A novel l-arabinose isomerase from Lactobacillus fermentum CGMCC2921 for d-tagatose production: Gene cloning, purification and characterization

Article abstract of DOI:10.1016/j.molcatb.2011.01.010

The araA gene encoding l-arabinose isomerase (l-AI) from the acidophilus bacterium Lactobacillus fermentum CGMCC2921 was cloned and over-expressed in Escherichia coli. The open reading frame of the l-AI consisted of 1425 nucleotides encoding 474 amino acid residues. The molecular mass of the enzyme was estimated to be approximately 53 kDa on SDS-PAGE. The purified recombinant enzyme showed maximum activity at 65 °C and pH 6.5, which were extremely suitable for industrial applications. It required divalent metal ions, either Mn²⁺ or Co²⁺, for enzymatic activity and thermostability improvement at higher temperatures. The enzyme was active and stable at acidic pH, it exhibited 83% of its maximal activity at pH 6.0 and retained 88% of the original activity after incubation at pH 6.0 for 24 h. Kinetic parameter study showed that the catalytic efficiency was relatively high, with a k _cat/K_m of 9.02 mM^-1 min^-1 for d-galactose. The purified L. fermentum CGMCC2921 l-AI converted d-galactose into d-tagatose with a high conversion rate of 55% with 1 mM Mn²⁺ after 12 h at 65 °C, suggesting its excellent potential in d-tagatose production.

Full text of DOI:10.1016/j.molcatb.2011.01.010

Journal of Molecular Catalysis B: Enzymatic 70 (2011) 1–7
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
A novel l-arabinose isomerase from Lactobacillus fermentum CGMCC2921 for
d-tagatose production: Gene cloning, purification and characterization
∗
Zheng Xu, Yujia Qing, Sha Li, Xiaohai Feng, Hong Xu , Pingkai Ouyang
State Key Laboratory of Materials-Oriented Chemical Engineering, College of Food Science and Light Industry, Nanjing University of Technology, No. 5 Xinmofan Road, Gulou District,
210009 Nanjing, China
a r t i c l e i n f o
a b s t r a c t
Article history:
The araA gene encoding l-arabinose isomerase (l-AI) from the acidophilus bacterium Lactobacillus fer-
mentum CGMCC2921 was cloned and over-expressed in Escherichia coli. The open reading frame of the
l-AI consisted of 1425 nucleotides encoding 474 amino acid residues. The molecular mass of the enzyme
was estimated to be approximately 53 kDa on SDS–PAGE. The purified recombinant enzyme showed max-
Received 16 September 2010
Received in revised form 5 December 2010
Accepted 23 January 2011
Available online 28 January 2011
◦
imum activity at 65 C and pH 6.5, which were extremely suitable for industrial applications. It required
2
+
2+
divalent metal ions, either Mn or Co , for enzymatic activity and thermostability improvement at
higher temperatures. The enzyme was active and stable at acidic pH, it exhibited 83% of its maximal activ-
ity at pH 6.0 and retained 88% of the original activity after incubation at pH 6.0 for 24 h. Kinetic parameter
Keywords:
l-Arabinose isomerase
Lactobacillus fermentum
Characterization
d-Tagatose
−
1
−1
study showed that the catalytic efficiency was relatively high, with a kcat/Km of 9.02 mM min for d-
galactose. The purified L. fermentum CGMCC2921 l-AI converted d-galactose into d-tagatose with a high
2
+
◦
conversion rate of 55% with 1 mM Mn after 12 h at 65 C, suggesting its excellent potential in d-tagatose
production.
©
2011 Published by Elsevier B.V.
1
. Introduction
As a matter of fact, the use of d-tagatose was limited mostly
due to its scarcity in nature and costly methods of production.
Recently, there has been great interest in the biological manu-
facture of d-tagatose from d-galactose with l-AIs. A number of
l-AIs have been identified from various microorganisms, such
as Geobacillus stearothermophilus, Geobacillus thermodenitrificans,
Thermotoga maritima, and Thermus sp. [8–11]. Nevertheless, l-
Als from those microbes were optimally active at an alkalescent
pH (7.5–8.5), which would cause browning reaction and forma-
tion of undesirable sub-products in d-tagatose production process
[12]. Other l-AIs, from E. coli, Lactobacillus gayonii, and Mycobac-
terium smegmatis [13–15] exhibited a lower optimum temperature
l-Arabinose isomerase (l-AI; EC 5.3.1.4) is an intracellular
enzyme that catalyzes the reversible isomerization of l-arabinose
to l-ribulose, a component of the pentose phosphate or the phos-
phoketolase pathway [1]. l-AIs are also referred to as d-galactose
isomerases due to their ability, in vitro, to isomerase d-galactose
into d-tagatose.
d-Tagatose is a hexoketose monosaccharide sweetener, which
is an isomer of d-galactose and rarely found in nature. The sweet-
ness of d-tagatose is equivalent to sucrose, but with only 38% of the
calories when compared in 10% solutions [2]. Physiological studies
demonstrated that d-tagatose consumption at recommended dose
not promote tooth decay or elicit any increase in blood glucose
level [3]. In addition, d-tagatose has been shown to have numer-
ous health and medical benefits, including treatment of obesity
◦
(30–45 C), which may be difficult to realize high conversion rate
of d-tagatose from d-galactose [16]. Thus, l-AI with high activ-
ity and stability at a moderately low pH and higher temperatures
would have the greatest potential for the production of the d-
tagatose.
[
4], prevention of dental caries, regulation of intestinal flora [5],
improvement of pregnancy and fetal development, and reduction
of symptoms of type 2 diabetes [6]. Based on these properties, d-
tagatose has attracted a great deal of attention in recent years as a
low calorie sugar-substituting sweetener, which was also approved
as a “generally recognized as safe (GRAS)” material under FDA reg-
ulations [7].
For the purpose of attaining l-AIs suitable for d-tagatose pro-
duction, new organisms carrying the acidophilic and thermostable
target enzyme need to be screened. Lactic acid bacteria are well
known for their acid tolerance, and previously reported l-AI from
Lactobacillus genera was extraordinarily thermostable [17]. Fur-
thermore, l-AI has not yet been characterized from Lactobacillus
fermentum. Thus, we chose L. fermentum CGMCC2921 as a can-
didate. Here we report the gene cloning, amino acid sequence
inspection, over-expression and characterization of a novel l-
arabinose isomerase (LFAI) from L. fermentum CGMCC2921.
∗ Corresponding author. Tel.: +86 25 83172061; fax: +86 25 83587326.
E-mail address: xuh@njut.edu.cn (H. Xu).
1
381-1177/$ – see front matter © 2011 Published by Elsevier B.V.
doi:10.1016/j.molcatb.2011.01.010

2
Z. Xu et al. / Journal of Molecular Catalysis B: Enzymatic 70 (2011) 1–7
2
2
. Materials and methods
Tagatose production was also confirmed by high-performance
liquid chromatography (HPLC) using Rezex RCM-Monosaccharide
column (300 mm × 7.8 mm). The products were separated by iso-
cratic elution with water at a flow rate of 0.5 mL/min and detected
with a refractive index detector (SHODEX RI-101). Solutions of d-
galactose and d-tagatose at 10 g/L each were used as standards. One
unit of l-AI activity was defined as the amount of enzyme catalyzing
the formation of 1 ␮mol keto-sugar per minute.
.1. Strains and materials
L. fermentum CGMCC2921, which was used as a source of
genomic DNA for araA gene cloning, was isolated from traditional
Chinese pickles and grown at 37 C in MRS medium. The host strains
Escherichia coli JM109 and E. coli BL21 (DE3) were obtained from
Novagen and grown in Luria–Bertani (LB) medium. Ex-taq DNA
polymerase, T4 DNA ligase, restriction endonucleases and pMD-
◦
2.5. Effect of temperature and pH on enzyme activity and stability
1
8T Vector were purchased from Takara Biotechnology (Takara,
China). All the other chemicals were of the highest reagent grade
and commercially available.
The temperature optimum of LFAI activity was measured by
◦
assaying the enzyme samples over the range of 30–90 C, at pH 6.5.
Three buffer systems (sodium acetate/phosphate/Tris–HCl) were
2.2. DNA amplification and subcloning of the l-AI gene
◦
used for measuring the pH optimum of enzyme activity at 65 C.
The thermal stability of LFAI was studied by incubating the enzyme
Genomic DNA was isolated from L. fermentum CGMCC2921
◦
2+
in phosphate buffer (pH 6.5) at 75 C in the presence of 1 mM Mn
,
using a Takara Bacterial Genomic DNA Extraction Kit (Takara,
China). Oligonucleotide primers specific for the full-length araA
gene were derived from the putative araA gene of L. fermentum
IFO3956 (Gene bank accession no. YP 001844370). The forward
2+
2+
2+
2
mM Co , 1 mM Mn plus 2 mM Co and without adding ions,
respectively. Samples were withdrawn at certain time intervals and
residual activity was estimated under standard assay conditions. To
determine the pH stability, the enzyme was incubated at various
ꢀ
primer was araAs, 5 -AGAGAATTCATGCGTAAGATGCAAGATTAC-
◦
pH values (5.0, 5.5 and 6.0) at 4 C for up to 24 h, the residual activity
ꢀ
ꢀ
3
5
(EcoRI site is underlined). The reverse primer was araAr,
was also estimated.
ꢀ
-AAGCTCGAGCTACTTGATGTTGATAAAGT-3 (XhoI site is under-
lined). The amplified 1.4 kb DNA fragment was cloned into the
pMD18-T Vector and transformed into E. coli JM109 competent
cells. Transformants containing the pMD18-T Vector harbouring
the araA gene were selected, plasmid DNA (pMD18-T-araA) was
isolated from the transformants and sequenced. To over-produce
LFAI in E. coli, an expression plasmid pET-araA was constructed
by ligation of gene araA, digested by EcoRI and XhoI from pMD-
2
.6. Effect of various metal ions on enzyme activity
Before studying the effects of metal ions on l-AI activity, the
purified enzyme was dialyzed against phosphate buffer contain-
ing 10 mM EDTA overnight at 4 C. Subsequently, the enzyme was
dialyzed against phosphate buffer to remove EDTA. Then, the enzy-
matic activity was assessed in the presence of several metal ions
◦
1
8T-araA, into the corresponding restriction sites of the pET-28a
(MgCl , MnCl , CoCl , ZnCl , CaCl , CuCl , NiCl , and BaCl ) with
2 2 2 2 2 2 2 2
plasmid (Novagen) and transformed into E. coli BL21 (DE3).
a final concentration of 1 mM. For the purpose of determining the
effect of Mn2+ and Co concentration on enzyme activity, the reac-
tions were performed using the EDTA-treated enzyme with the
addition of Mn2+ and Co at concentrations from 0.1 to 5 mM. Then
samples were taken for activity assays.
2+
2.3. Over-expression and purification of the recombinant l-AI
2+
E. coli BL21 (DE3) cells harbouring the pET-28a plasmid carry-
◦
ing the araA gene were grown at 37 C in LB medium containing
kanamycin (25 ␮g/mL) until the OD600 nm reached 0.5. Then, IPTG
◦
2.7. Determination of substrate specificity and kinetic parameters
was added at 1 mM and growth was carried out at 20 C for extra
1
2 h. Cells were harvested by centrifugation at 8000 × g for 10 min,
A substrate concentration of 50 mM was used to investigate the
substrate specificity of the enzyme. Reactions were carried out
under standard reaction conditions with different substrates (l-
arabinose, l-xylose, l-ribose, d-galactose, d-glucose, d-xylose, and
d-mannose). The values were compared to the enzyme activity in
the d-galactose solution.
washed with 50 mM phosphate buffer (pH 6.5). After sonication, the
lysates were centrifuged to remove the cell debris and the super-
natant was filtered through a 0.2 ␮m filter. The filtrate was loaded
on a Ni-NTA resin column equilibrated with equilibration buffer
(
300 mM NaCl, 50 mM NaH PO , pH 8.0). The column was then
2 4
washed with the same buffer containing 10 mM imidazole, and a
gradient of imidazole (from 50 mM to 250 mM) was applied to elute
the recombinant protein. The fractions containing enzyme activity
were pooled and dialyzed against phosphate buffer, and the dia-
Kinetic parameters of LFAI were determined in 50 mM phos-
phate buffer (pH 6.5), 1 mM Mn2+, 2 mM Co and 1–600 mM
substrate (d-galactose or l-arabinose). The samples were incubated
2+
◦
◦
at 65 C for 10 min. The enzyme reaction was stopped by chilling
lyzed enzyme preparation was stored at 4 C. Protein purity was
on ice, and the amount of d-tagatose (l-ribulose) was determined.
Kinetic parameters, such as Km (mM) and Vmax (U/mg protein) for
substrates were obtained using the Lineweaver–Burk equation. All
assays were performed in triplicate at least two separate times.
determined by SDS–PAGE analysis.
2.4. Analytical methods
Protein concentrations were determined by the Bradford
method using bovine serum albumin as a standard protein [18].
l-AI activity was measured by determining the amount of formed
d-tagatose (l-ribulose). Under standard conditions, the reac-
tion mixture of 1 mL contained 50 mM d-galactose (l-arabinose),
2.8. Analysis of the isomerization of d-galactose to d-tagatose
with LFAI
The conversion media (1 ml) contained 50 mM of d-galactose,
1 mM Mn2+ and 1 mg of the purified enzyme (9.98 U) in 50 mM
phosphate buffer (pH 6.5). The study of the kinetic conversion of
1
mM MnCl , 2 mM CoCl , 100 ␮L of enzyme preparation at a
2 2
suitable dilution and 50 mM phosphate buffer (pH 6.5). The
◦
◦
◦
◦
reaction mixture was incubated at 65 C for 10 min, followed
d-galactose was investigated until 24 h at 60 C, 65 C and 70 C.
Samples were taken periodically, and the concentration of the gen-
erated d-tagatose was determined by the cystein–carbazol–sulfuric
acid method and confirmed by HPLC as indicated in Section 2.4. The
by cooling samples on ice to stop the reaction. The generated
d-tagatose (l-ribulose) was determined by cysteine–carbazole
method [19], and the absorbance was measured at 560 nm. d-

Z. Xu et al. / Journal of Molecular Catalysis B: Enzymatic 70 (2011) 1–7
3
Fig. 1. Multiple sequence alignment of l-arabinose isomerases (l-AIs) from L. fermentum CGMCC2921 (LFAI), E. coli (ECAI), A. acidocaldarius (AAAI), B. subtilis (BSAI), Thermus
sp. (TSAI), G. stearothermophilus (GSAI), T. maritima (TMAI), and T. neapolitana (TNAI). The alignment was performed using Clustal X program. Strongly conserved or weakly
conserved residues are shaded dark blue or pink. The putative active residues (E306, E333, H350 and H450) are shaded red and marked as X at the top of the alignment. (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
bioconversion rate represents the ratio between the concentration
of d-tagatose formed and the initial d-galactose concentration.
the gene among L. fermentum strains. However, Its amino acid
sequence exhibits low identity with other compared mesophilic,
thermophilic and hyperthermophilic l-AIs (Fig. 1), demonstrating
that LFAI is a special member of the l-arabinose isomerase super-
family. The protein sequence of LFAI shows 49% amino acid identity
with l-AIs from Alicyclobacillus acidolcaldarius (AAY68209), G.
stearothermophilus (AAD45718) and Thermus sp. (AY225311); 46%
identity with l-AI from Bacillus subtilis (ACT82395); 44% identity
with l-AIs from T. maritima (NP 228089), Thermotoga neapolitana
2.9. Separation of d-tagatose from the conversion mixture
In order to purify the formed d-tagatose by LFAI from the con-
version mixture, an Amberite Ca2 column (35 mm × 550 mm) was
+
◦
used in a water warm system (60 C) to achieve chromatographic
separation. Identification of d-tagatose was performed by 1H NMR
(AY028379) and E. coli (AAA23463). As expected, essential catalytic
spectrometry (Bruker 400-MHz NMR spectrometer, 8% d-tagatose
◦
amino acids E306, E333, E350 and H450 as well as residues Q16,
L18, Y19, Q125, H128, M185, F279, Y335, M351, I373 and H449 that
contribute to substrate recognition and isomerization reaction are
perfectly conserved in LFAI [20].
in DMSO, 25 C).
2.10. Nucleotide sequence accession number
The nucleotide sequence of araA gene from L. fermentum
CGMCC2921 was submitted to GenBank under the accession num-
ber HM150718.
3.2. Expression and purification of the recombinant enzyme
In order to express the l-AI, gene araA was cloned into pET-28a
3
. Results and discussion
and successfully over-expressed in E. coli BL21 (DE3). The solu-
ble recombinant protein reached to over 20% of the total protein
in the cells. A single Ni2+ affinity chromatography step was used
to purify the l-AI to more than 90% purity with a yield of 75%.
SDS–PAGE analysis of the extracts of E. coli BL21 (DE3) harbour-
ing pET-araA induced by IPTG, compared with that of the control
E. coli BL21 (DE3) cells harbouring plasmid pET-28a, revealed the
presence of large amounts of protein around 53 kDa (data not
shown), which was in agreement with the predicted molecular
3.1. Sequence analysis of gene araA
The DNA sequence analysis of gene araA revealed an open read-
ing frame of 1425 bp, encoding a polypeptide of 474 amino acids
with a calculated isoelectric point of pH 4.82 and molecular mass
of 53428 Da. It shows 99% identity with the putative araA gene
from L. fermentum IFO 3956, suggesting a perfect conservation of

4
Z. Xu et al. / Journal of Molecular Catalysis B: Enzymatic 70 (2011) 1–7
mass of the LFAI protein. The molecular masses of l-AIs in other
bacteria were 55 kDa in M. smegmatis [15], 56 kDa in E. coli [1],
5
7 kDa in T. neapolitana [21] and 57 kDa in T. maritima [10]. The
purified enzyme exhibited l-AI activity of 9.98 U/mg at optimal con-
ditions which strongly supported the assumption that the putative
araA gene in L. fermentum CGMCC2921 corresponded to the l-AI
protein.
3.3. Effect of temperature and pH on enzyme activity
◦
The optimum temperature of the purified LFAI was 65 C,
◦
whereas 94% activity was remained at 60 C (Fig. 2A). For indus-
trial production of d-tagatose from d-galactose, isomerization
◦
performed at elevated temperatures (>60 C) offers several advan-
tages, such as higher conversion yield, better sugar solubility and
lower risk of microbial contamination [2]. However, higher tem-
◦
peratures (>70 C) introduce undesired effects such as browning
and unwanted by-products formation [22]. Thus industrial d-
◦
tagatose production is suggested to be carried out at 60–65 C
[
8,23].
Most previously characterized l-AIs have optimum pH in the
range of 7.0–8.5. Nevertheless, a slightly acidic pH range (∼6.0)
can reduce browning and formation of by-products and lower cost
for industrial applications, as observed in the case of acidophilic l-
AIs from Alicyclobacillus acidocaldarius [12] and Lactobacillus sakei
[
24]. Investigation of the effect of pH on LFAI activity showed that
while the enzyme was optimally active at pH 6.5, it exhibited 83%
of its maximum activity at pH 6.0 (Fig. 2B). It should be noticed
that besides L. fermentum CGMCC2921, other l-AIs from Lactobacil-
lus genera also displayed high relative activities at acidic pH, such
as L. plantarum NC8 l-AI (68% relative activity at pH 5.5) [17] and
L. sakei l-AI (80% relative activity at pH 3.0) [24]. From the view
◦
of practical application, LFAI shows high activity at 60–65 C and
pH 6.0, making it a promising candidate for industrial d-tagatose
production (Table 1).
3.4. Effect of divalent metal ions
Fig. 2. Effect of temperaturen (A) and pH (B) on the activity of LFAI. Assays were car-
ried out under standard conditions in the presence of 50 mM d-galactose. Activities
at the optimal temperature and pH were defined as 100%. Each value represents the
mean of triplicate measurements and varied from the mean by not more than 10%.
Filled square: 50 mM sodium acetate buffer; open circle: 50 mM phosphate buffer;
closed triangle: 50 mM Tris–HCl buffer.
In order to investigate the effect of divalent metal ions on LFAI
activity, the purified enzyme was dialyzed against 50 mM phos-
phate buffer (pH 6.5) containing 10 mM EDTA. No activity was
measurable in the absence of divalent metal ions. However, activity
2
+
2+
2+
was recovered when Mn or Co was added. The effect of Mn
and Co2+ concentration on LFAI activity was investigated. As can
be seen in Fig. 3, maximum stimulation of LFAI by Mn and Co
occurred at a concentration of 1 mM and 2 mM, respectively. Com-
2+
2+
The majority of l-AIs are metalloproteins involving metal ions
2
+
2+
2+
pared to Mn and Co , Cu strongly inhibited the d-galactose
for their optimal activity [20]. It was reported that mesophilic
2
+
2+
2+
isomerization, while Zn and Ba slightly enhanced the enzyme
activity. Other ions, such as Mg2+, Ca2+ and Ni had no effects on
LFAI activity (Table 2).
and thermophilic l-AIs require Mn as a cofactor to enhance
2+
the isomerization reaction rate, while the hyperthermophilic l-AIs
require Co2+ [1,2,10,21]. The addition of 1 mM Mn plus 2 mM Co
2+
2+
Table 1
Biochemical properties of LFAI and other reported microbial l-AIs.
◦
Bacterium
Optimum temp. ( C)
Optimum pH
Metal ion requirement
Reference
2
+
Lactobacillus sakei
Alicyclobacillus acidocaldarius
Lactobacillus gayonii
Lactobacillus plantarum SK-2
Mycobacterium smegmatis
Lactobacillus plantarum NC8
Geobacillus stearothermophilus
Thermotoga maritima
30–40
65
30–40
50
45
60
70
90
30
65
5.0–7.0
6.0–6.5
6.0–7.0
7.0
7.0–7.5
7.5
7.0–7.5
7.5
8.0
8.0
8.5
8.5
6.5
Mn , Mg2+
[24]
[12]
[14]
[29]
[15]
[17]
[30]
[10]
[13]
[3]
Mn , Co2+, Mg2+
2
+
2
+
Mn
Mn , Fe3+
2
+
Mn , Co , Mg2+
2
+
2+
2
+
2+
Mn , Co
Mn , Co , Mg2+
2
+
2+
2
+
2+
Mn , Co
2
+
2+
E. coli
Fe , Mn
2
2
2
+
+
+
Thermoanaerobacter mathranii
Thermus sp.
Geobacillus thermodenitrificans
L. fermentum CGMCC2921
Mn
Mn
Mn
60
70
65
[11]
[9]
This study
Mn2+, Co2+

Z. Xu et al. / Journal of Molecular Catalysis B: Enzymatic 70 (2011) 1–7
5
Table 3
Half-life (t1/2, min) of LFAI and other reported l-AIs at different temperatures.
Bacterium
Half-life (t1/2, min)
Reference
◦
Bacillus halodurans
Bacillus licheniformis
20 (70 C)
[30]
[25]
[9]
[30]
[31]
[21]
[10]
This study
◦
120 (50 C)
◦
Geobacillus thermodenitrificans
Geobacillus stearothermophilus
Bacillus stearothermophilus US100
Thermotoga neapolitana
Thermotoga maritima
30.5 (75 C)
◦
52 (80 C)
◦
◦
◦
110 (75 C)
120 (90 C)
185 (90 C)
◦
◦
L. fermentum CGMCC2921
30 (85 C)/220 (75 C)
greatly improved LFAI activity (4.2-fold), suggesting that both these
ions played important roles in the d-galactose isomerization by
LFAI.
3.5. Thermal and pH stabilities
The thermostability of LFAI was proved to be Mn2+ dependent.
Indeed, the enzyme was perfectly stable after a 2 h heating at 75 C
Fig. 3. Effect of Mn and Co2+ addition on LFAI activity. Open circle: concentration
2
+
◦
2+
2+
of Mn ion; closed circle: concentration of Co ion. Activitiy at the optimal concen-
2
+
2+
tration of Mn2+ was defined as 100%. Each value represents the mean of triplicate
measurements.
in the presence of either 1 mM Mn or 1 mM Mn plus 2 mM
2+
Co , since 85% and 83% of its maximum activity were retained,
respectively. On the contrary, in the absence of metal ions or in
the presence of only 2 mM Co2+, the enzyme was completely inac-
tivated after 60 min. This result suggests the involvement of Mn2+
ion in the enzyme stabilization at high temperatures besides its role
in the catalytic mechanism, whereas Co2+ seems to be essentially
implicated in the isomerization reaction. Compared to previously
reported l-AIs, LFAI showed a preferable thermostability in the
Table 2
Effect of different metal ions on the activity of LFAI.
Metal ion (1 mM)
Specific activity (U/mg protein)
Relative activity (%)
Nonea
EDTA
2.0
0
100
0
MgCl2
MnCl2
CoCl2
ZnCl2
CaCl2
2.2
6.0
4.0
2.6
2.1
0.4
2.2
2.4
8.5
109
298
201
129
103
20
107
120
420
2+
◦
presence of Mn , with a half-life time of 30 min at 80 C and
◦
2
20 min at 75 C (Table 3).
LFAI was stable at acidic pH since it retained 88% and 80% of its
original activity after 24 h of incubation at pH 6.0 and 5.5, respec-
tively. At pH 5.0, 55% of its activity was remained. In comparison, L.
plantarum NC8 l-AI remained 89% of its activity after 24 h of incu-
bation at pH 5.0 [17], and L. sakei l-AI had a half-life time of its
CuCl2
NiCl2
BaCl2
MnCl2 + CoCl2b
◦
The activity of EDTA-treated LFAI was assayed in the standard assay condition after
incubating of 1 mM various metal ions. Each value represents the mean of triplicate
measurements and varied from the mean by not more than 10%.
activity of 49 h at pH 5.0 and 47 h at pH 6.0 (under 35 C) [24].
a
3.6. Substrate specificity
The activity of native enzyme without EDTA treatment and metal ions addition
was set as 100%.
b
2+
2+
Mn and Co were added to the reaction mixture at 1 mM and 2 mM, respec-
The characterization of LFAI as an l-AI then allowed for the
investigation of its substrate specificity for various aldoses. LFAI had
a high preference for l-arabinose (220%) and d-galactose (relative
activity: 100%). Other aldoses, such as d-xylose (2.7%), d-mannose
tively.
(
2.5%), l-xylose (1.9%), d-glucose (1.7%), and l-ribose (0.7%) did
2+
2+
not serve as substrates for LFAI in the presence of Mn or Co
.
It was previously reported that l-AI from Bacillus licheniformis
ATCC14580 showed 2% enzyme activity for d-galactose compared
with l-arabinose [25], and B. subtilis str. 168 l-AI displayed sub-
strate specificity only towards l-arabinose [26]. Different from LFAI,
these l-AIs were ideal choice for enzymatic synthesis of l-ribulose
from l-arabinose.
3.7. Kinetic parameters determination
Values of kinetic constants were determined on the basis of the
Lineweaver–Burk plots. The Km was 29.9 mM for l-arabinose and
0.2 mM for d-galactose. Besides, the catalytic efficiency (kcat/Km)
6
−
1
−1
−1
−1
,
and Vmax was 19 mM min , 24.3 U/mg and 9.02 mM min
.8 U/mg for l-arabinose and d-galactose, respectively. Therefore,
9
the catalytic efficiency of LFAI increased 2.1-fold using l-arabinose
as a substrate compared with d-galactose. The LFAI catalyzes
the isomerization of d-galactose with a relatively high catalytic
efficiency (Table 4), showing a high substrate affinity towards
d-galactose, which makes it potential for d-tagatose production.
Fig. 4. Time course of d-tagatose production during LFAI-catalyzed isomerization of
◦
d-galactose. Closed circle: conversion curve at 65 C; open triangle: conversion curve
◦
◦
at 60 C; open square: conversion curve at 70 C.

6
Z. Xu et al. / Journal of Molecular Catalysis B: Enzymatic 70 (2011) 1–7
Table 4
Comparison of l-arabinose isomerase kinetic constants from various microbial origins.
Bacterium
aVmax (U/mg)
aKm (mM)
akcat/Km (mM−1 min−1
)
bkcat/Km (mM−1 min−1
)
Reference
Bacillus halodurans
1.3
6.9
7.0
37.6
NR
14.3
7.5
7.8
8.9
8.9
4.9
76
167
408
69.7
578
339
250
129
145
60
57
28.9
59
60
0.4
0.5
1.6
2.1
3.1
3.2
3.3
1.2
8.5
8.5
9.3
10.3
9.0
51.4
48
15.5
65
[30]
[9]
Geobacillus thermodenitrificans
Lactobacillus plantarum NC8
Geobacillus stearothermophilus (mutant enzyme)
Geobacillus thermodenitrificans (mutant enzyme)
Thermotoga neapolitana
Alicyclobacillus acidocaldarius
Geobacillus stearothermophilus
Thermotoga maritima
[17]
[32]
[33]
[21]
[12]
[30]
[10]
[31]
[34]
[24]
This study
136
58.1
41.5
61
74.8
71
NR
64.8
19
Bacillus stearothermophilus US100
Acidothermus cellulolytics
Lactobacillus sakei
L. fermentum CGMCC2921
9.8
NR: not reported.
a
Vmax, Km and kcat/Km for d-galactose.
kcat/Km for l-arabinose.
b
d-tagatose production might be difficult because they require Co2+
ion as a cofactor and cobalt cannot be used in nutritional applica-
tions [2,6,10].
Structure and mechanistic studies, to clarify the reason for sub-
strate choosing of LFAI, are presently under way.
3.8. d-Tagatose production by LFAI
3.9. Purification and identification of d-tagatose
The study of isomerization of d-galactose (50 mM) to d-tagatose
by LFAI at different temperatures at pH 6.5 demonstrated that the
Hong et al. reported a method for isolating d-tagatose (ketose)
ratio of conversion of d-galactose to d-tagatose after 12 h was 52
and 36% at 60 C and 70 C, respectively. The highest amount of
from mixtures with d-galactose (aldose), instead of employing
chemicals and organic solvents, ion-exchange chromatography
was utilized and d-tagatose with high purity was obtained [28].
We use Amberite column with a water solvent system to sepa-
rate d-tagatose so as to prevent environmental disadvantages. A
total of 20 mL reaction mixture was applied to the column at a
flow rate of 2 ml/min, then eluted by deionized water. Fractions
containing pure d-tagatose (confirmed by HPLC) were pooled and
concentrated by evaporation to dryness. The structure of puri-
fied d-tagatose was confirmed by 1H NMR spectrometry, 1H NMR
◦
◦
◦
2+
bioconversion was 55% at 65 C with 1 mM Mn (Fig. 4). The pro-
duction of d-tagatose from d-galactose was further proved by HPLC
analysis, and no by-products were observed (Fig. 5). The commer-
cial process using xylose isomerase as an enzyme similar to l-AI is
◦
carried out at around 60 C to limit color formation [27]. At this
temperature, thermophilic l-AIs exhibit higher conversion yield
than that of hyperthermophilic l-AIs [3,11,21]. Although hyper-
thermophilic l-AIs are more thermostable, their use in commercial
Fig. 5. HPLC analysis of the d-tagatose production. (A) d-Tagatose standard; (B) d-galactose standard; (C) products of isomerization by LFAI, the retention times of d-galactose
and d-tagatose were 14.76 and 18.56 min, respectively.

Z. Xu et al. / Journal of Molecular Catalysis B: Enzymatic 70 (2011) 1–7
7
(
2
5
400 MHz, DMSO) ı 3.24 (m, 1H), 3.28 (d, 1H), 3.36 (t, 1H), 3.44 (d,
H), 3.53 (s, 2H), 4.32 (d, 1H), 4.43 (d, 1H), 4.45 (t, 1H), 4.60 (d, 1H),
.33 (s, 1H).
[4] T.W. Donner, J.F. Wilber, D. Ostrowski, Diabetes Obes. Metab. 1 (1999) 285–291.
[
5] B. Buemann, S. Toubro, A. Raben, J. Blundell, A. Astrup, Br. J. Nutr. 84 (2000)
27–231.
2
[
[
[
[
6] D.K. Oh, Appl. Microbiol. Biotechnol. 76 (2007) 1–8.
7] G.V. Levin, J. Med. Food 5 (2002) 23–26.
8] H.J. Kim, S.A. Ryu, P. Kim, D.K. Oh, Biotechnol. Prog. 19 (2003) 400–404.
9] H.J. Kim, D.K. Oh, J. Biotechnol. 120 (2005) 162–173.
4
. Conclusion
[
[
[
10] D.W. Lee, H.J. Jang, E.A. Choe, B.C. Kim, S.J. Lee, S.B. Kim, Y.H. Hong, Y.R. Pyun,
Appl. Environ. Microbiol. 70 (2004) 1397–1404.
11] J.W. Kim, Y.W. Kim, H.J. Roh, H.Y. Kim, J.H. Cha, K.H. Park, C.S. Park, Biotechnol.
Lett. 25 (2003) 963–967.
12] S.J. Lee, D.W. Lee, E.A. Choe, Y.H. Hong, S.B. Kim, B.C. Kim, Y.R. Pyun, Appl.
Environ. Microbiol. 71 (2005) 7888–7896.
In summary, we have successfully cloned the araA gene from
strain L. fermentum CGMCC2921 and expressed as a recombinant
protein in E. coli. Compared with other l-AIs, LFAI exhibits not only
preferable thermostability at higher temperatures with Mn2 , but
also behaves relatively high activity and stability at acidic pH. The
successful identification and over-expression of the LFAI allows us
to characterize a novel l-AI showing high specificity towards d-
galactose and now sets the stage for more detailed investigation
of this enzyme. In addition, a feasible and environmental friendly
method for d-tagatose purification has been established. This work
will be of great value to both the efficient expression and the large
scale production of d-tagatose with E. coli as a host cell.
+
[13] S.H. Yoon, P. Kim, D.K. Oh, World J. Microbiol. Biotechnol. 19 (2003) 47–51.
[
[
14] T. Nakamatu, K. Yamanaka, Biochim. Biophys. Acta 178 (1969) 156–165.
15] K. Izumori, Y. Ueda, K. Yamanaka, J. Bacteriol. 133 (1978) 413–414.
[16] P. Kim, S.H. Yoon, M.J. Seo, D.K. Oh, J.H. Choi, Biotechnol. Appl. Biochem. 34
(2001) 99–102.
[
17] H. Chouayekh, W. Bejar, M. Rhimi, K. Jelleli, M. Mseddi, S. Bejar, FEMS Microbiol.
Lett. 277 (2007) 260–267.
[
18] M.M. Bradford, Anal. Biochem. 72 (1976) 248–254.
[19] Z. Dische, E. Borenfreund, J. Biol. Chem. 192 (1951) 583–587.
[
[
20] B.A. Manjasetty, M.R. Chance, J. Mol. Biol. 360 (2006) 297–309.
21] B.C. Kim, Y.H. Lee, H.S. Lee, D.W. Lee, E.A. Choe, Y.R. Pyun, FEMS Microbiol. Lett.
212 (2002) 121–126.
Acknowledgements
[22] M. Rhimi, N. Aghajari, M. Juy, H. Chouayekh, E. Maguin, R. Haser, S. Bejar,
Biochimie 91 (2009) 650–653.
[
[
23] D.K. Oh, H.J. Kim, S.A. Ryu, P. Kim, Biotechnol. Lett. 23 (2001) 1859–1862.
24] M. Rhimi, R. Ilhammami, G. Bajic, S. Boudebbouze, E. Maguin, R. Haser, N. Agha-
jari, Bioresour. Technol. 101 (2010) 9171–9177.
25] P. Prabhu, M.K. Tiwari, M. Jeya, P. Gunasekaran, I.W. Kim, J.K. Lee, Appl. Micro-
biol. Biotechnol. 81 (2008) 283–290.
26] J.H. Kim, P. Prabhu, M. Jeya, M.K. Tiwari, H.J. Moon, R.K. Singh, J.K. Lee, Appl.
Microbiol. Biotechnol. 85 (2010) 1839–1847.
27] B.S. Hartley, N. Hanlon, R.J. Jackson, M. Rangarajan, Biochim. Biophys. Acta 1543
This work was supported by the National Basic Research Pro-
gram of China (973) (2007CB714304), the National Nature Science
Foundation of China (20906050), the Natural Science Founda-
tion of the Jiangsu Higher Education Institutions of China (No.
[
[
[
[
0
(
8KJA180001), the Natural Science Foundation of Jiangsu Province
BK2009357) and the Key Projects in the National Science & Tech-
nology Pillar Program during the Eleventh Five-Year Plan Period
2008BAI63B07).
(
2000) 294–335.
28] Y.H. Hong, D.W. Lee, S.J. Lee, E.A. Choe, S.B. Kim, Y.H. Lee, C.I. Cheigh, Y.R. Pyun,
Biotechnol. Lett. 29 (2007) 569–574.
29] H. Zhang, B. Jiang, B. Pan, World J. Microbiol. Biotechnol. 23 (2007) 641–646.
30] D.W. Lee, E.A. Choe, S.B. Kim, S.H. Eom, Y.H. Hong, S.J. Lee, H.S. Lee, D.Y. Lee, Y.R.
Pyun, Arch. Biochem. Biophys. 434 (2005) 333–343.
(
[
[
References
[
31] M. Rhimi, S. Bejar, Biochim. Biophys. Acta 1760 (2006) 191–199.
[
[
[
1] J.W. Patrick, N. Lee, J. Biol. Chem. 243 (1968) 4312–4318.
2] P. Kim, Appl. Microbiol. Biotechnol. 65 (2004) 243–249.
3] F. Jorgensen, O.C. Hansen, P. Stougaard, Appl. Microbiol. Biotechnol. 64 (2004)
[32] H.J. Kim, J.H. Kim, H.J. Oh, D.K. Oh, J. Appl. Microbiol. 101 (2006) 213–221.
[33] H.J. Oh, H.J. Kim, D.K. Oh, Biotechnol. Lett. 28 (2006) 145–149.
[34] L. Cheng, W. Mu, T. Zhang, B. Jiang, Appl. Microbiol. Biotechnol. 86 (2010)
1089–1097.
8
16–822.