hydrolysates, enzyme activity, lactic acid biosynthesis, Lactobacillus delbrueckii subsp. bulgaricus, Streptococcus thermo-

philus, ultrafiltered fractions.

Introduction

Lactic acid (LA), one of the earliest known fermenta-tion products of microbial metabolism, is widely usedas natural preservative, acidulant and flavour agent infood, pharmaceutical and chemical industries (Michel-son et al., 2006; Silveira et al., 2010). Many microor-ganisms have been used for LA production, whereaslactic acid bacteria (LAB) are the most frequently used(Plessas et al., 2008). LAB are used worldwide in theindustrial manufacture of fermented food products,and their most important application in this respect isundoubtedly in the dairy industry (Kleerebezem et al.,2000). Streptococcus thermophilus (St) and Lactobacil-lus delbrueckii subsp. bulgaricus (Lb) are the mostcommonly used yoghurt starter cultures (Ginovartet al., 2002). The main role of St and Lb in yoghurtmanufacture is to acidify milk by producing a largeamount of LA from lactose (Fu & Mathews, 1999).Furthermore, LA can enhance shelf life and contributethe aroma and flavour of yoghurt (Leroy & De Vuyst,

2004). The enhancement of LA production of yogurtstarter could reduce yoghurt fermentation time anddecrease the contamination involving in yoghurt man-ufacture. St and Lb are both homofermentative LAB,possessing a relatively simple metabolism completelyfocused on the rapid conversion of sugar to LA(Hugenholtz & Kleerebezem, 1999; Leroy & De Vuyst,2004).Lactic acid bacteria (LAB) require a wide range of

growth factors including amino acids, peptides, vita-mins and fatty acids for their growth and biologicalactivity because of their fastidious nature (Kwon et al.,2000). A number of studies about the nutrients neces-sary for LA fermentation have been reported (Kwonet al., 2000; Mussatto et al., 2008; Plessas et al., 2008;Xu et al., 2008; Campos et al., 2009). Casein containsall the amino acids necessary for LAB growth and alsohas many oligopeptides with stimulative effect (Kunjiet al., 1996). It has been shown that the addition ofcasein hydrolysates can reduce the fermentation timeof yoghurt and improve the viability of LAB inyoghurt (Sodini et al., 2005; Zhang et al., 2010). Poch& Bezkorovainy (1991) also found that tryptic hydro-lysates of kappa-casein were the most potent growth

*Correspondent: Fax: +86 20 87113914;

e-mail: [email protected]

International Journal of Food Science and Technology 2013 48, 1474–1482

doi:10.1111/ijfs.12115

© 2013 The Authors. International Journal of Food Science and Technology © 2013 Institute of Food Science and Technology

1474

enhancer for Bifidobacterium. However, very little isknown about the mechanism of LA biosynthesisaffected by nitrogen source during long-term fermenta-tion by LAB. Several features of LAB, like their rela-tively simple metabolism, limited biosynthetic capacityand apparent lack of gene multiplicity, lead to advan-tages for illustrating mechanism of LA productioncompared with other microorganisms like yeast orfungi (Kleerebezem et al., 2000).

Therefore, the objective of this study was to investi-gate the influence of five ultrafiltered fractions (UFs)with molecular mass less than 3 kDa (kilo-daltons)from casein on growth performance, LA productionand enzyme activity involved in glycolysis pathwayduring 72-h fermentation by mixed cultures of Lb andSt. Furthermore, the mechanism of LA productionaffected, and the difference in the effect among fivefractions was clarified.

Materials and methods

Materials and chemicals

Sodium caseinate was purchased from Chr. Hansen(Guangzhou, China). Five proteolytic enzymes, that is,alcalase, flavourzyme, neutrase, papain and trypsin,with enzymatic activity of 6.9 9 105 U mL�1,4.9 9 105 U g�1, 2.7 9 105 U g�1, 5.1 9 105 U g�1

and 5.5 9 105 U g�1, respectively, were obtained fromNovozymes Biotechnology (Tianjin, China). All otherchemicals and solvents used in this study were ofanalytical grade and obtained from Sigma-Aldrich(St. Louis, MO, USA).

Preparation and ultrafiltration of casein hydrolysatesSodium caseinate was dissolved in deionised water at aconcentration of 100 mg mL�1. The suspension wasadjusted to an appropriate enzyme-to-substrate ratio(1:100). The hydrolysis temperature for alcalase, fla-vourzyme, neutrase, papain and trypsin was 45, 55, 45,50 and 55 °C, respectively. The reaction was per-formed for 8 h with papain, and other reactions wereperformed for 12 h. At the end of the hydrolysis per-iod, the reactions were terminated by water bath at95 °C for 10 min with stirring to ensure the inactiva-tion of the enzyme. The resultant slurry was cooled inan ice bath and then centrifuged at 6000 9 g for20 min at 4 °C (Centrifuge, Sigma, Munich, Germany)to eliminate the sediment. Then, all the supernatantswere subjected to ultrafiltration using a bioreactor sys-tem (Sartorius, Goettingen, Germany) equipped withan ultrafiltration membrane of 3-kDa cut-off. Finally,five UFs with molecular mass less than 3 kDa fromthe hydrolysates by alcalase, flavourzyme, neutrase,papain and trypsin, that is, UF-A, UF-F, UF-N, UF-P

and UF-T, were obtained. The protein contents ofUFs were determined by the Kjeldahl method.

Microorganisms and growth conditions

YC-380 (St and Lb in 1:1 ratio), a freeze-dried commer-cial starter culture (from Chr. Hansen), was usedthroughout this work. The stock culture was fermentedin the seed cultivation broth (14% skim milk powderculture, w w�1) for 5 h till coagulation. Then, fermenta-tion broth was prepared by inoculating De Man,Rogosa, Sharpe Agar (MRS) broth medium with 5% ofseed culture. The anaerobic fermentations were per-formed at 37 � 0.1 °C for 72 h (Anaerobic incubator;Shanghai Fuma Test Equipment, Shanghai, China). Toevaluate the influence of five UFs on the bacterialfermentation, UF-A, UF-F, UF-N, UF-P and UF-Twere supplemented to MRS broth medium at a proteinconcentration of 0.5% (w w�1), respectively. The med-ium without UFs supplement was used as the control.

Sampling

Samples were aseptically withdrawn before inocula-tion, immediately after inoculation (0 h) and at differ-ent time intervals (generally every 12 h). Samples wereimmediately analysed for viable cell counts. Theremainder of the fermentation medium was stored at�18 °C for other analysis.

Analytical methods

Bacterial enumerations were carried out using themethod of Oliveira et al. (2001). One millilitre of sam-ple was diluted with 9 mL of 0.1% sterile peptonewater. Afterwards, serial dilutions were done and thebacteria were counted applying the pour tube tech-nique. St colonies were enumerated in Elliker agar byincubating the tubes anaerobically at 37 °C for 48 h.Lb enumeration was carried out in MRS agar mediumafter pH adjustment at 5.4 by acetic acid addition andanaerobic incubation at 37 °C for 72 h. The bacterialcounts were expressed as log cfu mL�1.Cell concentrations were obtained by measuring the

optical density at 622 nm in a spectrophotometerUnico 2100 (Shanghai, China). Samples were dilutedto a reading range of 0.05–1.0 units, and the ODvalues were correlated with the cell concentrations(g L�1) by means of a calibration curve. One unit ofoptical density at 620 nm corresponded to 25.0 g drycell weight per litre.All pH measurements were made using pH-3E pH

meter with a combined glass electrode and temperatureprobe (Rex, Shanghai, China). The pH meter was cali-brated using standard buffer solutions (Merck) at pH4.0 and 7.0.

© 2013 The Authors

International Journal of Food Science and Technology © 2013 Institute of Food Science and Technology

International Journal of Food Science and Technology 2013

Lactic acid production by LAB: influence Q. Zhang et al. 1475

Glucose and LA concentrations were determined bya high-performance liquid chromatography (HPLC)system (Waters, Milford, MA, USA) equipped with arefractive index and dual k absorbance detectors,respectively. The glucose concentration was analysedon a Spherisorb� NH2 (5 lm, 250 9 4.6 mm) column(Waters) eluted with a constant gradient of 75% aceto-nitrile at a flow rate of 0.8 mL min�1. LA quantifica-tion was performed on an XBridgeTM C18 (5 lm,250 9 4.6 mm) column (Waters). The mobile phasewas used with 10 mM NaH2PO4 (pH 2.0) at a flowrate of 0.5 mL min�1, and a UV detector was adjustedto 210 nm. Samples were microfiltered using 0.45-lmsyringe filters prior to injection. The injection volumeof the sample was 20 lL. Peak identification wasbased on the relative retention time determined byinjection of standard solution. Quantification wasperformed using a calibration curve.

Preparation of cell extracts and enzyme assays

Preparation of cell extracts. Cell extracts were preparedaccording to the method of Walling et al. (2005), andthe protein concentrations of the extracts were evalu-ated by Lowry method with bovine serum albumin asstandard. The formation or consumption of NAD(P)Hwas determined by measuring the change in the absor-bance at 340 nm. The determinations were done in trip-licate. The blank contained the reaction buffer, thecofactors and the substrate but lacked the cell extract.

The glucokinase (GLK) activity was assayed with areaction mixture that contained 50 lM glycylglycine(pH 7.5), 5 lM MgCl2, 0.4 lM NADP, 0.01 mL glu-cose-6-phosphate dehydrogenase (18 U mL�1), 1 lMATP and 0.1 M glucose. The reaction was started bythe addition of cell extract (Velasco et al., 2007).

The phosphoglucose isomerase (PGI) reverse reactioncontained 50 lM potassium phosphate (pH 6.8), 5 lMMgCl2, 0.4 lM NADP, 0.01 mL glucose-6-phosphatedehydrogenase (180 U mL�1) and cell extract. The reac-tion was started by adding 2.5 lM fructose-6-phosphate(Velasco et al., 2007).

The reaction mixture for the 6-phosphofructokinase(PFK) assay contained 50 lM Tris–HCl (pH 7.5), 5 lMMgCl2, 50 lM KCl, 0.15 lM NADH, 1.25 lM ATP,50 lg aldolase, 20 lg triosephosphate isomerase/glyc-erol phosphate dehydrogenase and cell extract. Thereaction was started by adding 1 lM fructose-6-phos-phate (Velasco et al., 2007).

The reaction mixture for the pyruvate kinase (PYK)assay contained 100 lM Tris–HCl buffer (pH 7.6),1.5 lM EDTA, 100 lM KCl, 10 lM MgCl2, 0.2 lMNADH, 0.01 mL glucose-6-phosphate dehydrogenase(36 U mL�1), 3 lM ADP and 5 lM phosphoenolpyr-uvate. The reaction was started by the addition of cellextract (Susan-Resiga & Nowak, 2003).

The lactate dehydrogenase (LDH) reaction mixturecontained 50 lM Tris–HCl (pH 7.5), 5 lM MgCl2,0.5 lM NADH and 50 lM sodium pyruvate. The reac-tion was started by the addition of cell extract (Wallinget al., 2005).

Amino acid analysis

The total amino acid profiles of UF-A, UF-F, UF-N,UF-P and UF-T were assessed according to themethod of Dong et al. (2008). Amino acid composi-tions were determined by HPLC (Waters) equippedwith a PICO.TAG column (Waters). Total amino acidresidues were assessed after hydrolysis at 110 °C for24 h with 6 M hydrochloric acid prior to the derivati-sation with phenyl isothiocyanate. Alkaline hydrolysiswas also done for the determination of Trp level.

Statistical analysis

Analysis of variance and significant differencesbetween means were tested by one-way ANOVA usingSPSS software (version 13.0 for Windows, SPSS Inc.,Chicago, IL, USA). Statistical significance of Pearson’scorrelations values (r) was obtained with the SPSS13.0 package.

Results and discussion

Bacterial growth performance

To study the influence of five UFs (UF-A, UF-F,UF-N, UF-P and UF-T) from casein on growth per-formance of mixed cultures of St and Lb, cells weregrown in MRS broth medium with 20 g L�1 of glu-cose. The profiles of pH, cell biomass, bacterial viabil-ity and sugar consumption are shown in Fig. 1a–e.The results obtained indicated that all UFs displayedpositive effect on bacterial growth, significantly(P < 0.05) increasing the acidity rate and the maxi-mum biomass attained (Fig. 1a,b). On the other hand,the growth phases of the bacteria during fermentationprocess were independent of medium composition. Thestimulative effect of UF-P was significantly (P < 0.05)higher than other four UFs during exponential (12and 24 h) and stationary (36 and 48 h) growth phasesof mixed cultures. Besides, UF-P proportioned anincrease of 73.50% biomass production after 24 h ofincubation compared with the control.As shown in Fig. 1c,d, St counts were higher than

Lb at any given time period (except 0 h) in all fermen-tation courses, which was probably due to the fact thatSt is much more competitive in using the nutrientsthan Lb and Lb grows in chains (Rajagopal &Sandine, 1990). The results agreed with the findingspreviously published using the same strains

© 2013 The Authors

International Journal of Food Science and Technology © 2013 Institute of Food Science and Technology

International Journal of Food Science and Technology 2013

Lactic acid production by LAB: influence Q. Zhang et al.1476

(Radke-Mitchell & Sandine, 1986). In addition, Stcounts were significantly (P < 0.05) improved by theaddition of UFs in comparison with the control. Kunjiet al. (1996) also found that St grew better on themedia supplemented with the protein hydrolysatesbecause of its limited proteolytic activity. However, inthe case of Lb, no notable difference was found in bac-terial number between the control assay and the assayssupplemented with UFs at exponential stage (Fig. 1d).Microorganisms could use a wide variety of nitrogensources for growth but not all nitrogen sources sup-ported growth equally well (Sanchez & Demain, 2002;Safari et al., 2012). Oliveira et al. (2001) also foundthat the positive effect of casein hydrolysates on theviability of different LAB varied. Compared with thecontrol, a light increase in Lb counts occurred from 24to 36 h in each fermentation course supplemented withUF, which probably resulted from the protection ofmore numerous St in the presence of UFs. Several

studies on the positive interaction between St and Lbhave been reported: the growth of L. bulgaricus couldbe stimulated by formic acid and carbon dioxide pro-duced by S. thermophilus, and the specific peptides andamino acids produced by L. bulgaricus were stimula-tory for S. thermophilus (Driessen et al., 1982; Franc-ois et al., 2007). The lower bacterial count and pHvalue in the presence of UF-N were found in compari-son with UF-A (Fig. 1a,c) at 24 h. On the other hand,the highest bacterial count and lowest pH value wereobtained with UF-P among all the fermentationcourses at the same time. These results indicated thatthere was no good correlation between pH and bacte-rial viability (Zotta et al., 2009). From 36 to 72 h ofcultivation, St decreased to a greater extent than Lb inall fermentations, which was probably attributed tothe acid-sensitive nature of St relative to Lb (Radke-Mitchell & Sandine, 1986). At the decline stage, thedecrease in the counts of St and Lb was meliorated by

0 12 24 36 48 60 720.0

4

5

6

pH

Fermentation time (h)0 12 24 36 48 60 72

0

10

20

30

Dry

cel

l wei

ght (

g L–1

)

Fermentation time (h)

0 12 24 36 48 60 720.0

7

8

9

St c

ount

s (lo

g cf

u m

L–1)

Fermentation time (h)0 12 24 36 48 60 72

0.0

7

8

Lb c

ount

s (lo

g cf

u m

L–1)

Fermentation time (h)

0 12 24 36 48 60 72

0

5

10

15

20

Glu

ose

conc

entr

atio

n (g

L–1

)

Fermentation time (h)

(a) (b)

(c) (d)

(e)Figure 1 Fermentation profiles by mixed

cultures of Streptococcus thermophilus (St)

and Lactobacillus delbrueckii subsp. bulgari-

cus (Lb) with or without ultrafiltered frac-

tions (UFs) from casein, respectively, treated

with alcalase, flavourzyme, neutrase, papain

and trypsin. (a) pH, (b) dry cell weight,

(c) St viability, (d) Lb viability, (e) glucose

consumption. Symbols: (■) control, (◯)UF-A, (▼) UF-F, (h) UF-N, (*) UF-P,

( ) UF-T. Data are mean values of three

independent experiments.

© 2013 The Authors

International Journal of Food Science and Technology © 2013 Institute of Food Science and Technology

International Journal of Food Science and Technology 2013

Lactic acid production by LAB: influence Q. Zhang et al. 1477

the addition of UFs (Fig. 1c,d). Xu et al. (2008) alsofound the protection of vitamins for Lactobacillusparacasei NERCB 0401 from death at the end offermentation process.

As presented in Fig. 1e, the glucose utilisation bymixed cultures of St and Lb proceeded rapidly duringexponential phase in all cases and then the consump-tion rate slowed down, which suggested that the glu-cose metabolism of the bacteria started simultaneouslywith pH decrease and biomass accumulation. The ini-tial concentration of glucose in all MRS broth mediawas of 20 g L�1 and the increased glucose consump-tion was induced by the presence of UF-A, UF-F,UF-N, UF-P and UF-T. The sugar was not totallyconsumed, and concentrations of 2.0, 1.6, 2.9, 0.3 and0.3 g L�1, respectively, were found at the end of theincubation period. Glucose concentration of the con-trol at 72 h was 7.0 g L�1. Obviously, the glucose con-sumption was enhanced by the addition of UFs withinlimits.

Lactic acid production

The LA concentrations and the fermentative parame-ters obtained in the fermentation runs with or withoutUFs addition during 72 h of cultivation were summar-ised in Table 1. Compared with the control, higher LAformation was obtained with UFs supplement. LAconcentration was enhanced by 23.66%, 39.01%,29.74%, 49.64% and 47.40% with the supplement ofUF-A, UF-F, UF-N, UF-P and UF-T, respectively, atthe late exponential stage. In all cases, LA accumula-tion was initially fast, but slowed down after 24 h.The similar trend in glucose consumption was found

(Fig. 1e), which indicated that glucose metabolism ofthe mixed cultures started simultaneously with LAenrichment. Besides, compared with the control, theviability of St was obviously enhanced by the additionof UFs, but that of Lb stayed unchanged (Fig. 1c,d)during exponential phase. These results permitted toconclude that the increment in St was fundamental forthe increase in LA formation during that time.Regarding the LA production per gram of cell, the

YP/X (Mussatto et al., 2008) values clearly showed thatthe supplementation of UFs all led to a higher increasein total cell mass than LA productivity. Besides, com-pared with UF-N, a lower LA concentration andhigher biomass occurred in the presence of UF-A atthe early exponential stage (Fig. 1b and Table 1).These results indicated that only a certain part of thebiomass was always active, and obtaining high activecell concentration was essential for hyperproduction ofLA rather than just increasing the total cell mass(Richter & Nottelmann, 2004; Xu et al., 2008).Moreover, as LA accumulation in the fermentation

broth is unavoidable (as the major fermentation prod-uct), bacterial cultivation was carried out using initialglucose contents yielding LA concentrations underinhibitory levels (Y�a~nez et al., 2008). However, duringthe anaphase of fermentation process, the cell biomassdeclined more obviously in the medium without UFsaddition experiment than in the media with UFs addi-tion experiments. This was probably attributed to thefollowing facts. For most LAB, the inhibition by thisend product (LA) is only significantly in contentsabove 30–70 g L�1 (Amrane & Prigent, 1998). Andproteolysis could lower the negative effect of LA bythe production of NH3 by LAB. This NH3 could

Table 1 Fermentation parameters of lactic acid (LA) production by mixed cultures of Streptococcus thermophilus and Lactobacillus delbrueckii

subsp. bulgaricus in different fermentation media with or without ultrafiltered fractions (UFs) from casein respectively treated with alcalase, fla-

vourzyme, neutrase, papain and trypsin during 72 h. Data are mean values of three independent experiments

Parameters Fermentation time (h)

Fermentation media

Control UF-A UF-F UF-N UF-P UF-T

LA(g L�1) 12 9.15 � 0.49 9.85 � 0.22 10.9 � 0.34* 10.5 � 0.60 11.8 � 0.75* 11.30 � 0.31*

24 12.51 � 0.42 15.47 � 0.82* 17.39 � 0.33** 16.23 � 0.39** 18.72 � 0.71** 18.44 � 0.54**

36 12.61 � 0.75 15.75 � 0.41* 17.61 � 0.52** 16.40 � 0.89* 19.00 � 0.66** 18.80 � 0.48**

48 12.66 � 0.52 15.90 � 0.33* 17.72 � 0.35** 16.48 � 0.68* 19.08 � 0.86** 18.93 � 0.80**

60 12.69 � 0.72 15.94 � 0.70* 17.87 � 0.49** 16.50 � 0.38* 19.15 � 0.54** 18.97 � 0.47**

72 12.70 � 0.39 15.93 � 0.46* 17.85 � 0.72** 16.47 � 0.60* 19.14 � 0.52** 18.95 � 0.41**

YP/X (g g�1) 12 0.92 � 0.03 0.62 � 0.03* 0.68 � 0.01* 0.70 � 0.03* 0.67 � 0.02* 0.66 � 0.02*

24 0.88 � 0.02 0.69 � 0.05 0.73 � 0.05 0.71 � 0.02 0.76 � 0.04 0.75 � 0.04

36 0.85 � 0.01 0.67 � 0.01 0.72 � 0.03 0.68 � 0.05 0.75 � 0.03 0.76 � 0.05

48 0.84 � 0.04 0.67 � 0.01 0.71 � 0.02 0.68 � 0.01 0.71 � 0.02 0.72 � 0.05

60 1.00 � 0.04 0.69 � 0.05* 0.75 � 0.04* 0.70 � 0.02* 0.74 � 0.01* 0.75 � 0.02*

72 1.27 � 0.02 0.70 � 0.04* 0.75 � 0.02* 0.70 � 0.03* 0.75 � 0.01* 0.75 � 0.01*

*Values significantly different from control values (P < 0.05).

**Values significantly different from control values (P < 0.01).

© 2013 The Authors

International Journal of Food Science and Technology © 2013 Institute of Food Science and Technology

International Journal of Food Science and Technology 2013

Lactic acid production by LAB: influence Q. Zhang et al.1478

come from the deamination of amino acids. Further-more, the positive effect of UFs might counteract thekind of negative action to a certain extent.

Enzymatic activity analysis

The above results obtained in this study confirmed thestimulative effect of five UFs (UF-A, UF-F, UF-N,UF-P and UF-T) on LA production from mixed cul-tures of St and Lb. However, the mechanism of theaction of UFs on LA biosynthesis was still unclear.Hence, we carried out an investigation into the activityof several key enzymes involved in LA biosyntheticpathway, which could be helpful to clarify the reasonsof stimulating LA production by UFs supplement.

Lactic acid biosynthesis involves the enzymes ofglycolysis pathway, and GLK, PGI, PFK, PYK andLDH are the key enzymes of this pathway (Webster,2003). The metabolic map of the bacteria is presentedin Fig. 2. Hence, the activity of these five enzymes wasdetermined in detail during fermentation with or with-out UF supplement.

The GLK activity profile during the fermentationprocess is presented in Fig. 3a. This enzyme wasinvolved in the phosphorylation of glucose, whichcould be found intracellularly as a result of the hydro-lysis of maltose or after being transported by a perme-ase system (Velasco et al., 2007). Mixed cultures of Stand Lb exhibited higher GLK activity in the presenceof UFs in comparison with the control. Under allmedia conditions, the highest activity levels weredetected at 48 h and the highest GLK activity

(0.94 lM mg of protein�1 min�1) was obtained in thepresence of UF-P at that time.As shown in Fig. 3b, c, PGI and PFK, the two

enzymes responsible for the deviation of the carbonflux towards glycolysis (Velasco et al., 2007), displayedhigher activity levels compared with GLK. Measurableactivity levels of PGI were found irrespective of themedium composition. The PGI activity in all mediawas rather low in the earlier phase of fermentation,whereas it was maintained at a higher level duringmiddle and later phases. Obviously, compared with thecontrol, each UF supplement could enhance PFKactivity during whole fermentation process, showingstatistically different significance (P < 0.05). Further-more, a high correlation between LA production andPFK was found for UF-A, UF-F, UF-N, UF-P andUF-T (r = 0.95, 0.90, 0.87, 0.97 and 0.82, respectively),which was statistically significant (P < 0.05, 0.01, 0.05,0.01 and 0.01, respectively).Figure 3d shows the PYK activity profile during the

fermentation process. Like PFK, the PYK specificactivity was dependent on the nitrogen source. Fur-thermore, a good correlation was found between theactivity of PFK and PYK (r = 0.90, P < 0.01). As aresult, metabolic flux of glycolysis pathway wasenhanced at exponential phase, which was essential forhyperproduction of LA. Sheng & Wang (2002) previ-ously reported that vitamins could enhance the meta-bolic flux of glycolysis pathway through promotingglycolysis rate and accelerating electron transferring.As shown in Fig. 3e, LDH, possessing the function

to convert pyruvic acid to LA as the main end product(Kleerebezem et al., 2000), displayed a progressivedecrease along the incubation period (except at 36 h)in all fermentation courses. The activity levels of LDHwere very high compared with the above mentionedenzymes. Five UFs showed no influence on the LDHactivity during whole fermentation process, which indi-cated that the metabolic flux from pyruvic acid to LA wasnot improved by UFs supplement. Furthermore, the fer-mentation course with UF-A displayed the lowest LDHactivity among all the cases since 36 h. The LDH reactionacted as main electron sink under anaerobic conditions,resulting in homolactic fermentation (Kleerebezem et al.,2000). Nevertheless, a shift towards other metabolites fer-mentation had been observed under unusual conditions,which could induce low LDH activity (Hugenholtz &Kleerebezem, 1999). Cretenet et al. (2011) found that amoderate acid pH led to the over expression of manyenzymes as a stress response. In general, the LA biosynthesisis a complicated process.

Amino acid distribution of ultrafiltered fractions

The effect of five UFs on LA production of mixedcultures of St and Lb varied as above mentioned, so,

Glucose

ldh

Lactic acid

Pyruvate

pyk

pfk

Fructose-16dP

pgi

Fructose-6P

Glucose-6P

Glucoseglk

Figure 2 The biosynthetic pathway of lactic acid by homofermenta-

tive lactic acid bacteria during glucose metabolism.

© 2013 The Authors

International Journal of Food Science and Technology © 2013 Institute of Food Science and Technology

International Journal of Food Science and Technology 2013

Lactic acid production by LAB: influence Q. Zhang et al. 1479

the amino acid distribution of different UF was deter-mined (Fig. 4). Compared with UF-A, UF-F or UF-N, higher concentrations of Glu, His, Lys and Ser,which are all hydrophilic residues, were obtained inUF-P and UF-T, which probably led to the higher Stcounts at exponential stage in the presence of UF-Pand UF-T (Fig. 1c). Zourari et al. (1992) also reportedthat Glu and His were primarily required by St forgrowth. The contents Cys and Met, which both belongto sulphydryl residues, were the same level. And, theCys contents were very low among these five fractions.These results were conflicted with the previous findingthat sulphydryl groups were essential for stimulativeeffect for Bifidobacteria (Poch & Bezkorovainy, 1991).The highest levels of aromatic amino acid residuesincluding Phe, Trp and Tyr possessing inhibition activ-ity against St (Radke-Mitchell & Sandine, 1986) wereobserver in UF-N, which could explain the lowest via-bility of St with UF-N among all media with UFs

supplement at 12 and 24 h. Zhang et al. (2007) alsofound that the type and quantity of the nitrogensource play an important role in synthesising lacticacid and by-products in the fermentation by Rhizopusarrhizus 36017. In summary, peptide fractions frommilk with low molecular weight are important nitrogensource for LAB (Zourari et al., 1992).

Conclusion

LA production and bacterial growth of mixed culturesof St and Lb during 72 h of cultivation were stronglyaffected by five UFs (< 3 kDa) from casein (UF-A,UF-F, UF-N, UF-P and UF-T), especially UF-Pand UF-T. During LA production of the bacteriaenhanced by UFs, the metabolic flux of glycolysispathway was enhanced, through enlarging the activityof GLK, PFK and PYK. The difference in stimulativeeffect of five UFs for the bacteria was attributed to

12 24 36 48 60 72

0.0

0.5

1.0

1.5

GLK

(µM

mg

of p

rote

in–1

min

–1)

Fermentation time (h)

0.03

4

5

6

7

Fermentation time (h)

0.02

3

4

5

6

Fermentation time (h)

0

1

2

3

4

Fermentation time (h)

0.0

4

6

8

10

Fermentation time (h)

12 24 36 48 60 72

12 24 36 48 60 72 12 24 36 48 60 72

12 24 36 48 60 72

PGI (

µM m

g of

pro

tein

–1 m

in–1

)

PFK

(µM

mg

of p

rote

in–1

min

–1)

PYK

(µM

mg

of p

rote

in–1

min

–1)

LDH

(µM

mg

of p

rote

in–1

min

–1)

(a) (b)

(c) (d)

(e)

Figure 3 Activity of enzymes involved in

glycolysis pathway leading to the biosynthe-

sis of lactic acid in cell extracts of mixed cul-

tures of Streptococcus thermophilus and

Lactobacillus delbrueckii subsp. bulgaricus

grown in MRS broth with or without ultra-

filtered fractions (UFs) from casein, respec-

tively, treated with alcalase, flavourzyme,

neutrase, papain and trypsin. (a) glucokinase

(GLK), (b) phosphoglucose isomerase

(PGI), (c) 6-phosphofructokinase (PFK), (d)

pyruvate kinase (PYK), (e)lactate dehydro-

genase (LDH). Symbols: (■) control, (◯)UF-A, (▼) UF-F, (h) UF-N, (*) UF-P,

(◄) UF-T. Data are mean values of three

independent experiments.

© 2013 The Authors

International Journal of Food Science and Technology © 2013 Institute of Food Science and Technology

International Journal of Food Science and Technology 2013

Lactic acid production by LAB: influence Q. Zhang et al.1480

their diverse amino acid distribution. The resultsobtained from this research also could contribute theselection of casein hydrolysates addition to enhanceyoghurt fermentation. On the basis of observationobtained from current study, further work on yoghurtmanufacture with the UFs addition assay would beperformed.

Acknowledgments

The authors gratefully acknowledge the financial sup-ports from the Science and Technology Program ofGuangdong Province (No. 2008A010900001 and No.2008A010900017), China.

References

Amrane, A. & Prigent, Y. (1998). Influence of an initial addition oflactic acid on growth, acid production and their coupling forbatch cultures of Lactobacillus helveticus. Bioprocess Engineering,19, 307–312.

Campos, F.M., Figueiredo, A.R., Hogg, T.A. & Couto, J.A. (2009).Effect of phenolic acids on glucose and organic acid metabolism bylactic acid bacteria from wine. Food Microbiology, 26, 409–414.

Cretenet, M., Laroute, V., Ulv�e, V. et al. (2011). Dynamic analysisof the Lactococcus lactis transcriptome in cheeses made from milkconcentrated by ultrafiltration reveals multiple strategies of adapta-tion to stresses. Applied Environmental Microbiology, 77, 247–257.

Dong, S.Y., Zeng, M.Y., Wang, D.F., Liu, Z.Y., Zhao, Y.H. &Yang, H.C. (2008). Antioxidant and biochemical properties of pro-tein hydrolysates prepared from Silver carp (Hypophthalmichthysmolitrix). Food Chemistry, 107, 1485–1493.

Driessen, F.M., Kingma, F. & Stadhouders, J. (1982). Evidence thatLactobacillus bulgaricus in yogurt is stimulated by carbon dioxide

produced by Streptococcus thermophilus. Netherlands Milk andDairy Journal, 36, 135–139.

Francois, Z.N., Hoda, N.E., Florence, F.A., Paul, M.F., F�elicit�e,T.M. & Soda, M.E. (2007). Biochemical properties of some ther-mophilic lactic acid bacteria strains from traditional fermentedmilk relevant to their technological performance as starter culture.Biotechnology, 6, 14–21.

Fu, W. & Mathews, A.P. (1999). Lactic acid production from lactoseby Lactobacillus plantarum: kinetic model and effects of pH, sub-strate, and oxygen. Biochemical Engineering Journal, 3, 163–170.

Ginovart, M., L�opez, D., Valls, J. & Silbert, M. (2002). Simulationmodelling of bacterial growth in yoghurt. International Journal ofFood Microbiology, 73, 415–425.

Hugenholtz, J. & Kleerebezem, M. (1999). Metabolic engineering oflactic acid bacteria: overview of the approaches and results ofpathway rerouting involved in food fermentations. Current Opinionin Biotechnology, 10, 492–497.

Kleerebezem, M., Hols, P. & Hugenholtz, J. (2000). Lactic acid bac-teria as a cell factory: rerouting of carbon metabolism in Lactococ-cus lactis by metabolic engineering. Enzyme and MicrobialTechnology, 26, 840–848.

Kunji, E.R.S., Mierau, I., Hagting, A., Poolman, B. & Konings,W.N. (1996). The proteolytic systems of lactic acid bacteria. Anto-nie van Leeuwenhoek, 70, 187–221.

Kwon, S., Lee, P.C., Lee, E.G., Chang, Y.K. & Nam Chang, N.(2000). Production of lactic acid by Lactobacillus rhamnosus withvitamin-supplemented soybean hydrolysate. Enzyme and MicrobialTechnology, 26, 209–215.

Leroy, F. & De Vuyst, L. (2004). Lactic acid bacteria as functionalstarter cultures for the food fermentation and industry. Trends inFood Science and Technology, 15, 67–78.

Michelson, T., Kask, K., J~ogi, E., Talpsep, E., Suitso, I. & Nurk, A.(2006). L(+)-Lactic acid producer Bacillus coagulans SIM-7 DSM14043 and its comparison with Lactobacillus delbrueckii ssp. lactisDSM 20073. Enzyme and Microbial Technology, 39, 861–867.

Mussatto, S.I., Fernandes, M., Mancilha, I.M. & Roberto, M.I.(2008). Effects of medium supplementation and pH control on lac-tic acid production from brewer’s spent grain. Biochemical Engi-neering Journal, 40, 437–444.

Oliveira, M.N., Sodini, I., Remeuf, F. & Corrieu, G. (2001). Effect ofmilk supplementation and culture composition on acidification, tex-tural properties and microbiological stability of fermented milks con-taining probiotic bacteria. International Dairy Journal, 11, 935–942.

Plessas, S., Bosnea, L., Psarianos, C., Koutinas, A.A., Marchant,R. & Banat, I.M. (2008). Lactic acid production by mixed cul-tures of Kluyveromyces marxianus, Lactobacillus delbrueckii ssp.Bulgaricus and Lactobacillus helveticus. Bioresource Technology,99, 5951–5955.

Poch, M. & Bezkorovainy, A. (1991). Bovine Milk j-Casein trypsindigest is a growth enhancer for the genus Bifidobacterium. Journalof Agricultural and Food Chemistry, 39, 73–77.

Radke-Mitchell, L.C. & Sandine, W.E. (1986). Influence of tempera-ture on associative growth of Streptococcus thermophilus and Lac-tobacillus bulgaricus. Journal of Dairy Science, 69, 2558–2568.

Rajagopal, S.N. & Sandine, W.E. (1990). Associative growth andproteolysis of Streptococcus thermophilus and Lactobacillus bulgari-cus in skim milk. Journal of Dairy Science, 73, 894–899.

Richter, K. & Nottelmann, S. (2004). An empiric steady state modelof lactate production in continuous fermentation with total cellretention. Engineering in Life Science, 4, 426–432.

Safari, R., Motamedzadegan, A., Ovissipour, M., Regenstein, J.M.,Gildberg, A. & Rasco, B. (2012). Use of hydrolysates from Yellow-fin Tuna (Thunnus albacares) heads as a complex nitrogen sourcefor lactic acid bacteria. Food and Bioprocess Technology, 5, 73–79.

Sanchez, S. & Demain, A.L. (2002). Metabolic regulation of fermen-tation processes. Enzyme and Microbial Technology, 31, 895–906.

Sheng, T. & Wang, J.Y. (2002). Biochemistry, 3rd edn. Beijing:Higher Education Press.

0

20

40

60

80

100

ValTyrTrpThrSerProPheM

etLysLeuIleH

isG

lyG

luC

ysA

snA

rg

Con

cent

ratio

n (m

g 10

0 g–

1 )

Fermentation time (h)

Ala

Figure 4 The amino acid distribution of five ultrafiltered fractions

(UFs) from casein treated with alcalase, flavourzyme, neutrase,

papain and trypsin, respectively. Symbols: ( ) UF-A, ( ) UF-F,

( ) UF-N, ( ) UF-P, ( ) UF-T. Data are mean values of three

independent experiments.

© 2013 The Authors

International Journal of Food Science and Technology © 2013 Institute of Food Science and Technology

International Journal of Food Science and Technology 2013

Lactic acid production by LAB: influence Q. Zhang et al. 1481

Silveira, M.S., Fontes, C.P.M.L., Guilherme, A.A., Fernandes,F.A.N. & Rodrigues, S. (2010). Cashew apple juice as substrate forlactic acid production. Food and Bioprocess Technology, 3, 1–7.

Sodini, I., Lucas, A., Tissier, J.P. & Corrieu, G. (2005). Physicalproperties and microstructure of yoghurts supplemented with milkprotein hydrolysates. International Dairy Journal, 15, 29–35.

Susan-Resiga, D. & Nowak, T. (2003). The proton transfer step cat-alyzed by yeast pyruvate kinase. The Journal of Biological Chemis-try, 278, 12660–12671.

Velasco, S.E., Yebra, M.J., Monedero, V., Ibarburu, I., Due~nas,M.T. & Irastorza, A. (2007). Influence of the carbohydrate sourceon b-glucan production and enzyme activities involved in sugarmetabolism in Pediococcus parvulus 2.6. International Journal ofFood Microbiology, 115, 325–334.

Walling, E., Dols-Lafargue, M. & Lonvaud-Funel, A. (2005). Glucosefermentation kinetics and exopolysaccharide production by ropyPediococcus damnosus IOEB8801. Food Microbiology, 22, 71–78.

Webster, K.A. (2003). Evolution of the coordinate regulation of gly-colytic enzyme genes by hypoxia. The Journal of Experimental Biol-ogy, 206, 2911–2922.

Xu, G.Q., Chu, J., Zhuang, Y.P., Wang, Y.H. & Zhang, S.L.(2008). Effects of vitamins on the lactic acid biosynthesis of Lacto-

bacillus paracasei NERCB 0401. Biochemical Engineering Journal,38, 189–197.

Y�a~nez, R., Marques, S., G�ırio, F.M. & Roseiro, J.C. (2008). Theeffect of acid stress on lactate production and growth kinetics inLactobacillus rhamnosus cultures. Process Biochemistry, 43,356–361.

Zhang, Z.Y., Jin, B. & Kelly, J.M. (2007). Production of lactic acidand byproducts from waste potato starch by Rhizopus arrhizus:role of nitrogen sources. World Journal of Microbiology & Biotech-nology, 23, 229–236.

Zhang, Q.L., Zhao, M.M., Qu, D.J., Zhao, H.F. & Zhao, Q.Z.(2010). Effect of papain-hydrolysed casein peptides on the fermen-tation kinetics, microbiological survival and physicochemical prop-erties of yoghurt. International Journal of Food Science &Technology, 45, 2379–2386.

Zotta, T., Asterinou, K., Rossano, R., Ricciardi, A., Varcamonti,M. & Parente, E. (2009). Effect of inactivation of stress responseregulators on the growth and survival of Streptococcus thermophi-lus Sfi39. International Journal of Food Microbiology, 129, 211–220.

Zourari, A., Accolas, J.P. & Desmazeaud, M.J. (1992). Metabolismand biochemical characteristics of yogurt bacteria. A review. Lait,72, 1–34.

© 2013 The Authors

International Journal of Food Science and Technology © 2013 Institute of Food Science and Technology

International Journal of Food Science and Technology 2013

Lactic acid production by LAB: influence Q. Zhang et al.1482

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具