Isolation of lactic acid bacteria from Malaysian foods and assessment

of the isolates for industrial potential

¤
Ahmad Faris Mohd Adnan, Irene K.P. Tan
Institute of Biological Sciences, University of Malaya, 50603 Kuala Lumpur, Malaysia

Abstract

Two traditional fermented food ‘tapai’ (fermented tapioca) and ‘tempoyak’ (fermented durian Xesh), chilli puree and fresh goat’s milk
were used as sources for the isolation of lactic acid bacteria (LAB). A total of 126 isolates were obtained and by sequential screening for
catalase activity and Gram-staining, 55 were determined to be LAB out of which 16 were established to be homofermentative by the gel
plug test. Seven isolates were identiWed by use of the API 50CHL kit and two lactobacilli strains and one lactococci strain were selected to
study their growth and lactic acid production proWles in a time course experiment. The lactobacilli strains, both isolated from ‘tapai’, pro-
duced higher amounts of cells and lactic acid from glucose as compared to the lactococci strain isolated from fresh goat’s milk.
 2006 Elsevier Ltd. All rights reserved.

Keywords: Chilli puree; Fermented foods; Goat’s milk; Homofermenters; Lactic acid bacteria; Tapai; Tempoyak

1. Introduction thesis, e.g., as a by-product in the petrochemical industries,

The isolation and screening of microorganisms from nat-
ural sources has always been the most powerful means for
obtaining useful and genetically-stable strains for industri-
ally-important products. Lactic acid bacteria (LAB) are
important in the food and dairy industries because the lactic
acid and other organic acids produced by these bacteria act
as natural preservatives as well as Xavour enhancers. LAB
Wnd increasing acceptance as probiotics which aid in stimu-
lating immune responses, preventing infection by entero-
pathogenic bacteria, and treating and preventing diarrhoea
(Reid, 1999). In non-food applications, lactic acid is the
building blocks used in the chemical polymerisation of poly
(lactic acid) (PLA), a biodegradable and biocompatible
material which is a good alternative to synthetic plastics.
PLA have found use as surgical sutures, orthopaedic
implants, in drug delivery systems and disposable consumer
products. Lactic acid could be produced by chemical syn-
*
Corresponding author. Tel.: +60 3 79676792; fax: +60 3 79674178.
E-mail address: itan@um.edu.my (I.K.P. Tan).
or it could be made by microbial fermentation. The fermen-
tation process is becoming more relevant because the raw
materials used in fermentation are renewable in contrast
to petrochemicals. Furthermore, the fermentation process
could produce optically pure isomers of lactic acid. Pure
isomers, L or D lactic acid, are more valuable than the race-
mic DL form because each isomer has its own applications in
the cosmetics and pharmaceuticals industries. In PLA, the
ratio of L and D lactic acid inXuences the degradability of
the polymers (Kharras et al., 1993), therefore it is easier to
manufacture PLA with speciWc properties, e.g., degradabil-
ity, if L and D lactic acid are supplied separately. LAB that
have industrial potential should be homofermentative, i.e.,
they produce only lactic acid. Heterofermentative LAB pro-
duce carbon dioxide and other types of organic acid besides
lactic acid, and are therefore deemed to be lower-yielding
strains. Recovery of lactic acid and subsequent puriWcation
would also be easier in a homofermentative process. Other
desirable properties for an industrially-useful LAB are the
synthesis of pure isomers and tolerance to high product
concentration and high temperature (42 °C or higher) as
these external conditions themselves would be useful to
minimise contamination of the culture by other micro-
organisms (Tsai et al., 1993).
The objective of this study was to isolate lactic acid bac-
teria from some Malaysian food and to screen these isolates
for desirable traits such as homofermentative ability, pro-
duction of single isomeric form of lactic acid, and tolerance
to high concentration of lactic acid, low pH and high tem-
perature.
2. Methods
2.1. Isolation of LAB
Two local fermented food, ‘tapai’ (fermented tapioca) and
‘tempoyak’ (fermented durian Xesh), ‘chili boh’ (chilli puree)
and fresh goat’s milk were all bought from a local market.
Each of these (10 g) was separately blended with 100 ml
0.85% NaCl solution; 10 ml of this blended food was added
to 100 ml MRS broth (de Man et al., 1960) in a 250 ml Erlen-
meyer Xask. The broths containing tapai and tempoyak were
enriched with mannitol, sorbitol, lactose, sucrose or glucose,
at 2% w/v concentration. The chilli boh broth was enriched
with glucose (2%) only. The Xasks were shaken in an incuba-
tor (Hotech 718 Orbital shaker) at 100 rpm, 37 °C for 7 days.
The fresh goat’s milk was incubated on its own, i.e., without
mixing with MRS broth and with no sugar enrichment at
100 rpm, 45 °C for 7 days; then aliquots of the culture from
each of the Xasks were diluted serially to 106 times and 0.1 ml
was spread evenly on MRS agar plates. The plates were incu-
bated for 3 days at 37 °C. Bacterial colonies that developed
on the plates were individually picked and streaked on fresh
MRS agar plates by dilution-streaking to obtain single
colonies. This procedure was repeated in order to purify
the isolates, which were maintained on MRS agar slants
for immediate use and in 20% glycerol for storage at ¡20 °C.
Each of the isolates was Wrst tested for catalase by plac-
ing a drop of 3% hydrogen peroxide solution on the cells.
Immediate formation of bubbles indicated the presence of
catalase in the cells. Only those isolates which were cata-
lase-negative were Gram-stained, and only those which
were Gram-positive were put through the gel plug test
(Gibson and Ab-del Malek, 1945) to determine whether the
isolate produced carbon dioxide during fermentation. An
isolate was deemed to be a homofermentative lactic acid
producer if no gas was produced. Based on the results,
seven homofermentative isolates were selected for further
studies. These isolates were identiWed by the API 50CHL
identiWcation kit (Biomerieux).
2.2. Determination of lactic acid isomers
Each isolate was cultured in 100 ml basal MRS medium
contained in a 250 ml Erlenmeyer Xask at 100 rpm, 37 °C for
48 h. The cells were spun down and the supernatant was
passed through 0.22
m micropore Wlter (Millipore). The
recovery and puriWcation of lactic acid from the superna-
tant was based on Vaccari et al. (1993) with some modiW-
cations. The Wltered supernatant was passed through a
column packed with a strong anion exchange resin in car-
bonate form, Amberlite IRA-400 (Fluka Chemika), to
which lactic acid was adsorbed. Two volumes of 35 ml ultra-
pure water were passed through the column to remove
unadsorbed compounds. Next, ammonium carbonate solu-
tion was passed through and lactic acid was recovered in
the eluate as ammonium lactate. The eluate was then passed
through a strong cation exchange resin in hydrogen form,
Amberlite IR-120(H) (BDH Chemicals), to release the free
lactic acid in the eluate which was later concentrated by
vacuum evaporation under 1 atm pressure. The isomeric
form of the puriWed lactic acid was determined by thin layer
chromatography (Cecchi and Malaspina, 1991). Each of the
lactic acid samples and standards, D and L lactic acid (Fluka
Chemika), was dissolved in 500
l acetone and spotted on
the plate. The plate was placed in a glass development tank
containing 90% v/v dioxane (Fluka Chemika) in water as
the mobile phase and the chromatography was run for
40 min. The D and L lactic acid isomers appeared as dark
blue spots and diVerentiated by their respective Rf values.
2.3. Rapid screening for tolerance to high temperatures,
high concentrations of lactic acid and sodium chloride,
and low pH
A basal MRS medium was used in these series of studies
(de Man et al., 1960) but without beef extract, and with
0.17 g/l bromocresol purple added as pH indicator (pH 7).
A lowering of this pH would change the medium from pur-
ple to yellow, and was taken to indicate cell growth because
the production of lactic acid is growth-related. No change
in the color of the medium was taken to indicate no cell
growth because no acids were released. Universal bottles
with screw caps were each Wlled with 20 ml of the basal
MRS medium and autoclaved. An 18 h culture of each iso-
late was used as the inoculum whereby the cells were spun
down, resuspended in 0.85% saline, and 50
l of the suspen-
sion was inoculated into each test bottle. The temperatures
tested were 15, 30, 37, 45 and 50 °C, the concentrations of
lactic acid tested were 2.5%, 5%, 7.5%, 10% and 15% (w/v),
and the concentrations of NaCl tested were 1.5%, 2.5%, 5%,
7.5% and 10% (w/v). Three pH were tested, i.e., 4.5, 7 and 9.
The basal MRS medium was adjusted with 1 M phosphoric
acid and 1 M NaOH to prepare these initial pH. The bottles
were placed in water baths with reciprocal shaking, set at
the speciWc test temperatures or at 37 °C for the tests on pH
and concentrations of lactic acid and NaCl. At the end of
48 h, the color change and turbidity of each bottle was
noted as a simple indication of growth or no-growth. Each
treatment was tested with triplicate bottles.
2.4. Growth and lactic acid production proWles from time
course studies
Based on the screening studies for tolerance to various
environmental conditions, two lactobacilli isolates,
TapLac and TapSuc, and one lactococci isolate, FM, were
selected for a time-course study to generate their growth
and lactic acid production proWles. Erlenmeyer Xasks
(1000 ml) each containing 400 ml basal MRS medium, pH
7, were used. A 50 ml, 18 h culture was used as inoculum
whereby the cells were spun down and washed twice with
0.85% saline before being transferred to the test Xasks
which were run in triplicate for each isolate tested. The
Xasks were incubated at 37 °C at 100 rpm (LH Fermenta-
tion). The cultures were sampled at six hourly intervals
until the 54th hour. At each sampling time, 20 ml of the
cultures were removed aseptically from each Xask and
assayed for biomass (cell dry weight), pH, lactic acid and
residual glucose.
2.5. Assay of lactic acid
A 1 ml aliquot of the sample cultures was centrifuged
(Sigma Laborzentrifugen, Model 2K15) at 14 000g for
10 min at 4 °C after which the supernatant was diluted ten-
fold with ultra pure water. The lactic acid in the diluted
supernatant was quantiWed by high performance liquid
chromatography (HPLC) which comprised a pump
(Waters 501), absorbance detector (Waters, Model 486)
set at 214 nm, system interface module and autosampler
(Waters, Model 717 Plus) all operating on a Waters Mille-
nium32 software. The column used was an ion exchange
chromatographic column, Resolve 5
m spherical C18
3.9 £ 150 mm (Waters) maintained at room temperature.
Analysis was performed isocratically at a Xow rate of
0.5 ml/min. The mobile phase used was 0.5% diammonium
hydrogen phosphate (BDH) with pH adjusted to pH 3 with
60% phosphoric acid (Ajax). L-lactic acid (Fluka) in con-
centrations ranging from 0.03% to 0.5% w/v, each contain-
ing 0.25% v/v propionic acid (Sigma) as internal standard,
were assayed by the HPLC to produce the standard curve
(lactic acid concentration versus peak area).
2.6. Assay of residual glucose
The diluted supernatant as above was also used for the
assay of residual glucose. An enzymatic kit, Glucose (HK)
procedure no. 16-UV (Sigma Diagnostics), was used for the
assay.
Table 1
Some characteristics of the seven homofermentative LAB isolates
Isolate Food source from Sugar used in the
name which isolated enrichment medium
FM Fresh goat’s milk None
TapLac Tapai Lactose
TapSuc Tapai Sucrose
TemSor Tempoyak Sorbitol
TemGlu Tempoyak Glucose
TemMan Tempoyak Mannitol
CB Chili Boh Glucose
3. Results and discussion
3.1. Screening for lactic acid bacteria
One hundred twenty six bacterial isolates were obtained
from the four types of food and diVerent sugars used in the
enrichment medium. Of these, 55 tested Gram-positive and
catalase-negative, and all but one was rod-shaped. The one
remaining LAB isolate was coccus-shaped. Sixteen of these
isolates did not produce gas in the gel plug test and were,
therefore, deemed to be homofermenters. Based on the food
from which the isolate was obtained and the sugar used in
the enrichment process, seven of these 16 were selected for
characterization and the tolerance tests. Table 1 lists these
seven isolates, giving details about the food origin, the sugar
used in the enrichment process, cell morphology, identiWca-
tion by the API 50CHL kit, probability of Wt to the closest
species, and the type of lactic acid isomer produced. The iso-
lates from tapai and tempoyak, regardless of the sugar used
in the enrichment medium, were identiWed as Lactobacillus
casei, while the isolate from chili boh was identiWed as Lac-
tobacillus plantarum. The isolate obtained from fresh goat’s
milk was identiWed as Lactococcus lactis. Isolates identiWed
as L. casei and L. lactis produced only the L-form of the lac-
tic acid while the isolate identiWed as L. plantarum produced
a mixture of D and L isomers of lactic acid. LAB strains that
produce a single isomeric form of lactic acid are more desir-
able as industrial strains compared to those strains that
produce a racemic mixture of lactic acid. This is because
the material properties of poly (lactic acid) are governed by
the ratio of L and D isomers in the polymer, so the polymeri-
zation process would be better controlled when the lactic
acid used as starting material is in single isomeric forms.
3.2. Rapid screening for tolerance to high temperatures,
high concentrations of lactic acid and sodium chloride,
and low pH
Table 2 shows the tolerance of each of the seven selected
isolates to the environmental conditions tested. Five strains
of the Lactobacillus sp. grew at 15–45 °C, while the one iso-
lated from tempoyak in a glucose-enriched medium, Tem-
Glu, could not grow at 15 °C. The Lactococcus sp., FM,
isolated from fresh goat’s milk grew at higher temperatures
Cell IdentiWcation by IdentiWcation Lactic acid isomer
morphology API 50CHL kit probabilities (%) produced
Cocci Lactococcus lactis 95.9 L
Rod Lactobacillus casei 99.3 L
Rod L. casei 99.6 L
Rod L. casei 95.9 L
Rod L. casei 95.9 L
Rod L. casei 94.5 L
Rod L. plantarum 99.9 DL
Table 2 (Liu et al., 1998). Some cells overcome this eVect by regulat-
Tolerance of the seven LAB isolates to ranges of temperatures, lactic acid, ing the osmotic pressure between the inside and outside
NaCl concentrations, and pH
of the cell (Kashket, 1987). There are reports describing
Environmental Isolate name strains of lactococci (Uguen et al., 1999) and lactobacilli
conditions
FM TapLac TapSuc TemSor TemGlu TemMan CB (Hutkins et al., 1987; Glaasker et al., 1998) showing
Temperature decreased growth rate with increasing osmolarity of the
15 °C ¡ + + + ¡ + + medium. Uguen et al. (1999) also noted increased amount
30 °C + + + + + + + of glycine betaine, an osmolyte, in lactococci cells when
37 °C + + + + + + +
they were grown in high NaCl concentration. Liu et al.
45 °C + + + + + + +
50 °C + ¡ ¡ ¡ ¡ ¡ ¡ (1998) reported that the uptake of glycine betaine was
induced in cells as an adaptive measure to withstand
Lactic acid concentration (% w/v)
2.5 + + + + + + +
increasing external osmotic pressure. The lactococci isolate,
5 ¡ + + + + + + FM, could be similarly protected to be able to grow at
7.5 ¡ + + ¡ ¡ ¡ ¡ higher NaCl concentration compared to the other isolates
10 ¡ ¡ ¡ ¡ ¡ ¡ ¡ which were lactobacilli. During industrial fermentation, as
15 ¡ ¡ ¡ ¡ ¡ ¡ ¡ lactic acid is being produced by the cells, alkali would be
NaCl concentration (% w/v) pumped into the broth to prevent excessive reduction in
1.5 + + + + + + + pH. Thus, the free acid would be converted to its salt form
2.5 + + + + + + +
which would in turn increase the osmotic pressure on the
5 + + + + + + +
7.5 + ¡ ¡ ¡ ¡ ¡ ¡ cells. Therefore, an LAB strain with high osmotolerance
10 ¡ ¡ ¡ ¡ ¡ ¡ ¡ would be desirable as an industrial strain.
pH
All the isolates, except FM, could grow at pH 4.5. The
4.5 ¡ + + + + + + inability of FM to grow at low pH was consistent with its
7 + + + + + + + inability to grow at high lactic acid concentration, and was
9 + + ¡ + + + + attributed to the low tolerance of L. lactis to free acid (H+)
Methods are described in Section 2.4. compared to Lactobacillus spp. (Kashket, 1987). LAB are
+ indicates colour change from purple to yellow, taken to equate growth. acidophilic but while that means a tolerance to low pH, the
¡ indicates no colour change from purple, taken to equate no growth. latter should be diVerentiated from a situation in which a
high concentration of free acids (H+) also exists because the
ranging from 30 to 50 °C with no growth indicated at 15 °C. free acids could inhibit growth (Amrane and Prigent, 1999).
Wouters et al. (2000) noted reduced glycolytic activity lead- On the other hand, all the isolates except TapSuc, a lacto-
ing to reduced production of lactic acid in L. lactis at low bacilli, could grow in alkaline environment, pH 9. Rhee and
temperature. The ability to grow at high temperature is a Pack (1980) observed that Lactobacillus bulgaricus also
desirable trait as it could translate to increased rate of could not tolerate high pH.
growth and lactic acid production. At the same time, a high
fermentation temperature reduces contamination by other 3.3. Growth and lactic acid production proWles from
microorganisms. time-course studies
FM, however, was the least tolerant to high concentra-
tions of lactic acid as growth was indicated only at 2.5% In the Wrst 6 h, FM, the lactococci strain, grew faster
and not at 5% and higher. The two isolates from tapai, than TapLac and TapSuc, the lactobacilli strains, based on
TapLac and TapSuc, were the most tolerant of high lactic cell dry weight (cdw) measurements. After this, the growth
acid concentrations as they were able to grow at 7.5%. The of FM leveled oV at about 1.7 g/l cdw while TapLac and
other four isolates could tolerate up to 5% lactic acid con- TapSuc continued to register biomass increase until 18 h
centration. None of the seven isolates grew at 10% and 15% before their growth too leveled oV at around 3.4 and
lactic acid concentrations. A higher tolerance to lactic acid 3.2 g/l cdw, respectively (Fig. 1). The low biomass produced
is a desirable trait for an industrial strain of LAB as it could by FM with respect to TapLac and TapSuc correlated well
produce more lactic acid in the fermentation broth without with the lower amount of lactic acid produced by FM
prematurely aVecting itself adversely. (Fig. 2) resulting in a higher pH in the medium (Fig. 3) and
In contrast to its low tolerance to lactic acid, FM was the lower consumption of glucose (Fig. 4). These results were
most tolerant to high NaCl concentration compared to the consistent with those shown in Table 2 where FM was
other six isolates. FM grew in concentrations up to 7.5% found to be unable to tolerate high lactic acid concentra-
while the rest could grow up to 5%. None of the isolates tions and low pH compared to TapLac and TapSuc.
could grow in 10% NaCl. This test gave an indication of the While the growth proWles of TapLac and TapSuc did not
osmotolerance level of a LAB strain. Bacterial cells culti- appear to be diVerent from one another (Fig. 1), the
vated in a high salt concentration would experience a loss amount of lactic acid produced by TapSuc was higher than
of turgor pressure, which would then aVect the physiology, TapLac for the Wrst 24 h (Fig. 2). The production of lactic
enzyme activity, water activity and metabolism of the cells acid by TapSuc also peaked faster at 18 h compared to
 4.00 cose consumption (Fig. 4) of the two lactobacilli strains,
3.50 respectively. At 36 h, the yield of lactic acid from glucose
Cell Dry Weight (g/l)

3.00 was highest with TapLac at 1.9, followed by TapSuc at 1.7,

2.50
and FM at 1.6. This indicated that TapLac had the highest
eYciency in converting glucose to lactic acid. TapLac and
2.00
TapSuc had similar tolerance levels to high temperature
1.50
(up to 45 °C), lactic acid concentration (7.5%), NaCl
1.00 FM concentration (5%), and to low pH (4.5), but the faster pro-
TapLac
0.50 duction of lactic acid by TapSuc might give it a slight
TapSuc
0.00 advantage over TapLac. This diVerence between TapSuc
0 6 12 18 24 30 36 42 48 54 60
Time (Hours) and TapLac was apparent under current experimental con-
ditions, i.e., using MRS medium in which glucose was the
Fig. 1. Biomass produced by three LAB isolates over 54 h. Methods are primary carbon source and the medium was not pH-con-
described in Section 2.5.
trolled. In developing the fermentation process to industrial
120 level, cheaper sources of carbon are necessary and the
Lactic Acid Produced (mM)

100 medium would need to be pH-controlled by incorporating

neutralizing agents such as calcium carbonate into the
80
medium to reduce the inhibitory eVects of free lactic acid on
60 the producer cells. Under such conditions, the growth and
40 lactic acid production of TapSuc and TapLac would need
FM to be re-evaluated.
20 TapLac
TapSuc The time-course study was conducted to compare the
0
0 6 12 18 24 30 36 42 48 54 60
growth and lactic acid producing capacity of two lacto-
Time (Hours) bacilli strains and a lactococci strain, the Wndings of which
supported the results of the rapid screening tests on toler-
Fig. 2. Lactic acid produced by three LAB isolates over 54 h. Methods are
described in Section 2.5. ance of the strains to a range of environmental factors.
There was little doubt that the amount of biomass and lac-
7.5 tic acid produced by the respective strains were limited by
7.0 FM the accumulation of lactic acid in the fermentation broth
6.5 TapLac and the prevailing pH, and reXected the diVerent tolerance
6.0 TapSuc
End Point pH

levels of the strains. A pH-controlled medium would proba-

5.5
bly allow for higher accumulation of lactic acid and would
5.0
be appropriate for future studies where optimization of the
4.5
4.0
fermentation process was the focus.
3.5
3.0 4. Conclusions
2.5
0 6 12 18 24 30 36 42 48 54 60
Time (Hours) This study described the sequential steps of isolating
Fig. 3. pH of the cultures of three LAB isolates over 54 h. Methods are bacteria from some Malaysian food, screening the isolates
described in Section 2.5. for LAB traits, selection of isolates based on a series of tests
for industrially-desirable traits, and Wnally compared the
70 growth and lactic acid production proWles between the lac-
Glucose Consumption (mM)

60 tobacilli and lactococci strains.

50
40
Acknowledgements
30
20 FM
TapLac We thank the Malaysian Ministry of Science, Tech-
10 TapSuc nology and Innovation, for funding this work under the
0 IntensiWcation of Research in Priority Areas (IRPA) pro-
0 6 12 18 24 30 36 42 48 54 60
Time (Hours) gramme, Grant No. 09-02-03-1015.
Fig. 4. Glucose consumption (mM) by three LAB isolates over 54 h. Meth-
ods are described in Section 2.5. References
TapLac in which the production of lactic acid peaked at Amrane, A., Prigent, Y., 1999. DiVerentiation of pH and free lactic acid
36 h. These lactic acid production proWles corresponded eVects on the various growth and production phases of Lactobacillus
well to the proWles on pH of the cultures (Fig. 3) and glu- helveticus. J. Chem. Technol. Biotechnol. 74, 33–40.
Cecchi, L., Malaspina, P., 1991. Lactic acid enantiomers: Separation by
thin layer chromatography on silica gel plates impregnated with Cu2+.
Anal. Biochem. 192, 219–221.
de Man, J.C., Rogosa, M., Sharpe, M.E., 1960. Medium for the cultivation
of lactobacilli. J. Appl. Bacteriol. 23, 130–135.
Gibson, T., Ab-del Malek, T., 1945. The formation of carbon dioxide by
lactic acid bacteria and Bacillus licheniformis and a cultural method of
detecting the process. J. Dairy Res. 14, 35–44.
Glaasker, E., Tjan, F.S.B., Steeg, P.F.T., Konings, W.N., Poolman, B., 1998.
Physiological response of Lactobacillus plantarum to salt and nonelec-
trolyte stress. J. Bacteriol. 180, 4718–4723.
Hutkins, R.W., Ellefson, W.L., Kashket, E.R., 1987. Betaine transport
imparts osmotolerance on a strain of Lactobacillus acidophilus. Appl.
Environ. Microbiol. 43, 2275–2281.
Kashket, E.R., 1987. Bioenergetics of lactic acid bacteria: Cytoplasmic pH
and osmotolerance. FEMS Microbiol. Rev. 46, 233–244.
Kharras, G.B., Sanchez-Riera, F., Severson, D.K., 1993. Polymers of lactic
acid. In: Molby, D.B. (Ed.), Plastics from microbes: Microbial synthesis
of polymers and polymer precursors. Hanser Publ., pp. 93–137.
Liu, S.Q., Asmundson, R.V., Gopal, P.K., Holland, R., Crow, V.L., 1998.
InXuence of reduced water activity on lactose metabolism by Lacto-
coccus lactis subsp. cremoris at diVerent pH values. Appl. Environ.
Microbiol. 64, 2111–2116.
Reid, G., 1999. The scientiWc basis for probiotic strains of Lactobacillus.
Appl. Environ. Microbiol. 65, 3763–3766.
Rhee, S.K., Pack, M.Y., 1980. EVect of environmental pH on fermentation
balance of Lactobacillus bulgaricus. J. Bacteriol. 144, 217–221.
Tsai, S.P., Coleman, R.D., Moon, S.H., Schneider, K.A., Millard, C.S.,
1993. Strain screening and development for lactic acid fermentation.
Appl. Biochem. Biotechnol. (39/40), 323–335.
Uguen, P., Hamelin, J., Le Pennec, J., Blanco, C., 1999. InXuence of
osmolarity and the presence of osmoprotectant on Lactococcus lactis
growth and bacteriocin production. Appl. Environ. Microbiol. 63,
291–293.
Vaccari, G., Gonzalez-Vara, A., Campi, A.L., Dosi, E., Brigidi, P., Mat-
teuzzi, D., 1993. Fermentative production of L-lactic acid by Lacto-
bacillus casei DSM 20011 and product recovery using ion exchange
resins. Appl. Microbiol. Biotechnol. 40, 23–27.
Wouters, J.A., Kamphuis, H.H., Hugenholtz, J., Kuipers, P., De Vos,
W.M., Abee, T., 2000. Changes in glycolytic activity of Lactococcus
lactis induced by low temperature. Appl. Environ. Microbiol. 66,
3686–3691.