Biotechnol. Appl. Biochem.
(2002) 36, 1–6 (Printed in Great Britain)
Production and characterization of a thermostable
β-galactosidase from Bacillus coagulans RCS3
Navneet Batra*, Jagtar Singh*, Uttam C. Banerjee†, Pratap R. Patnaik‡
and Ranbir C. Sobti*1
*Department of Biotechnology, Panjab University, Chandigarh, India, †Department of Biotechnology, National Institute of
Pharmaceutical Education and Research, Mohali, India, and ‡Institute of Microbial Technology, Sector 39-A, Chandigarh, India
A strain of Bacillus coagulans RCS3 isolated from a
hot-water spring produced significant β-galactosidase
activity at 10 days of growth in a flask. While enzyme
production was maximum at 50 mC, the highest activity
was at 65 mC, where the half-life was 2 h. A 2 mC decrease
in temperature increased the half-life to 15 h without
significantly changing the activity, suggesting that
63 mC is the temperature of preference compared
with 65 mC for a combination of good activity and
stability. The β-galactosidase was also stable over
pH 5–8, with peak activity at pH 6–7. It was strongly
and competitively inhibited by the hydrolysis product
galactose. Bivalent cations (Cu2+, Ni2+ and Hg2+) in the
concentration range of 0.5– 2.0 mM also inhibited
enzyme activity. Both lactose solution and whey could
be hydrolysed substantially within 36 h at 50 mC. The
thermostability and pH-stability and good hydrolytic
capability make this enzyme potentially useful in the
dairy industry.
Introduction
The ability of β-galactosidase (lactase ; EC 3.2.1.23) to
hydrolyse the β1 4-D-galactosidase linkage is one of the
most promising applications of enzymes in food processing
[1–5]. The enzyme hydrolyses lactose in milk to glucose and
galactose, thereby increasing its digestibility and sweetness.
This supplements the low levels of intestinal lactase, which
limit the digestion of milk and can thus lead to gastrointestinal
diseases [6–11]. Lactase may therefore be usefully applied to
whey and its constituents to make higher-solid syrups,
alcoholic beverages, animal feed and food desserts. A
decrease in lactose can increase sweetness and decrease
grittiness [3,4,12].
β-Galactosidase is widely distributed in Nature and is
produced by animals, plants and micro-organisms. While
different microbial sources [3,13,14] have been studied in
respect of this enzyme, very few products show good
activity at high temperature. Thermostability is an important
property, not only in food processing, but also in other
biotechnological applications, because it decreases con-
tamination and increases the shelf life of the enzyme. The
1
present study describes a new thermophilic strain of Bacillus
coagulans with thermostable β-galactosidase that shows
promise for the efficient hydrolysis of the lactose in milk and
whey.
Materials and methods
Isolation and growth of the thermophilic micro-
organism
A total of 16 clones of thermophilic microbes producing
β-galactosidase were isolated from hot-water springs at
Manikaran (Himachal Pradesh, India). The temperature of
the aqueous environment, the source of samples, ranged
from 45 to 90 mC, while the soil temperature was approxi-
mately 42 mC. The clones were numbered RCS1–RCS16. As
a result of monitoring the production and stability of
β-galactosidase obtained from different clones (results not
shown), RCS3 was selected for further exploration.
Strain RCS3, a Gram-positive, spore-forming microbe
was identified as Bacillus coagulans by the Microbial Type
Culture Collection and Gene Bank, Institute of Microbial
Technology, Chandigarh, India, and assigned the accession
number 3244. The culture medium for enzyme production
contained, per litre : yeast extract, 5.0 g ; lactose, 10.0 g ;
K2HPO4, 2.5 g ; and MgSO4 : 7H2O, 0.5 g. The pH of the
medium was adjusted to 6.8–7.0. The cultures were grown
in Ehrlenmeyer flasks at a shaking speed of 150 rev.\min with
temperature maintained at 40 mC. After 10 days, the cells
were harvested by centrifugation at 10 000 rev.\min (C24
centrifuge ; Remi Equipments, New Delhi, India). The cell-
free supernatant was adjusted to 30–70 % saturation with
(NH4)2SO4. The precipitates were removed by centrifugation
at 15 000 rev.\min for 1 h and dissolved in 0.1 M sodium
phosphate buffer, pH 7.0, followed by dialysis overnight
against two changes of the buffer. The dialysed samples were
used as sources of enzyme for characterization.
Key words : lactose hydrolysis, inhibition, thermostability.
Abbreviations used : ONP, o-nitrophenol ; ONPG, o-nitrophenyl
β-D- galactopyranoside.
1
To whom correspondence should be sent (e-mail
rcsobti!panjabuniv.chd.nic.in).
# 2002 Portland Press Ltd
2 N. Batra and others

Enzyme assay partially purified β-galactosidase to 25 ml of reaction mix-

The enzyme activity was studied using the technique
described by Griffith and Muir [15] with certain modifica-
tions. The assay mixture (3.0 ml) containing 2.0 ml of 5.0 mM
o-nitrophenyl β-D-galactopyranoside (ONPG) in 0.1 M so-
dium phosphate buffer, pH 7.0, and 1.0 ml of diluted enzyme
was incubated at 60 mC for 10 min. A 2.0 ml portion of
0.5 mM ice-cold Na2CO3 was then added and the A420 was
read in a DU 640B spectrophotometer (Beckman) with
appropriate substrate and enzyme blanks. o-Nitrophenol
(ONP) was used as standard.
To analyse the intracellular β-galactosidase, the cells
were washed twice with buffer and treated with toluene\
acetone mixture (2 : 1, v\v). After 5 min, the cells were
centrifuged at 9000 rev.\min for 20 min, suspended in the
buffer and assayed as described above with a suitable blank.
The activity of the enzyme was expressed in units, 1 unit
of enzyme being the amount that produces 1 µmol of
ONP\min per ml at pH 7.0 and 60 mC. The values shown in
the Figures are the means of three measurements in each
case.
Protein estimation
Protein concentration was determined by the method of
Lowry et al. [16]. To prepare the standard curve, BSA was
used.
Optimum pH value and stable pH range
The optimum pH for β-galactosidase activity was studied by
assaying enzyme activity using the following buffers (0.1 M) :
potassium acetate, pH 4.0–5.0 ; sodium phosphate, pH
6.0–7.0 ; and glycine\NaOH, pH 9.0–10.0 (with suitable
dilution of enzyme in the same buffer).
Stability profiles with respect to pH were obtained by
incubating the enzyme in buffers at different pH values
(4 –10) for 24 h at 37 mC, followed by measurement of the
residual activity.
ture. The flasks were incubated at 50 mC and 100 rev.\min.
Aliquots (0.5 ml) were removed at regular intervals and
the glucose concentration was determined as described
for the Glucose (Trinder) Kit no. 135-100 (Sigma). The
percentage hydrolysis of lactose was calculated in terms of
percentage of maximum efficiency. Lactose solution (1 %)
was mixed with β-galactosidase and, after a sufficient period
of incubation and hydrolysis, the final solution was precipi-
tated with acetone. The supernatant, along with standard
sugars (lactose, glucose and galactose), was spotted on Kie-
selgel 60 P254 plates (Merck, Darmstadt, Germany). The
spots were developed in chloroform\acetone\water (6 : 7 : 1,
by vol.). The plates were dried and sprayed with aniline\
diphenylamine\phosphoric acid (1 : 1 : 1, v\w\v) in acetone
and incubated at 100 mC for 10 min.
Results and discussion
The time course of production of β-galactosidase is shown in
Figure 1. After 10 days, cultures of Bacillus coagulans RCS3
yielded 6.26 units of extracellular enzyme\ml of broth
culture. Following a lag period of 24 h, the activity of the
enzyme increased rapidly and continued to increase after 10
days. By contrast, intracellular β-galactosidase activity, esti-
mated from cell lysates, was nearly constant at about
0.42 units\ml throughout the fermentation period. This
observation and the constancy of cell density after 24 h
(Figure 1) suggest that β-galactosidase synthesis by this
organism occurs largely after the cells have matured and it
is secreted out.
Both growth and enzyme production remained stable

Optimum temperature and stability

The optimum temperature was determined by incubating
the assay mixture at different temperatures (30–90 mC).
Stability of β-galactosidase was followed similarly by incubat-
ing the enzyme at different temperatures, withdrawing
samples periodically and determining residual lactase activity
at 60 mC.

Hydrolysis of lactose
Figure 1 Time course of β-galactosidase production by Bacillus coagulans
Hydrolysis of lactose (1–5 % in phosphate buffer, pH 7.0) or RCS3 using 1 % lactose medium
whey (obtained from the local market and containing 3.5 %
Abbreviations : I.U., unit ; d, day.
lactose) was performed by adding 5 ml (32.2 units) of

# 2002 Portland Press Ltd


Figure 2 Effect of initial pH on growth and production of β-galactosidase
from Bacillus coagulans RCS3
Figure 3 Effect of temperature on the production of β-galactosidase by
Bacillus coagulans RCS3
over a range of the initial pH 5–8 (Figure 2). Although the pH
of the final medium after growth was 8.4 –8.9, for all starting
values, maximum enzyme activity was attained with an initial
pH of 6.8. The change of pH during fermentation followed
that of cell growth, i.e. an early increase and then fairly
constant values (Figure 1).
To study the effect of temperature of enzyme pro-
duction, fermentations were carried out in the range
25–60 mC. While β-galactosidase activity increased with time
for all temperatures, 50 mC was the best temperature at all
times (Figure 3). An interesting feature of Figure 3 is the large
New thermostable β-galactosidase 3
Figure 4 Optimum temperature and thermal stability of β-galactosidase
from Bacillus coagulans RCS3
increase in enzyme activity from 30 to 40 mC. From the data
of the day 10, this increase was 360 %, whereas the further
increase until the optimum temperature was just 7 %.
Similarly, the activities fall drastically on heating the culture
broth beyond 50 mC. For industrial production, these
observations imply that it might be preferable to operate
somewhat below 50 mC so that disturbances or limitations of
control actions do not cause a sharp decrease in enzyme
synthesis. This approach is analogous to the use of subop-
timal dilution rates in continuous fermentations so as to
avoid wash-out of cells.
The optimum pH for enzyme production was 6.0–7.0.
This is comparable with pH 6.0–6.4 reported for B.
stearothermophilus [17] and B. coagulans [18,19], and 7.0
for B. subtilis [20]. While the activity declined sharply on
either side of this optimal range, a sudden increase was
observed at pH 9.0, suggesting a change in conformation.
This increase in activity from pH 8 to pH 9 was 23 % and
was reproducible, indicating that it was not an experimental
error. This enzyme was more than 90 % stable within pH
range 6–9, after which there was a small, but perceptible, fall
to 86 % of the peak activity at pH 10. Just as for production,
the activity fell sharply below pH 6.0. This might be due to
denaturation of the enzyme. Similar observations have been
reported for β-galactosidase from Bacillus sp. TA 11[21] and
B. macerans [22].
Figure 4 displays the optimum assay temperature and
stability at different temperatures for enzyme generated at
the optimum temperature. The optimum temperature was
# 2002 Portland Press Ltd
4 N. Batra and others

65 mC, but more than 90 % of activity was retained at 63 mC

and 67 mC, thus allowing small variations. While B. circulans
has a similar optimum, this is the upper end of the range
55–65 mC [23]. Moreover, B. circulans is mesophilic, and for
thermophilic organisms the reported optima are a few
degrees lower than for this strain [18,19,23]. From the data
in Figure 4, the half-life at 65 mC was 120 min, whereas at
63 mC it was 15 h. The dramatic improvement in stability and
the detection of more than 90 % of the peak activity at 63 mC
imply that 63 mC may be the best overall temperature. The
half-life again increased significantly to 41 h at 60 mC, but
the activity was almost entirely lost at 70 mC. Griffiths and
Muir [15] observed a half-life of 83 min at 65 mC for β-
galactosidase from a Bacillus sp. and a Thermusspecies [24,26]
have been reported to produce more stable β-galactosidase
than that from Bacillus sp. [21,22].
As described above, β-galactosidase activity was as-
sayed by monitoring the hydrolysis of ONPG at concentra-
tions from 1 mM to 25 mM. Lineweaver–Burk plots (Figure
5) yielded straight lines with a common intercept on the 1\v
Figure 5 Lineweaver–Burk plot for Bacillus coagulans RCS3 β-galactosidase axis and different slopes for different values of inhibitor
for hydrolysis of ONPG
concentration. This indicates competitive inhibition by
galactose, which agrees with the observations of most other
authors [15,24,26–27] ; however, non-competitive inhibition
Table 1 Effect of concentration of carbohydrates on the activity of
β-galactosidase
in Aspergillus oryzae has been reported by Shukla and Caplin
[28]. The product of lactose hydrolysis, i.e. galactose and
Carbohydrate Concentrationa Relative activity glucose, inhibited the RCS3 β-galactosidase activity to
Control – 100n0 different extents (Table 1), as has also been observed in
Glucose 0n01 91n4 other microbial β-galactosidases also [21,24] ; however, in
0n05 89n1 Thermoanaerobacter TP6 B1 [25], glucose acts as an activator
0n10 87n9
over the concentration range 1–10 %. The inhibition was
Galactose 0n01 47n8
0n05 16n6 most pronounced in the case of galactose, which inhibited
0n10 8n6 91 % of the activity at a concentration of 0.1 M, while only
0n20 8n7 12 % of the activity was inhibited by 0.1 M glucose. To
Lactose 0n01 94n8 evaluate the values of the parameters in the hydrolysis rate
Raffinose 0n01 96n5
Maltose 0n01 92n6 equation, the following equation [27] for competitive
Sorbitol 0n01 96n9 inhibition was regressed non-linearly to the data :
Fructose 0n01 99n6
Sucrose 0n01 100n4 Vmaxs
Starch 5n0 % 100n3 vl (1)
[Km\(1+i\Ki)]+s
a
All concentrations are in M, except for starch.

Table 2 Percentage deviations of reaction velocities predicted by eqn. (1) from corresponding experimental values

i is the inhibitor (galactose) concentration in mM.

Percentage deviation
[ONPG]
(mM) i… 0 2 4 6 8 10 15 20

0n1 3n71 k1n70 k1n15 k4n07 k2n24 2n05 k7n19 k4n52

0n2 k5n11 k7n44 k5n95 k7n64 k7n15 k2n50 k8n16 k13n79
0n3 5n38 2n57 2n32 0n96 k0n91 4n05 k1n98 0n06
0n4 k2n17 k0n98 k2n70 k3n91 k1n24 k1n15 2n22 1n59
0n5 k0n24 3n42 2n65 6n92 2n88 k1n11 1n03 2n58
0n8 0n83 1n64 4n08 2n58 4n39 0n30 3n29 2n59
1n0 k0n43 k1n99 k3n00 k2n61 k2n95 0n03 k2n08 k1n86

# 2002 Portland Press Ltd

 New thermostable β-galactosidase 5

Figure 7 Progress of enzymic hydrolysis of lactose (1–5 %) and whey

Figure 6 Experimental and predicted velocities for hydrolysis of ONPG by
permeate at 50 mC
β-galactosidase from Bacillus coagulans RCS3

Table 3 Inhibition and activation by metal ions (a) and other agents (b) of
β-galactosidase from Bacillus coagulans RCS3 deviations (Table 2) and from both the linearized (Line-
(a) weaver–Burk) plots (Figure 5) and the original non-linear
plots (Figure 6).
Relative activity
Table 3 shows the effect of different univalent and
Metal ion 0n5 mM 1n0 mM 2n0 mM bivalent cations (Table 3a) and other specific inhibitors and
Mg 2+
97n0 98n6 101n4
activators (Table 3b). No bivalent cation tested stimulated
Ca2+ 103n7 105n6 107n5 the enzyme activity, as also observed by Choi et al. [21] for
Mn2+a 102n3 104n7 105n6 a different thermophilic bacillus. Cu2+, Hg2+ and Na2+ acted as
Mn2+ 111n1 115n6 119n6
Al3+ 95n2 100n6 99n5
strong inhibitors, while Mn2+ in the form of chloride salt was
Na+ 97n1 98n4 99n7 a weak activator. A particularly useful property of our β-
K+ 94n2 95n4 97n3 galactosidase is the absence of inhibition by Ca2+, which is an
Co2+ 96n9 96n6 92n4
Fe2+ 90n7 90n0 82n9
important component of milk. Although reducing agents like
Cu2+ 102n3 9n7 8n9 dithiothreitol, 2-mercaptoethanol and cysteine have been
Ni2+ 56n0 40n2 23n7 reported [24– 26] to activate the enzyme, no such activation
Hg2+ 36n8 34n8 34n3
was observed in β-galactosidase from B. coagulans RCS3 ;
(b)
however, the chelating agent EDTA acted as weak activator
Relative activity at 20 mM.
The effectiveness of the enzyme in hydrolysing lactose
Other agent 5n0 mM 10n0 mM 20n0 mM
solutions and whey are examined in Figure 7. Hydrolysis of
EDTA 101n8 108n2 116n3 lactose was indicated by the formation of glucose and
Dithiothreitol 106n5 106n3 105n3 galactose on TLC, and it signifies the ability of β-galactosidase
2-Mercaptoethanol 95n2 110n3 99n2
Cysteine 94n3 95n5 97n3 to break the β1 4 linkage. More than 90 % hydrolysis of a
Control 100n0 100n0 100n0 1 % lactose solution could be accomplished in 36 h, and
a
As sulphate ; all others are chlorides.
substantial hydrolysis (77 %) was possible even at a con-
centration of 5 %. In spite of the presence of other
constituents, 65 % hydrolysis of commercial whey occurred
The values obtained for the parameters were : Vmax l within the same duration. A decrease in hydrolysis with
1.75 µmol\min per mg of protein, Km l 5030 µM and Ki l increasing lactose content should be expected, since lactose
7040 µM. Using these values, the predicted velocities agree and its products, galactose and glucose, inhibit the process
well with experimental values, as seen from the percentage (Table 1).

# 2002 Portland Press Ltd

6 N. Batra and others

Conclusions 8 Buller, H. A. and Grand, R. J. (1990) Annu. Rev. Med. 44,

The present strain of B. coagulans RCS3 can grow and 141–148
produce β-galactosidase at elevated temperatures (up to 9 Khedkar, C. D., Mantri, J. M. and Khedkar, G. D. (1994) Cult.
50 mC) over the range pH 5–8. The enzyme has maximum Dairy Product J. 29, 1–14
activity at 65 mC and pH 6.8, with a half-life of 2 h at this 10 Solomons, N. W. (1996) Rev. Invest. Clin. 48 (Suppl.), 1–13
temperature and 15 h at 63 mC. It can hydrolyse lactose 11 Inman Felton, A. E. (1999) J. Am. Diet. Assoc. 99, 481–489
solutions and whey efficiently and is not inhibited by Ca2+, a 12 Zadow, J. G. (1992) in Whey and lactose processing (Zadow
prime constituent of milk. However, hydrolysis is inhibited J. G., ed.), pp. 361–408, Elsevier Science Publishers Ltd., Oxford
by the product galactose and therefore fed-batch operation 13 Blankenship, L. C. and Wells, P. A. (1974) J. Milk Food Technol.
is to be preferred for controlled feeding of lactose. The 37, 199–202
advantages mentioned above and the extracellular secretion 14 Bacerra, M. and Siso, M. I. G. (1996) Enzyme Microb. Technol.
of the enzyme make β-galactosidase from B. coagulans RCS3 19, 39–44
a good candidate for the dairy industry. 15 Griffiths, M. W. and Muir, D. D. (1978) J. Sci. Food Agric. 29,
753–761
16 Lowry, O. H., Rosenbrough, N. J., Farr, A. L. and Randall, R. J.
Acknowledgments (1951) J. Biol. Chem. 193, 265–275
17 Goodman, R. and Pederson, D. M. (1976) Can. J. Microbiol 22,
N. B. and J. S. thank the University Grants Commission and 817–825
the Council of Scientific and Industrial Research (CSIR), 18 Long, M. E. and Lee, C. K. (1979) U.S. Patent 4,179,335
New Delhi, India, respectively for providing research fellow- 19 Lewin, R. E. and Mahoney, R. R. (1981) Antonie Van Leeuwen-
ships. hoek 47, 53–64
20 Anema, P. J. (1964) Biochem. Biophys. Acta 89, 495–502
21 Choi, J. Y., Kim, II, H., Lee, B. H. and Lee, J. S. (1995) Biotechnol.
References Appl. Biochem. 22, 191–201
22 Miyazaki, Y. (1988) Agric. Biol. Chem. 52, 625–631
1 Shukla, T. P. (1975) CRC Crit. Rev. Food Technol. 5, 325–356 23 Iida, T., Ozaki, S., Kotaka, T. (1980) U.S. Patent 4,237,230
2 Belem, M. A. and Lee, B. H. (1998) Cric. Rev. Food. Sci. Nutr. 24 Cowan, D. A., Daniel, R. M., Martin, A. M. and Morgan, H. W.
38, 565–598 (1984) Biotechnol. Bioeng. 26, 1441–1145
3 Richmond, M. L., Gray J. I. and Stine, C. M. (1981) J. Dairy Sci. 25 Lind, D. L., Daniel, R. M., Cowan, D. A. and Morgan, H. W.
64, 1759–1771 (1989) Enzyme Microb. Technol. 11, 180–186
4 Gekas, V. and Lopez Leiva, M. (1985) Process Biochem. 20, 26 Berger, J., Lee, B. H., Lacroix, C. (1997) Biotechnol. Appl.
2–12 Biochem. 25, 29–41
5 Garcia-Garibay, M. and Gomez-Ruiz, L. (1996) Rev. Invest. Clin. 27 Bowden, A. C. (1974) Biochem. J. 137, 143–144
48 (Suppl.), 51–61 28 Shukla, H., Chaplin, M. (1993) Enzyme Microb. Technol. 15,
6 Vera, T. H., Marteau, P. and Korpela, R. (2000) J. Am. Coll. Nutr. 297–299
19 (Suppl.), 165S–175S
7 Fischer, L., Scheckermann, C. and Wagner, F. (1995) Appl.
Received 8 November 2001\25 February 2002; accepted 11 March 2002
Environ. Microbiol. 61, 1497–1501

# 2002 Portland Press Ltd