Biochemical characterization of a novel glucose isomerase from Anoxybacillus gonensis G2T that displays a high level of activity and thermal stability

Article abstract of DOI:10.1016/j.molcatb.2013.08.019

In the continuing search for novel enzymes suitable for the production of high fructose corn syrup (HFCS), a new glucose isomerase (GI) from the thermophile Anoxybacillus gonensis G2^T is described. The gene encoding this GI (AgoG2GI) was cloned and then engineered for heterologous expression in Escherichia coli. The recombinant enzyme was purified from the heat treated cell-free extract by anion exchange chromatography followed by hydrophobic interaction chromatography. The purified enzyme showed optimal activity at 85 C and pH 6.5. The steady state parameters of K_m and k_cat with d-glucose were found to be 146.08 ± 9.50 mM and 36.47 ± 2.01 (1/s), respectively. l-arabinose, d-ribose and d-mannose also served as substrates for the enzyme with comparable kinetic parameters. AgoG2GI requires the divalent cations of Co²⁺, Mn²⁺ and Mg²⁺ for its maximal activity and thermostability. The results reported here are indicative of a new GI with desirable kinetics and stability parameters for the efficient production of HFCS at industrial scale.

Full text of DOI:10.1016/j.molcatb.2013.08.019

Journal of Molecular Catalysis B: Enzymatic 97 (2013) 215–224
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Biochemical characterization of a novel glucose isomerase from
Anoxybacillus gonensis G2T that displays a high level of activity and
thermal stability
Hakan Karaoglua, Derya Yanmisb, Fulya Ay Salc, Ayhan Celikd, Sabriye Canakcic,
Ali Osman Belduzc,∗
a Department of Basic Sciences, Faculty of Fisheries and Aquatic Sciences, Recep Tayyip Erdogan University, 53100 Rize, Turkey
b Department of Biology, Faculty of Sciences, Ataturk University, 25240 Erzurum, Turkey
c Department of Biology, Faculty of Sciences, Karadeniz Technical University, 61080 Trabzon, Turkey
d Department of Chemistry, Gebze Institute of Technology, 41400 Kocaeli, Turkey
a r t i c l e i n f o
a b s t r a c t
Article history:
In the continuing search for novel enzymes suitable for the production of high fructose corn syrup
(HFCS), a new glucose isomerase (GI) from the thermophile Anoxybacillus gonensis G2T is described.
The gene encoding this GI (AgoG2GI) was cloned and then engineered for heterologous expression in
Escherichia coli. The recombinant enzyme was purified from the heat treated cell-free extract by anion
exchange chromatography followed by hydrophobic interaction chromatography. The purified enzyme
showed optimal activity at 85 ◦C and pH 6.5. The steady state parameters of Km and kcat with d-glucose
were found to be 146.08 9.50 mM and 36.47 2.01 (1/s), respectively. l-arabinose, d-ribose and d-
mannose also served as substrates for the enzyme with comparable kinetic parameters. AgoG2GI requires
the divalent cations of Co2+, Mn2+ and Mg2+ for its maximal activity and thermostability. The results
reported here are indicative of a new GI with desirable kinetics and stability parameters for the efficient
production of HFCS at industrial scale.
Received 15 February 2013
Received in revised form 26 August 2013
Accepted 27 August 2013
Available online 5 September 2013
Keywords:
Anoxybacillus gonensis
Thermophile
XylA
Xylose isomerase
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
required fructose syrup concentration of 55%. It has been theorized
that isomerization at elevated temperature (at about 95 ◦C) and
Xylose isomerase (d-xylose ketol-isomerase E.C 5.3.1.5) cat-
alyzes the isomerization of d-xylose to xylulose on the first step of
xylose metabolism in many microorganisms [1]. Interconversion
between xylose and xylulose provides a nutritional requirement
in saprophytic bacteria that flourish on rotting plant matter and
also assists in the bioconversion of hemicellulose to ethanol [2].
Xylose isomerase is also known as glucose isomerase (GI) due to its
involvement in the reversible isomerization of glucose to its respec-
tive ketose (fructose) in the presence of divalent metal ions (Fig. 1).
This isomerization reaction is important in industrial processes
such as the production of high fructose corn syrup (HFCS) in the
food industry [3,4]. In the existing processes for HFCS production,
non-thermostable xylose (glucose) isomerases from mesophilic
microorganisms are utilized in an immobilized enzyme reactor pro-
ducing 40% to 42% fructose syrup at an operating temperature of
50 ◦C and 58 ◦C with residencetimes of <1 h [5]. In this process, addi-
tional expensive chromatographic steps are necessary to obtain the
lower pH (at about 4.5–5.5) are required to achieve the desired
fructose level in the syrup due to the isomerization reaction equi-
librium [6]. If this could be achieved, additional concentration steps
would be unnecessary [7]. For this reason, researchers have sought
to isolate the microorganisms that are capable of producing GI with
high levels of activity and stability at elevated temperature. In this
context, several genes encoding these enzymes have been reported
from Escherichia coli [8], Bacillus subtilis [9], Clostridium species [10],
Streptomyces species [11], Ampullariella species [12], Actinoplanes
missouriensis [13] and Thermus thermophilus [14]. Furthermore, GI is
also an ideal enzyme for studying structure–function relationships
from an academic perspective [2].
In brief, an industrially valuable GI must have a lower pH opti-
mum, a higher temperature optimum, a resistance to inhibition by
Ca2+, and an elevated affinity for glucose over currently employed
enzymes. Therefore, there is currently significant research effort
devoted to identifying a thermostable and acid stable GI with
higher affinity for glucose. What is more, advances in recombi-
nant DNA technology and protein engineering have opened new
and encouraging possibilities for development of an economically
viable commercial process.
∗
Corresponding author. Tel.: +90 462 377 2522; fax: +90 462 325 3195.
E-mail addresses: belduz@ktu.edu.tr, biyoloji2008@hotmail.com (A.O. Belduz).
1381-1177/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcatb.2013.08.019

216
H. Karaoglu et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 215–224
Fig. 1. Reactions catalyzed by GI.
Further in this direction, we report here, for the first time,
2.4. Cloning of the xylA gene
a novel glucose isomerase (named as AgoG2GI) from the ther-
mophile Anoxybacillus gonensis G2T, a xylanolytic, sporulating,
Gram-positive, rod-shaped, facultative anaerobe and moderately
thermophilic bacterium growing naturally at 55 ◦C to 60 ◦C in the
thermal spring at Balıkesir, Turkey [15]. This study describes the
biochemical characterization of AgoG2GI together with its kinetic
parameters with various substrates. The potential application of
AgoG2GI in the food industry is also discussed. We believe that the
identification of a thermophilic GI along with recently developed
recombinant systems for heterologous gene expression, with the
possibility for further enhancement of enzyme activity by protein
engineering, will be of interest to both industry and academia.
PCR was used to amplify a portion of the A. gonensis G2T xylA
gene with degenerate primers (Table 1). Degenerate primers F1, F2,
R1, R2 and R3 were designed based on the conserved sequence of
the other xylose isomerases. Genomic DNA was purified using the
Wizard Genomic DNA Purification Kit (Promega, Madison, WI, USA)
according to the manufacturer’s instructions. Taq DNA polymerase
was used to perform PCR with A. gonensis G2T genomic DNA as
the template. The PCR fragment was subsequently cloned into a
pGEM-T vector system I and sequenced by Macrogen (Seoul, Korea).
Sequence analysis revealed the PCR produced a 534 bp fragment of
the xyl gene.
Inverse PCR was performed to clone the complete xyl gene
according to the methods described by Trigilia et al. [17]. Two and
a half micrograms of genomic DNA were separately digested with
restriction enzymes (NcoI, MboI, HinfI, AvaII, BamHI, HindIII, SacI,
PstI, Sa3AI, SalI, PspGI, AvaI, BfaI, EcoRI, or XbaI), which did not
digest the cloned 534 fragment. Digestions were performed accord-
ing to the manufacturer’s instructions overnight in a final volume
of 50 ␮l. The enzymes were heat-inactivated at 70 ◦C for 10 min.
The mixtures were then self-ligated overnight at 16 ◦C with 1 U/␮l
T4 DNA ligase in a total volume of 400 ␮l. The ligation mixture was
ethanol precipitated. The primers for inverse PCR (Table 1) were
designed according to the partial sequence encoding xylA gene and
to the inverse PCR fragments for the next inverse PCR. The PCR
products were checked by agarose gel electrophoresis and cloned
into pGEM-T easy vector and then sequenced as described above.
The xylA gene was amplified directly from genomic DNA of
A. gonensis G2T using a pair of primers (Xyla Ex F1-Xyla Ex R1
or Xyla Ex F1-Xyla Ex R2) designed according to the sequence of
xylose isomerase from A. gonensis G2T. The restriction sites NcoI and
BamHI or HindIII (for HisTag) were incorporated into the forward
and reverse primer sequence, respectively. Taq DNA polymerase
was used to perform PCR with A. gonensis G2T genomic DNA as the
template.
2. Materials and methods
2.1. Substrates and chemicals
The chemicals were purchased commercially from Merck A.G.
(Darmstadt, Germany), Sigma Chem. Co. (St. Louis, MO, USA),
Fluka Chemie A.G. (Buchs, Switzerland), Acumedia Manufacturers,
Inc. (Baltimore, Maryland, USA), and Aldrich-Chemie (Steinheim,
Germany). The Wizard Genomic DNA Purification Kit, Wizard Plus
SV Minipreps DNA Purification System, MagneHis Protein Purifica-
tion System, Taq DNA Polymerase, dNTP, and all of the restriction
enzymes were purchased from Promega Corp. (Madison, WI, USA).
All chemicals were reagent grade, and all solutions were made with
distilled and deionized water.
2.2. Strains, vectors, and media
A. gonensis G2T NCIMB 13933T, E. coli BL21 (DE3):pLysS was
purchased from Novagen (Madison, WI, USA) and XL1-Blue from
Stratagene (La Jolla, CA, USA). The plasmid vectors used were
pGEMT-Easy (Promega) and pET28 (a-c)+ (Novagen). All E. coli
strains containing recombinant plasmids were cultured in Luria
broth (LB) medium supplemented with 50 ␮g/ml ampicillin or
kanamycin, as appropriate, at 37 ◦C and pH 7.4, unless otherwise
stated.
The 1326 bp fragment was cloned into pET28(a-c)+ to generate
pAgoG2GI or pAgoG2GI-his. The recombinant plasmids were then
used to transform the expression strain E. coli BL21 (DE3):pLysS.
Molecular Evolutionary Genetics Analysis version 5.05 (MEGA
5.05) was used for multiple sequence alignments and for construc-
tion of a phylogenetic tree using the neighbor-joining method.
2.3. Plate assays for detection of thermostable glucose isomerase
in G2T
2.5. Overexpression of the enzyme
Plate assay was carried out according to Lee et al. [16]. The plates
were incubated for 5 to 6 days at 55 ◦C for A. gonensis G2T. The same
plates containing 1% xylose were also used of xylose utilization
for E. coli xyl-5 mutants (E. coli HB101) at 37 ◦C. Glucose isomer-
ase activities were determined as a dark brown halo around the
colonies.
E. coli BL21 cells harboring pAgoG2GI or pAgoG2GI-his were
grown to an optical density (OD) at 600 nm of about 0.6. Het-
erologous gene expression was then induced by addition of 1 mM
iso-propyl-␤-d-thiogalactopyranoside (IPTG) and the culture was
continued for a further 4 h. Cells were harvested by centrifugation

H. Karaoglu et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 215–224
217
Table 1
Degenerate, inverse-PCR and expression primers which successfully amplified xylA gene of A. gonensis G2T.
Primers
Sequence
Remarks
F1 (degenerate primer)
F2 (degenerate primer)
R1 (degenerate primer)
R2 (degenerate primer)
R3 (degenerate primer)
XylaIF01
5ꢀ-gTVYTBTggggYggVMgHgARgg-3ꢀ
5ꢀ-ATHgARCCNAARCCNAWRgARCC-3ꢀ
5ꢀ-ggRAAYTCRTCBgTRTCCCAKCC-3ꢀ
5ꢀ-NgCRTCRAARTTBANNCCDCC-3ꢀ
5ꢀ-SYRTCCATBSMDSCDAYRTg-3ꢀ
5ꢀ-AgACgCggTTCgTTTgAACC-3ꢀ
5ꢀ-ACgTTTCATAgCCTTCACg-3ꢀ
For partial cloning of xylA gene
Inverse-PCR
XylaIF02
XylaIF03
XylaIF04
XylaIF05
XylaIR01
XylaIR02
XylaIR03
XylaIR04
XylaIR05
5ꢀ-AgCAgCAAAggTgAAAAAAgg-3ꢀ
5ꢀ-TggACTAAAAgTggCgTATCg-3ꢀ
5ꢀ-ATCATTgATgggAAggCCgAC-3ꢀ
5ꢀ-gCATAgTCAACCgCCATATgC-3ꢀ
5ꢀ-AggCggATTgAACTTTgATgC-3ꢀ
5ꢀ-TgCCARTAVgCRAYVgMRADDCg-3ꢀ
5ꢀ-TCCAAAgCAACTTggTTTTgC-3ꢀ
5ꢀ-TgCACCATgTACAAAACgAgg-3ꢀ
5ꢀ-gCgCAAATgCTCCTCCATCg-3ꢀ
5ꢀ-ggCCAATTCCTTCTgTATAgC-3ꢀ
5ꢀ-CgAgCTCCATggCgTATTTTgAAAACg-3ꢀ
XylaIR06
XylaIR07
Xyla Ex F1
For cloning into pET28(a)+
NcoI
Xyla Ex R1
Xyla Ex R2
5ꢀ-CCAAgCTTACgAgCTACACAAACTTC-3ꢀ
HindIII
5ꢀ-CggATCCgTTACTATCATTAACgAgC-3ꢀ
BamHI
at 11,000 rpm for 5 min. The cells were disrupted using a Sartorius
Labsonic M sonicator at 0.6 cycle scale (80% amplitude) to release
intracellular proteins. The cell-free extract was clarified by centrifu-
gation at 14,800 rpm for 15 min to remove cell debris, and assayed
for glucose isomerase activity.
AgoG2GI-recHis, respectively. The protein, which contained a His
tag (AgoG2GI-recHis), was purified using the MagneHisTM Protein
Purification System (Promega).
Protein concentration was determined using the method of
Bradford [18]. Bovine serum albumin was used as a standard in
this procedure.
A. gonensis G2T for expression of glucose isomerase was induced
with 0.5% final d-xylose concentration at OD 0.6 (600 nm). The
expressed enzyme from type stain G2 was designated as AgoG2GI-
wt (wild-type enzyme), whereas the enzyme expressed from E. coli
BL21/pAgoG2GI or pAgoG2GI-his was designated as AgoG2GI-rec or
AgoG2GI-recHis.
2.7. Activity assay of GI
Cell extracts prepared by sonication and purified preparations
were used as enzyme sources. Glucose isomerase activity was
measured by incubating a reaction mixture that contained 10 mM
MgSO4, 1 mM CoCl2, 0.2 M glucose and the enzyme in 50 mM MOPS
buffer (pH 6.5) at 85 ◦C for 30 min in 100 ␮l reaction volume. The
reaction was stopped with the addition of 100 ␮l 0.5 M perchlo-
ric acid, 40 ␮l of 1.5% cysteine hydrochloride, and 40 ␮l of 0.12%
carbosol, respectively. After 1.2 ml of 70% sulfuric acid was added,
the mixture was vortexed and then incubated at room tempera-
ture for 30 min. Absorbance was measured at 560 nm for fructose.
The amount of fructose formed after the enzyme reaction was esti-
mated using the cysteine–carbazole–sulfuric acid method [19]. One
unit of activity was defined as the amount of enzyme that released
1 ␮mol of fructose/minute under the assay conditions described
above.
2.6. Purification of the enzymes
The purification procedure was performed at room tempera-
ture. A crude extract of AgoG2GI-wt (30 ml) was heated at 75 ◦C
for 15 min and precipitated proteins were removed. In the first
step of purification of AgoG2GI-wt, ammonium sulfate precipi-
tation was carried out as follows: solid (NH4)2SO4 was added
to the heat-treated extracts to give 60% saturation, and the pre-
cipitate was removed by centrifugation (12,500 rpm for 30 min).
(NH4)2S04 was added further to the supernatant to give 70% satura-
tion. This precipitate was collected and dialyzed overnight against
the same buffer. In the subsequent purification step, the dialyzed
enzyme preparation was loaded onto a diethylaminoethyl (DEAE)-
Sepharose anion-exchange column. After loading the sample the
column was extensively washed to remove unbound material.
Bound proteins were then eluted on a linear salt gradient from
0 M to 0.6 M NaCl in 20 mM MOPS buffer (pH 6.5) at a flow rate
of 1 ml/min. Fractions containing AgoG2GI-wt were pooled, con-
centrated by ultrafiltration (Sartorious, 10,000 MWCO filters) and
loaded onto a Phenyl-Sepharose-6 Fast Flow hydrophobic inter-
action column (Sigma) equilibrated in buffer containing 1.3 M
(NH4)SO4. Proteins were eluted on a linear gradient from 1.3 M to
0 M (NH4)SO4 at a flow rate of 0.5 ml/min.
2.8. Determination of Km and Vmax values
The kinetic parameters Vmax (␮mol/min/mg), Km (Michaelis
constant, mM) and kcat (1/s) were determined from
Michaelis–Menten plots of specific activities at various substrate
(d-glucose, l-arabinose, d-mannose, d-ribose) concentrations
using the OriginPro8.1 program (OriginLab Data Analysis and
Graphing Software). Typically, duplicate measurements at 6 to 10
different concentrations of substrate spanning 0–700 mM were
used to determine the value of Km
.
AgoG2GI-rec (crude extract 13 ml) was purified according to the
procedures described above, except for the ammonium sulfate pre-
cipitation.
The enzyme purified directly from A. gonensis was desig-
nated as AgoG2GI-wt while the enzyme with and without HisTag
from pET28(a-c)+ in E. coli designated as AgoG2GI-recHis and
2.9. Determination of submolecular mass by SDS-PAGE
SDS-PAGE was performed as described by Maniatis et al. [20].
Subunit molecular mass values were estimated by SDS-PAGE using
marker proteins (Promega).

218
H. Karaoglu et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 215–224
Fig. 2. Phylogenetic analysis of AgoG2GI and other amino acid sequences of the tested XIs reported in the GenBank database. The phylogenetic tree was constructed using
MEGA 5.05 software with the neighbor-joining method. Bar, 0.1 substitutions per nucleotide.
2.10. Determination of submolecular mass by MALDI-TOF
The percentage of residual activity of glucose isomerase was calcu-
lated by comparison with untreated enzyme.
Mass spectrometry of AgoG2GI-rec was performed by MALDI-
TOF using a MicroFlex LT (Bruker). Protein sample was diluted
in 0.1% formic acid to approximately 0.05 mg/ml and mixed with
an equal volume of matrix (saturated solution of sinapinic acid
in 50% acetonitrile, 0.1% formic acid) on a stainless steel surface.
The samples were air dried at room temperature to crystallize. The
machine was operated in positive ion mode and calibrated with
bovine serum albumin.
The following buffers (200 mM) were used: sodium acetate (pH
5.0–6.0), potassium phosphate (pH 6.0–7.0), tris–HCl (pH 7.0–9.0),
and glycine–NaOH buffer (pH 9.0–10.0).
2.14. Effect of metal ions
It is reported that Co2+, Mn2+, and Mg2+ bivalent metal ions are
activators of glucose isomerases [2]. The activator effects of var-
ious metal ions on the glucose isomerase activity of AgoG2GI-wt
and AgoG2GI-rec was assayed at optimum reaction conditions. To
determine the effect of these metal ions on glucose isomerase activ-
ity, metal ions were removed from the purified glucose isomerase
by treatment with 5 mM EDTA at 60 ◦C for 1 h followed by dial-
ysis against 50 mM MOPS buffer (pH 6.5) overnight with several
changes of buffer [21]. MOPS buffer containing 0.1 mM, 0.5 mM,
1 mM, 2 mM, 4 mM, 10 mM, 20 mM, 50 mM and 100 mM of biva-
lent metal ions such as Co2+, Mn2+, and Mg2+ chloride salts were
added separately to the dialyzed enzyme preparation and assayed
glucose isomerase activity as described above. The glucose isomer-
ase activity of the enzyme without metal ion was defined as 100%.
The residual activity (%) was assayed spectrophotometrically.
The inhibitor effects of various metal ions on the glucose isom-
erase activity of AgoG2GI-wt and AgoG2GI-rec were assayed at
2.11. Determination of molecular mass with HPLC
The apparent molecular mass of the native enzyme was deter-
mined by high performance liquid chromatography (HPLC) under
the following conditions: column, SuperdexTM 200 10/300 GL Gel
filtration column (Amersham Pharmacia Biotech); mobile phase,
50 mM phosphate buffer; flow rate, 0.5 ml/min. The eluted com-
pound was detected by UV index. Carbonic anhydrase (29 kDa),
bovine serum albumin (66 kDa), alcohol dehydrogenase (150 kDa),
␤-amylase (200 kDa), and apoferritin (443 kDa) were used as the
molecular mass marker proteins.
2.12. Effects of temperature on activity and stability
optimum reaction conditions. The metal ions Cd2+, Ca2+, Hg2+, Ni2+
,
The effects of temperature on AgoG2GI-wt and AgoG2GI-rec
activities were determined spectrophotometrically using d-glucose
as the substrate. The enzymatic reactions at various temperatures
over the range of 25 ◦C to 100 ◦C were performed using the enzyme
activity assay described previously, and the results were expressed
as relative activity (%) obtained at optimum temperature.
The effect of temperature on AgoG2GI-wt and AgoG2GI-rec
stability was determined by measuring the residual activity (%)
after pre-incubation in 50 mM MOPS (pH 6.5) at 4 ◦C, 30 ◦C and
85 ◦C. Samples of assay mixtures were taken at intervals and resid-
ual enzyme activity was determined. The percentage of residual
activity of glucose isomerase was calculated by comparison with
untreated enzyme.
Zn2+, Fe2+ and Cu2+ were added as chloride or sulfate salts to the
enzyme solution in amounts of 0.1 mM, 1 mM, 5 mM, and 10 mM.
The glucose isomerase activity of the enzyme without metal ion
was defined as the 100% level. The residual activity (%) was assayed
spectrophotometrically.
The nucleotide sequence has been submitted to the Gen-
Bank/EMBL Data Bank with accession number JQ768452.
3. Results
3.1. Detection of thermostable glucose isomerase in
Anoxybacillus gonensis G2T
The presence of glucose isomerase activity in the A. gonensis G2T
and E. coli HB101 strains was studied by plate assay. A. gonensis G2T
showed a dark brown halo around the light colonies, indicating
glucose isomerase activity. The E. coli HB101 strain had no activity.
2.13. Effects of pH on activity and stability
The optimum pH of the enzyme was measured at 85 ◦C and
560 nm by using buffer solutions of different pH values. Their rela-
tive activities (%) were measured, and the results were expressed as
relative activity (%) obtained at pH. Other parameters were deter-
mined in optimum pH conditions.
3.2. Cloning of AgoG2 gene
To determine the stability of the enzyme at pH values between
5.0 and 10.5, pre-incubation was performed at each pH value at
4 ◦C for 300 h and at 85 ◦C for 8 h. Samples of assay mixtures were
taken at intervals, and residual enzyme activity was determined.
In this work, using two degenerate primers, we obtained a
534 bp fragment of the XylA gene; the remaining portion of the
gene sequence was subsequently obtained by using inverse PCRs.
The whole gene was analyzed by using the Advanced Blast Program

H. Karaoglu et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 215–224
219
Table 2
Summary of purification steps: (a) AgoG2GI-wt (b) AgoG2GI-rec.
(a)
Purification step
Volume (ml)
[Protein] (mg/ml)
Total protein (mg)
Total activity (U)
Specific activity (U/mg)
Yield (%)
Fold
Cell extract
Heat treatment
(NH4)2SO4 sedimentation
DEAE-Sepharose
Phenyl-Sepharose 6 Fast flow
30
24
19
45
17
11.53
2.50
1.80
0.24
0.33
345.9
60.1
34.2
10.8
5.5
423
255
175
147
116
1.22
4.25
5.11
13.61
20.98
100
1
3.47
4.18
11.12
17.14
60.2
41.3
34.7
27.4
(b)
Cell extract
Heat treatment
DEAE-sepharose
Phenyl-sepharose 6 fast flow
13
12
33
30
17.38
5.74
1.21
0.85
225.9
68.8
39.8
25.5
1730
1251
1045
752
7.66
100.0
1.00
2.37
3.43
3.85
18.17
26.25
29.50
72.3
60.4
43.5
1
2
3
1
2
3
(b)
(a)
Fig. 3. (a) SDS-PAGE showing purified recombinant AgoG2GI-rec enzyme contained HisTag tail. (1) Crude cell extract of E. coli BL21 DE3 expressing recombinant AgoG2GI-
rec enzyme. (2) AgoG2GI-rec after heat shock. (3) The recombinant enzyme purified by MagneHis Protein Purification System. (b) SDS-PAGE showing purified recombinant
AgoG2GI-rec enzyme without HisTag tail. (1) Crude cell extract of E. coli BL21 DE3 expressing recombinant AgoG2GI-rec enzyme. (2) AgoG2GI-rec after heat shock (3)
AgoG2GI-rec after ion-exchange column and hydrophobic interaction chromatography.
of GenBank (NCBI, NIH, Washington DC) and analysis revealed
the presence of a 1326 bp open reading frame (ORF) encoding a
hypothetical 441 amino acid protein with a molecular mass of
49,976 Da (calculated by ProtParam, www.expasy.org), identified
by a database enquiry (BLASTP).
addition of IPTG resulted in a high expression of soluble glu-
cose isomerase activity of 7.66 U/mg for AgoG2GIrec but a low
expression at 0.05 U/mg for AgoG2GI-recHis. Due to the low activ-
ity of AgoG2GI-recHis, purification studies were initiated with
AgoG2GI-rec.
In A. gonensis G2T, the glucose isomerase gene was not expressed
unless d-xylose was present in the medium. AgoG2GI-wt was
expressed in A. gonensis G2T with specific activity of 1.22 U/mg in
the presence of 0.5% xylose. The purification procedures applied
for both enzymes (AgoG2GI-rec and AgoG2GI-wt) were the same
except for ammonium sulfate precipitation and are described under
Section 2. Because the enzyme was stable for several hours at
room temperature, all purification applications were performed
at room temperature. Table 2 shows the effects of all purifica-
tion techniques on the specific activities, fold purification and
yield. Because of the thermostability of the recombinant enzyme,
heat treatment of the cell-free extract at 75 ◦C for 15 min was a
very efficient purification step for thermostable glucose isomer-
ase from either E. coli or A. gonensis G2T. Indeed, AgoG2GI-wt and
AgoG2GI-rec were estimated to be approximately 60% and 72%
pure after this heat treatment step, respectively. A 17.14 and 3.85-
fold purification and a yield of 27.4% and 43.5% for AgoG2GI-wt
and AgoG2GI-rec, respectively, were obtained from the cell-free
extracts. Fig. 3 gives the SDS-PAGE analysis of AgoG2GI-rec and
AgoG2GI-wt.
3.3. Comparison of the protein sequences
The phylogenetic tree of AgoG2GI and other glucose isomerase
amino acid sequences indicates that A. gonensis G2T glucose isom-
erase clearly belongs to xylose isomerase family II, which includes
the enzymes from E. coli, Bacillus sp., Lactobacillus sp., Lactococcus
sp., Thermoanaerobacterium thermosulfurigenes and Thermotoga sp.
[22]. Family I glucose isomerases are shorter by 40 to 50 residues at
the N-terminal end; Streptomyces sp., Actinoplanes sp., Ampullariella
sp., Arthrobacter sp. and T. thermophilus are the examples of family
I (Fig. 2) [23].
3.4. Expression and purification of glucose isomerase gene
The gene encoding AgoG2GI was engineered for expression to
give pAgoG2GI-rec or pAgoG2GI-recHis. In both cases, expression
of the glucose isomerase gene was under the control of a T7 pro-
moter. The two constructs were used to transform E. coli BL21(DE3).
Overexpression of the cloned glucose isomerase induced by the

220
H. Karaoglu et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 215–224
(a)
100
80
60
40
20
0
100
80
60
40
20
0
(b)
AgoG2GI-wt
AgoG2GI-rec
AgoG2GI-wt
AgoG2GI-rec
0
100
200
Time (Min)
300
400
20 30 40 50 60 70 80 90 100 110
Temperature (ºC)
Fig. 4. (a) Effect of temperature on activity of the purified glucose isomerases. (b) Effect of temperature on stability of the purified glucose isomerases. The percentage relative
enzyme activity was calculated by comparison with unincubated enzyme.
3.5. Determination of molecular mass
3.7. Effects of temperature on activity and stability
Analysis of the soluble fractions on SDS-PAGE revealed a band
below the 50 kDa protein marker that did not correspond to the
anticipated molecular mass of 50 kDa for AgoG2GI. The discrep-
ancy between the theoretical mass and observed mass was resolved
by MALDI-TOF analysis. The molecular mass of AgoG2GI-wt and
AgoG2GI-rec were determined as 50.047 kDa. The apparent solu-
tion molecular mass of AgoG2GI was determined to be 180 kDa
according to gel filtration analysis using a SuperdexTM 200 10/300
GL column (Amersham Pharmacia Biotech).
The effects of temperature on glucose isomerase activity and
stability were tested spectrophotometrically using d-glucose as
substrate. The optimum temperature for glucose isomerase activity
was observed to be at 85 ◦C (Fig. 4a).
AgoG2GI-rec and AgoG2GI-wt retained 100% of their activity at
room temperature and 4 ◦C (data not shown). The enzymes lost
more than 50% of their activity at 85 ◦C after 3 h (Fig. 4b). The
enzymes also require Co2+ or Mg2+ and Mn2+ for high thermal
stability (data not shown).
3.6. Characterization of the cloned glucose isomerase
3.8. Effects of pH on activity and stability
The native (AgoG2GI-wt) and expressed glucose isomerases
(AgoG2GI-rec) displayed identical molecular weights and compo-
sitions, pH and temperature activity optima, thermostabilities, and
metal ion requirements.
The optimum pH for the glucose isomerase activities of
AgoG2GI-rec and AgoG2GI-wt were observed to be 6.5 (Fig. 5a).
The enzymes were active and stable in the broad pH range of
5.0 to 9.0 at 4 ◦C. When assayed at various pH values at 85 ◦C, the
purified enzymes showed the most stability at the pH range of 8 to
9. Fig. 5b shows the effect of pH on stability of AgoG2GI-rec enzyme.
The effect of pH on stability of AgoG2GI-wt was not shown since
the results of AgoG2GI-wt are in accordance with AgoG2GI-rec.
The glucose isomerases exhibited simple Michaelis–Menten
kinetics. Based on the Michaelis–Menten equation, the apparent
enzyme Km, Vmax, and kcat values for d-glucose were determined as
146.08 9.50 mM, 43.72 1.01 ␮mol/min/mg protein and 36.47
2.01 (1/s), respectively, for AgoG2GI-wt, and 138.37 7.63 mM,
40.51 0.81 ␮mol/min/mg protein and 33.79 (1/s), respectively for
AgoG2GI-rec. The Km, Vmax and kcat values did not significantly
change for the recombinant and native enzymes. The initial rates
were calculated by measuring the absorption at 560 nm for several
concentrations of glucose over a range of 0 mM to 700 mM.
The Km and Vmax values for AgoG2GI-rec toward different sub-
strates, l-arabinose, d-ribose and d-mannose, were determined
to be 58.74 9.5 mM, 0.0413 ␮mol/min/mg; 80.32 10.06 mM,
0.85 ␮mol/min/mg, and 121.8 25.33 mM, 0.0276 ␮mol/min/mg,
respectively. kcat values for AgoG2GI-rec toward l-arabinose, d-
ribose, and d-mannose, were determined to be 0.0344, 0.709 and
0.023 (1/s), respectively.
3.9. Effect of metal ions
The effects of various metal ions on glucose isomerase activ-
ity was assayed at 85 ◦C (pH 6.5) using d-glucose as substrate for
AgoG2GI-wt and AgoG2GI-rec. Co2+ was the most required metal
ion for glucose isomerase activity, whereas Mg2+ and Mn2+ slightly
enhanced enzyme activity (Table 3). A minimum concentration of
1 mM CoCl2 was required to achieve maximum glucose isomerase
activity. Higher Mg2+ or Mn2+ concentrations did not increase the
activity to the level of 1 mM CoCl2. In the absence of metal ions
(Co+2, Mg+2, and Mn+2) the enzyme lost all of its original activity at
85 ◦C over a 2 h period. However, in the presence of 1 mM Co2+ or
Table 3
The activator effects of various metal ions on the glucose isomerase activity of AgoG2GI-wt and AgoG2GI-rec.
Concentration of metal ions (mM)
AgoG2GI-wt relative activity (%)
AgoG2GI-rec relative activity (%)
Co+2
Mg+2
Mn+2
Co+2
Mg+2
Mn+2
0
0
0
0
0
0
0
0.1
0.5
1
2
4
10
20
50
100
64 1.7
83 3.2
100 1.5
98 4.5
96 6.3
83 4.6
59 10.1
57 8.7
39 9.8
11 1.4
10 1.5
10 1.1
12 0.3
18 2.7
24 3.2
22 2.1
26 0.6
24 1.2
21 2.4
35 2.1
36 3.4
35 7.5
24 2.1
26 1.2
21 0.4
21 0.7
17 1.1
61 2.1
82 1.4
100 1.5
93 8.9
93 7.5
88 5.8
55 3.5
45 2.1
32 1.6
10 1.3
15 0.7
14 2.1
13 0.8
17 0.9
28 2.1
32 2.7
25 2.6
26 1.4
19 1.1
35 2.6
43 2.9
34 1.3
28 0.4
28 0.7
26 1.1
31 2.1
20 0.4

H. Karaoglu et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 215–224
221
(a) 100
AgoG2GI-wt
AgoG2GI-rec
80
60
40
20
0
4
5
6
7
8
9
10
11
pH
(b)
5
4°C
85°C
5,5
6
120
100
80
60
40
20
0
110
100
90
80
70
pH 5
pH 6
pH 7
pH 8
pH 9
6,5
7
7,5
8
8,5
9
60
50
9,5
10
10,5
0
5
10
0
100
200
Time (hour)
300
400
Time (hour)
Fig. 5. (a) Effect of pH on the activity of the purified glucose isomerases. (b) Effect of pH on stability of AgoG2GI-rec enzyme at 4 ◦C and 85 ◦C. The percentage relative enzyme
activity was calculated by comparison with unincubated enzyme.
Mn2+, the enzymes retained 100% and 36% (or 43%) of their original
activities, respectively. The activities of purified glucose isomerases
were dependent mainly on Co2+ rather than Mn2+ or Mg2+ at 85 ◦C.
Mg2+ and Mn2+ are determined as the activators of AgoG2GI-wt
and AgoG2GI-rec as with other GIs [24]. AgoG2GI-wt and AgoG2GI-
rec were inhibited by the addition of chloride and sulfate salts
of Cd2+, Ca2+, Hg2+, Ni2+, Zn2+, Fe2+ and Cu2+ bivalent metal ions
(Table 4).
Heterologous proteins overproduced in E. coli are often gen-
erated in the form of insoluble inclusion bodies [25]. Whether a
protein precipitates in the cytoplasm or remains soluble, however,
depends also on the nature of the protein being overproduced [26].
In the case of the recombinant thermostable glucose isomerase,
the enzyme remained soluble, which may be an important factor
in attempts to scale up its production [16]. However, expressing
enzymes from hyperthermophiles in mesophilic hosts raises ques-
tions about the properties of the recombinant enzyme versus the
native enzyme. In addition to the problems normally encountered
in expressing recombinant proteins, the correct folding of a hyper-
thermophilic protein at significantly lower temperatures is a key
concern [23]. Therefore, we purified the enzyme not only from
E. coli BL21/pAgoGI-rec but also from A. gonensis G2T and compared
the kinetic parameters of both enzymes. We found no significant
differences, suggesting that the host organism did not affect the
refolding of the enzyme and that both enzymes are of the same
size.
Glucose isomerase is a tetramer, and the asymmetric unit of
the crystal contains a dimer. Each subunit contains two domains.
The main domain is a parallel-stranded ␣ ␤ barrel. The C-terminal
domain is a loop structure consisting of five helical segments
and is involved in intermolecular contacts between subunits [2].
AgoG2GI-recHis have the His tag at the C-terminus in the resulting
recombinant enzyme. Overexpression of the cloned glucose isom-
erase resulted in a high expression of soluble glucose isomerase
activity for AgoG2GIrec but a low expression for AgoG2GI-recHis.
These results suggest that the C-terminal region is important
for intersubunit contacts, which may explain why the His tag is
unfavorable for enzyme activity. Due to the low activity of AgoG2GI-
recHis, characterization studies were initiated with AgoG2GI-rec.
4. Discussion
The screening method, based on a specific assay that detected
conversion of glucose to fructose on agar plates, facilitated the iden-
tification of glucose isomerase activity in A. gonensis G2T [16]. Here,
we have described the cloning, expression, purification and bio-
chemical characterization of the A. gonensis G2T glucose isomerase
gene (xylA). This is the first example of GI from A. gonensis G2T and
the genus Anoxybacillus.
The A. gonensis G2T glucose isomerase clearly belongs to glu-
cose isomerase family II, which includes the enzymes from Bacillus
stearothermophilus (91%), Geobacillus kaustophilus HTA426 (90%),
Geobacillus sp. C56-T3 (89%), Bacillus halodurans C-125 (79%), Bacil-
lus megaterium DSM 319 (77%) and Bacillus coagulans 36D1 (75%),
with the indicated amino acid sequence similarity, and over 70%
for many other bacteria. Enzymes of family I are shorter than
those of family II by 40 to 50 residues at the N-terminal end. The
A. gonensis G2T enzyme is different from the enzymes belong to
family I glucose isomerases. According to phylogenetic analysis of
AgoG2GI and other glucose isomerase amino acid sequences, the A.
gonensis G2T glucose isomerase clearly belongs to xylose isomerase
family II.

222
H. Karaoglu et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 215–224
The molecular mass estimated from SDS-PAGE analysis was
inconsistent with the anticipated value based on the correspond-
ing protein sequence. Such a discrepancy may be due to the
nature of the protein under investigation. Sizing accuracy of SDS-
PAGE depends on the protein characteristics, such as amino acid
sequence, isoelectric point, structure, and the presence of certain
side chains or prosthetic groups. Consequently, discrepancies may
arise for certain proteins, such as glycosylated proteins, which do
not truly migrate according to their molecular weight [27]. The
apparent inconsistency between theoretical mass and observed
mass estimated from SDS-PAGE was resolved by MALDI-TOF analy-
sis, which confirmed the size of both AgoG2GI-wt and AgoG2GI-rec
to be 50.047 kDa.
AgoG2GI-wt and AgoG2GI-rec showed optimal activity at 85 ◦C
when incubated with d-glucose for 30 min. Glucose isomerases
described in the literature are generally operate at an optimum
temperature between 60 ◦C and 80 ◦C with a few exceptions: [2]
the optimum temperature for Streptomyces sp.[28] and Bacillus
sp.[29] was observed to be 85 ◦C; the optimum temperature of
the glucose isomerase of Thermotoga neapolitana was reported as
95 ◦C [23], the highest Topt glucose isomerase. The thermodynamic
equilibrium fructose/glucose ratio increases with temperature so
commercial processes are carried out with immobilized enzymes
at around 60 ◦C, which is the effective upper limit of their ther-
mostability. These processes yield 45% fructose/55% glucose syrups,
but 55% fructose syrups are desirable for food and soft drinks use,
so the former are fortified with chromatographically purified fruc-
tose. This expensive step would be superfluous if the isomerization
could be performed at 90–95 ◦C where 55% fructose syrups could,
in theory, be produced directly. Hence, there has been a widespread
search for glucose isomerases that are stable at these elevated tem-
peratures [6]. We can conclude that AgoG2GI is among the highest
Topt glucose isomerases described to date. AgoG2GI has activity
of more than 50% at 90–95 ◦C. This finding has great potential in
biotechnological terms because the higher temperatures drive the
equilibrium of the isomerization of glucose toward fructose for-
mation. Thus, the Topt 85 of AgoG2GI makes it one of the favored
enzymes for HFCS production [30].
The AgoG2GIs lost 50% of their original activity at 85 ◦C after
3 h. However, full activity was retained after an extended period
of storage at +4 ◦C. Bacillus sp. glucose isomerase retained 100%
of its activity at the temperature range of 50 ◦C to 80 ◦C and lost
its activity above 80 ◦C within 30 min [29]. Bacillus TX-3 glucose
isomerase lost more than 20% of its activity at 80 ◦C within 10 min
[31]. The half-life of Thermoanaerobacterium sp. glucose isomerase
is 1 h at 82 ◦C [21]. Based on these data, it is suggested that AgoG2GI
is more stable than these glucose isomerases at 85 ◦C.
The activity of glucose isomerase was tested in the pH range
of 5.0 to 11.0. The optimum activities of many glucose isomerases
were observed at pHs higher than 6.5, ranging generally between
pH 7.0 and 9.0 [2]. Glucose isomerases from Streptomyces sp., T. ther-
mophilus, and Clostridium thermosulfurogenes showed pH optimum
at 7.0 for d-glucose substrate [14,16,28]. However, commercial
application of glucose isomerase requires an acidic pH optimum to
enable starch liquefaction and isomerization to be performed in a
single step. In addition, the acidic conditions in industrial processes
decrease the formation of undesired compounds like browning
products (mannose, psicose and other acidic compounds). Fig. 5
shows the pH dependence of the enzymatic activity, measured
after 30 min at 85 ◦C for d-glucose substrate, demonstrating high-
est activities around pH 6.5. This enzyme can be thought suitable
for industrial applications, like Lactobacillus brevis having a lower
pH optimum between 6.0 and 7.0 [2]. To determine the stability of
the enzyme at pH values of 5.0 and 10.5, pre-incubation was per-
formed at each pH value at 4 ◦C for 300 h and at 85 ◦C for 8 h. The
enzymes were active and stable in a broad pH range between 5.0

H. Karaoglu et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 215–224
223
and 9.0 at 4 ◦C (Fig. 5). When assayed at various pH values at 85 ◦C,
the purified enzymes showed the most stability at the pH range of
8 to 9 (Fig. 5). The pH-optima of activity and stability are different
for AgoG2GI and this difference occurs in some cases according to
Talley and Alexov [32].
The pH stability of an enzyme is very important to predict its
storage condition, because the enzyme can lose activity after being
stored at different pH values. According to these results, glucose
isomerase can be stored in a broad range of pH conditions for a
long period of time without significant loss of activity.
AgoG2GI has very high kcat value for glucose. The kcat value of
AgoG2GI was determined to be 36.47 2.01 (1/s), which is greater
than that of Streptomyces chibaensis (30.72/s) [47], A. missourien-
sis (25.3/s) [48], Arthrobacter sp. (19.9/s) [41] and T. thermophilus
(0.833/s) [49]. In general, the kcat value of AgoG2GI is the second
best among the published kcat values of other species except the
species of Bacillus (53/s) [29]. This is almost the highest turnover
rate among any of the wild-type glucose isomerases characterized
so far. Moreover, AgoG2GI shows a better catalytic efficiency with
respect to glucose. It remains to be seen if the catalytic efficiency
for the industrial substrate can be improved by site-directed muta-
genesis, as has been done for other xylose isomerases [24].
Substrate orientation, especially the hydroxyl group (–OH) of
sugar is speculated to define the preferential substrate for iso-
merases [50]. In addition to d-glucose, AgoG2GI also isomerizes
d-ribose, l-arabinose, and d-mannose, which are non-preferential
substrates for glucose isomerases. kcat values for AgoG2GI-rec
toward l-arabinose, d-ribose and d-mannose were determined to
be 0.0344, 0.709 and 0.023 (1/s), respectively. AgoG2GI has a higher
kcat value for glucose than the other substrates. According to these
results, l-arabinose, d-ribose and d-mannose are non-preferential
substrates for this glucose isomerase.
The high thermostability of the A. gonensis G2T glucose isomer-
ase made it possible to use heat treatment of crude cell extracts as
one of the most efficient purification steps when the enzyme was
produced in mesophilic hosts. Thermal treatment caused the dena-
turation of mesophilic host proteins. The presence of metal ions
(Mg2+ and Co2+) and a high protein concentration in cell extracts
were essential for optimal recovery of the enzyme during heat
treatment [33]. The activity of xylA from G2 was more dependent on
Co2+ than on Mn2+ or Mg2+ at 85 ◦C. Co2+ increased thermal stabil-
ity effectively, whereas Mg2+ had a low effect and Mn2+ decreased
thermal stability (data not shown). Co2+ is supposed to be more
important for stabilization of the multimeric structure, resulting in
thermostability of the enzyme [34].
It has long been known that all glucose isomerases require either
Co2, Mg2+ or Mn2+ ions for activity and stability. We can now see
that all of these enzymes have essentially similar metal ion require-
ments, but the activity and stability depend on pH, temperature,
substrate and the presence of other ions [6]. AgoGI showed no
measurable activity in the absence of metal ions. In the presence
of Co+2, AgoGI has the highest activity, as is the case for glucose
isomerases from B. coagulans [35,36], Bacillus sp. [29,37], Thermus
aquaticus [38] and Streptomyces violaceoruber [39]. The effects of
Mg2+, Co2+ or Mn2+ on AgoGI are consistent with Vieille et al. for
glucose isomerase from Bacillus licheniformis [24].
The behavior of AgoGIs for some metal ions was examined
by using chloride and sulfate salts of each metal at concentra-
tions of 0 mM to 10 mM. AgoGIs were inhibited by divalent ions
of heavy metals, strongly by Cd2+, Hg2+, Zn2+, Fe2+ and Cu2+, but
only moderately by Ni2+ and a competitive inhibitor, Ca2+, which is
in accordance with the results of Lehmacher and Bisswanger [38],
Rangarajan and Hartley [40], Smith et al. [41], Suekane et al. [42],
Chen and Anderson [43] and Kwon et al. [37].
The domestic HFCS market in 2011 sustained the strong
momentum of demand, production and sales from 2010. The indus-
try developed rapidly and the production recorded 2.36 million
tons in 2011, whereas it had been merely 400,000 t in 2006. The
total amount of HFCS produced by glucose isomerase exceeds a mil-
lion tons per year [44]. Acid-stable glucose isomerases, which are
resistant to inhibition by Ca2+, are useful in a uni-pH process. The
combination of saccharification and isomerization is an ideal devel-
opment in the progress of HFCS production, and it is likely to be in
operation once an acid-stable, thermostable and Ca2+-tolerant GI
is discovered. Ca2+ showed slight inhibition on AgoGI and has high
optimum temperature but the optimum pH for AgoGI was observed
to be 6.5. This can be improved by protein engineering and can be
used for commercial production of HFCS [2].
The AgoG2GI-wt or AgoG2GI-rec glucose isomerase has kinetic
characteristics for glucose isomerization similar to those of other
glucose isomerases with similar amino acid sequences. With
respect to glucose, however, the AgoG2GI-wt or AgoG2GI-rec
enzymes have very low Km and a higher Vmax (146.08 mM and
43.72 1.01 ␮mol/min/mg protein) than most thermophilic glu-
cose isomerases. The Km values for GIs from Arthrobacter sp.,
A. missouriensis, Bifidobacterium adolescentis, Streptomyces sp. and
Bacillus sp. were 210 mM [41], 290 mM [45], 398 mM [46], 400 mM
[28] and 142 mM [29], respectively.
5. Conclusion
In conclusion, a new d-glucose isomerase (AgoG2GI) was iso-
lated, cloned and characterized from A. gonensis G2T. Our analysis
indicates that this enzyme is among the best d-xylose–glucose
isomerases described so far. Special features of AgoG2GI is its rel-
atively high catalytic activity toward the substrate d-glucose with
relatively high Vmax, low Km, high kcat, and high kcat/Km values. The
enzyme preferentially uses glucose as a substrate.
Given the promising characteristics of AgoG2GI, it may com-
pete with the enzymes currently used in industry. However, further
experiments are needed before practical industrial utility can be
attained i.e., lowering the Topt to 60 ◦C, which is the temperature
used in the industrial process.
Acknowledgments
We are grateful to the Karadeniz Technical University Research
Foundation (Grant no. 2003.111.04.6) and the Scientific and
Research Council of Turkey (TUBITAK, Grant no 104T472 and TBAG-
AY/395(104T380)) for financial support. We also thank Dr. Gareth
Roberts (University of Edinburgh, Scotland) for his critical reading
of this manuscript and his constructive comments.
References
[1] M.G. Wovcha, D.L. Steuerwald, K.E. Brooks, App. Environ. Microb. 45 (1983)
1402–1404.
[2] S.H. Bhosale, M.B. Rao, V.V. Deshpande, Microbiol. Rev. 60 (1996) 280–300.
[3] V.J. Jensen, S. Rugh, Methods Enzymol. 136 (1987) 356–370.
[4] A. De Raadt, M. Ebner, C.W. Ekhart, M. Fechter, A. Lechner, M. Strobi, A.E. Stutz,
Catal. Today 22 (1994) 549–561.
[5] W.D. Crabb, J.K. Shetty, Opin. Microbiol. 2 (1999) 252–256.
[6] B.S. Hartley, N. Hanlon, R.J. Jackson, M. Rangarajan, Biochim. Biophys. Acta 1543
(2000) 294–335.
[7] E.B. Da silva, A.U. De souza, S.G. De Souza, A.E. Rodrigues, Adsorption 11 (2005)
847–851.
[8] G.D. Schellenberg, A. Sarthy, A.E. Larson, M.P. Backer, J.W. Crabb, M. Lidstrom,
B.D. Hall, C.E. Furlong, J. Biol. Chem. 259 (1984) 6826–6832.
[9] M. Wilhelm, C.P. Hollenberg, Nucleic Acids Res. 13 (1985) 5717–5722.
[10] K. Dekker, H. Yamagata, K. Sakaguchi, S. Udaka, Agric. Biol. Chem. 55 (1991)
221–227.
[11] G.J. Joo, J.H. Shin, G.Y. Heo, Y.M. Kim, I.K. Rhee, J. Microbiol. 43 (2005) 34–37.
[12] G.C. Saari, A.A. Kumar, G.H. Kawasaki, M.Y. Insley, P.J. O’Hara, J. Bacteriol. 169
(1987) 612–618.
[13] R. Amore, C.P. Hollenberg, Nucleic Acids Res. 17 (1989) 75–115.
[14] K. Dekker, H. Yamagata, K. Sakaguchi, S. Udaka, J. Bacteriol. 173 (1991)
3078–3083.

224
H. Karaoglu et al. / Journal of Molecular Catalysis B: Enzymatic 97 (2013) 215–224
[15] A.O. Belduz, S. Dulger, Z. Demirbag, Int. J. Syst. Evol. Microbiol. 53 (2003)
1315–1320.
[33] C.Y. Lee, M. Bagdasarian, M.H. Meng, J.G. Zeikus, J. Biol. Chem. 265 (1990)
19082–19090.
[16] C.Y. Lee, L. Bhatnagar, B.C. Saha, Y.E. Lee, M. Takagi, T. Imanaka, M. Bagdasarian,
J.G. Zeikus, App. Environ. Microb. 56 (1990) 2638–2643.
[17] T. Trigilia, M.G. Peterson, D.J. Kemp, Nucleic Acids Res. 16 (1988) 81–86.
[18] M.M. Bradford, Anal. Biochem. 72 (1976) 248–254.
[19] Z. Dische, E. Borenfreund, J. Biol. Chem. 192 (1951) 583–587.
[20] T. Maniatis, E.F. Fritsch, J. Sambrook, Molecular Cloning: A Laboratory Manual,
Cold Spring Harbor Laboratory, New York, 1982, pp. 545.
[21] S.Y. Liu, J. Wiegel, F.C. Gherardini, J. Bacteriol. 178 (1996) 5938–5945.
[22] C. Vieille, J.M. Hess, R.M. Kelly, J.G. Zeikus, Appl. Environ. Microbiol. 61 (1995)
1867–1875.
[34] J. Cha, Y. Cho, R.D. Whitaker, H.L. Carrell, J.P. Glusker, P.A. Karplus, C.A. Batt, J.
Biol. Chem. 269 (1994) 2687–2694.
[35] G. Danno, Agric. Biol. Chem. 34 (1970) 1805–1814.
[36] G. Danno, Agric. Biol. Chem. 35 (1971) 997–1006.
[37] H.J. Kwon, M. Kitada, K. Horikoshi, Agric. Biol. Chem. 51 (1987) 1983–1989.
[38] A. Lehmacher, H. Bisswanger, J. Gen. Microbiol. 136 (1990) 679–686.
[39] M. Callens, H. Kersters-Hilderson, W. Vangrysperre, C.K. Debruyne, Enzyme
Microb. Technol. 10 (1988) 695–700.
[40] M. Rangarajan, B.S. Hartley, Biochem. J. 283 (1992) 223–233.
[41] C.A. Smith, M. Rangarajan, B.S. Hartley, Biochem. J. 277 (1991) 255–261.
[42] M. Suekane, M. Tamura, C. Tomimura, Agric. Biol. Chem. 42 (1978) 909–917.
[43] W.P. Chen, A.W. Anderson, Appl. Environ. Microbiol. 38 (1979), 1111-111.
[44] T. Johannes, M.R. Simurdiak, H. Zhao, in: S. Lee (Ed.), Biocatalysis, Encyclopedia
of Chemical Processing, New York, 2006, pp. 101–110.
[23] J.M. Hess, V. Tchernajenko, C. Vieille, J.G. Zeikus, R.M. Kelly, Appl. Environ.
Microbiol. 64 (1998) 2357–2360.
[24] C. Vieille, K.L. Epting, R.M. Kelly, J.G. Zeikus, Eur. J. Biochem. 268 (2001)
6291–6301.
[25] F. Baneyx, M. Mujacic, Nat. Biotechnol. 22 (2004) 1399–1408.
[26] G. Bueli, N. Panayotatos, in: W. Reznikoff, L. Gold (Eds.), Mechanism and
Practice, Maximizing Gene Expression, Butterworths Publishers, Boston, 1986,
pp. 345–363.
[27] H. Goetz, M. Kuschel, T. Wulff, C. Sauber, C. Miller, S. Fisher, C. Woodward, J.
Biochem. Biophys. Methods 60 (2004) 281–293.
[28] C.U. Inyang, U. Gebhart, S.K.C. Obi, H. Bisswanger, Appl. Microbiol. Biotechnol.
43 (1995) 632–638.
[29] J. Chauthaiwale, M.B. Rao, Appl. Environ. Microbiol. 60 (1994) 4495–4499.
[30] L.D. Kluskens, J. Zeilstra, A.C. Geerling, W.M. De vos, J. Van der Oost, Environ.
Technol. 31 (2010) 1083–1090.
[45] J. Jenkins, J. Janin, F. Rey, M. Chiadmi, H. Van Tilbeurgh, I. Lasters, M. DeMaeyer,
D. VanBelle, S.J. Wodak, M. Lauwereys, P. Stanssens, N.T. Mrabet, J. Snauwaert,
G. Matthyssens, G.A. Lambeir, Biochemistry 31 (1992) 5449–5458.
[46] Y. Kawai, H. Konishi, H. Horitsu, H. Sakurai, K. Takamizawa, T. Suzuki, K. Kawai,
Biosci. Biotechnol. Biochem. 58 (1994) 691–694.
[47] G.J. Joo, J.H. Shin, G.Y. Heo, Y.Y. Kwak, J.H. Choi, I.K. Rhee, Agric. Chem. Bio-
technol. 44 (2001) 113–118.
[48] J. Karimaki, T. Parkkinen, H. Santa, O. Pastinen, M. Leisola, J. Rouvinen, O.
Turunen, Protein Eng. Des. Sel. 17 (2004) 861–869.
[49] A. Lonn, M. Gardonyi, W. Van Zyl, B. Hahn-Hagerdal, R.C. Otero, Eur. J. Biochem.
269 (2002) 157–163.
[31] M. Kitada, Y. Dobashi, K. Horikoshi, Agric. Biol. Chem. 53 (1989) 1461–1468.
[32] K. Talley, E. Alexov, Proteins 78 (2010) 2699–2706.
[50] D.H. Patel, E.J. Cho, H.M. Kim, I.S. Choi, H.J. Bae, Protein Eng. Des. Sel. (2012)
1–6.