FSAB: A new fructose-6-phosphate aldolase from Escherichia coli. Cloning, over-expression and comparative kinetic characterization with FSAA

Article abstract of DOI:10.1016/j.molcatb.2012.02.010

Fructose-6-phosphate aldolase B (FSAB) from Escherichia coli was successfully over-expressed as His-tagged recombinant protein. A decameric protein was observed as for FSAA. Unlike FSAA, FSAB is not thermally stable at temperatures higher than 60 °C. The 70% identity between the two aldolases has allowed the generation of a 3D structure which has shown a high similarity of the two active sites. Full kinetic studies towards several substrates have revealed that FSAB catalytic activity is very close to FSAA activity, corroborated by the similarity of their active sites. FSAB has been able to react with three known donors (dihydroxyacetone, hydroxyacetone and glycolaldehyde) but always slightly slower than FSAA.

Full text of DOI:10.1016/j.molcatb.2012.02.010

Journal of Molecular Catalysis B: Enzymatic 84 (2012) 9–14
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
FSAB: A new fructose-6-phosphate aldolase from Escherichia coli. Cloning,
over-expression and comparative kinetic characterization with FSAA
Israel Sánchez-Morenoa,b , Lionel Nautona,b , Vincent Thérya,b , Agnès Pinetc,d,e , Jean-Louis Petitc,d,e
,
Véronique de Berardinisc,d,e, Anne K. Samlandf, Christine Guérard-Hélainea,b,∗, Marielle Lemairea,b,∗
a Clermont Université, Université Blaise Pascal, ICCF, BP 10448, F-63000 Clermont-Ferrand, France
b CNRS, UMR 6296, BP 80026, F-63177 Aubière, France
c CEA, IG, Génoscope, Evry, France
d CNRS, UMR 8030, Evry, France
e Université d’Evry Val d’Essone, Evry, France
f
Institute of Microbiology, Universität Stuttgart, 70550 Stuttgart, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Fructose-6-phosphate aldolase B (FSAB) from Escherichia coli was successfully over-expressed as His-
tagged recombinant protein. A decameric protein was observed as for FSAA. Unlike FSAA, FSAB is not
thermally stable at temperatures higher than 60 ◦C. The 70% identity between the two aldolases has
allowed the generation of a 3D structure which has shown a high similarity of the two active sites. Full
kinetic studies towards several substrates have revealed that FSAB catalytic activity is very close to FSAA
activity, corroborated by the similarity of their active sites. FSAB has been able to react with three known
donors (dihydroxyacetone, hydroxyacetone and glycolaldehyde) but always slightly slower than FSAA.
© 2012 Elsevier B.V. All rights reserved.
Received 9 December 2011
Received in revised form 14 February 2012
Accepted 21 February 2012
Available online 3 March 2012
Keywords:
Fructose-6-phosphate aldolase B
C
C bond formation
Recombinant aldolase
Enzyme characterization
Substrate specificity
1. Introduction
aldolases. Moreover, FSAA presents the unprecedented advan-
tage to react with four different unphosphorylated donors (DHA,
Aldol reaction is one of the most useful tools for the synthetic
chemist. This reaction offers powerful strategies towards the syn-
thesis of both enantiopure natural and novel poly-hydroxylated
compounds. Aldolases have been used in the synthesis of chiral
compounds by stereoselectively catalysing the reversible forma-
tion of C C bonds. These enzymes catalyse aldol additions between
nucleophilic donors (ketones or aldehydes) and electrophilic accep-
tors (aldehydes) [1]. One of the most important recent discoveries
in the aldolase family was fructose-6-phosphate aldolase (FSAA,
gene mipB from Escherichia coli) which was first reported by
Sprenger in 2001 [2]. This enzyme has the striking advantage
of accepting unphosphorylated dihydroxyacetone (DHA) as donor
instead of the expensive and unstable dihydroxyacetone phos-
phate (DHAP) strictly used by the well-known DHAP-dependent
hydroxyacetone, 1-hydroxybutan-2-one and glycolaldehyde) [3]
providing an access to a large variety of sugars and analogues
(Scheme 1).
Thus, FSAA has been efficiently used to synthesize different
polyhydroxylated compounds such as iminocyclitols and car-
bohydrates [3a,b,4]. To broaden the scope of such biocatalyst
for synthetic applications, mutants were constructed by rational
design. Indeed, A129S mutant [5] conferred a more hydrophilic
active site to FSAA, increasing the enzyme affinity for DHA. Then the
double mutant A129S/A165G [6] was designed to better accommo-
date C-␣ substituted aldehydes in the active site. It should also be
noticed that using the scaffold of a transaldolase (TALB, gene talB
from E. coli) a new efficient DHA-dependent aldolase was designed
by a single amino acid replacement, F178Y [7]. This work together
with some FSAA results has been recently reviewed [1b]. Another
approach to provide organic chemists with useful biocatalysts, dif-
ferent from protein engineering, is to study aldolases that can be
found in nature. Notably, at the same period as Sprenger’s group
discovered FSAA, another protein codified in the genome of E. coli
K-12 (gene talC) was identified as a new FSA (named FSAB) [2b]. Due
to its lower specific activity towards DHA compared to FSAA, this
∗
Corresponding authors at: Clermont Université, Université Blaise Pascal, ICCF,
BP 10448, F-63000 Clermont-Ferrand, France. Tel.: +33 4 73 40 75 84;
fax: +33 4 73 40 77 17.
E-mail addresses: christine.helaine@univ-bpclermont.fr (C. Guérard-Hélaine),
marielle.lemaire@univ-bpclermont.fr (M. Lemaire).
1381-1177/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2012.02.010

10
I. Sánchez-Moreno et al. / Journal of Molecular Catalysis B: Enzymatic 84 (2012) 9–14
Scheme 1. Aldol reaction catalysed by fructose-6-phosphate aldolase (FSA).
enzyme was not further investigated. Since then, other donors like
hydroxyacetone and glycolaldehyde have been described as much
better donors than DHA [3c]. In this way, a complete study of FSAB
is of interest, in particular with the above cited donor substrates.
This work presents the biochemical characterization of FSAB and
a comparison of kinetic parameters between FSAA and B for several
substrates.
to 30 ◦C. The culture was incubated for a further period of 12 h.
Cells were harvested by centrifugation, washed twice and resus-
pended in buffer A (50 mM NaH2PO4, 300 mM NaCl, pH 8.0). Cell
suspension was disrupted by sonication and the cell lysate was
centrifuged at 10,000 × g for 20 min. Clear supernatant was loaded
onto a Ni2+–NTA–agarose column (Qiagen, h = 1.5 cm; Ø = 2.5 cm)
pre-equilibrated with buffer A containing imidazole 20 mM (buffer
B). The column was washed with buffer B, and the retained proteins
were eluted with the same buffer containing 250 mM of imidazole.
The purity of fractions was assessed by SDS-PAGE. Eluted fraction
containing pure protein was dialysed against buffer for desalting
and imidazole removal. The protein was dialysed against water (pH
close to 8.0) to remove buffer salts before the lyophilization pro-
cess. This was also used when protein freeze-drying was necessary
without loss of activity. Untagged FSAA was expressed and puri-
fied as previously described [5] and untagged FSAB as described for
FSAA by Schneider et al. [7] but omitting the heat treatment.
2. Materials and methods
2.1. General
Expression E. coli strains BL21(DE3)pLysE and BL21(DE3)pLysS
cells were from Invitrogen. Plasmids pET22b(+) and pET11a were
from Novagen. Plasmid pET22b(+) (Novagen) modified for ligation
independent cloning [8] was done by Génoscope [8b]. d-Fructose-
6-phosphate dipotassium salt and d,l-glyceraldehyde 3-phosphate
diethyl acetal barium salt were purchased from Sigma–Aldrich. d,l-
Glyceraldehyde 3-phosphate diethyl acetal barium salt was depro-
tected according to the protocole provided by Sigma–Aldrich. The
enzymes phosphoglucose isomerase, glucose-6-phosphate dehy-
drogenase, ␣-glycerophosphate dehydrogenase-triosephosphate
isomerase from rabbit muscle, and glycerol dehydrogenase from
Cellulomonas sp. were also purchased from Sigma–Aldrich.
2.4. Biochemical characterization
SDS-PAGE was performed as described by Laemmli [10] using
15% polyacrylamide gels. Electrophoresis samples were always run
under reducing conditions and gels were stained with Coomassie
brilliant blue R-250. Protein gels were quantified by densitometry
using the image analysis software GeneTools (Syngene). Apparent
proteins molecular weight in their native state was determined by
gel filtration on a Superdex 200 column (Amersham Biosciences)
as previously described [9].
2.2. Bacterial strains and vectors
Genes encoding FSAA (mipB, GeneBank Accesion number
BAA13552.1) and FSAB (talC, GeneBank Accesion number
P32669.1) enzymes from E. coli K12 (strain MC4100) were
amplified by PCR using specific primers (PfsaA 5PRIM: 5ꢀ-
aaagaaggagataggatcatgcatcatcaccatcaccatgaactgtatctggatacttcaga-
cg-3ꢀ; PfsaA 3PRIM 5ꢀ-gtgtaatggatagtgatcttaaatcgacgttctg cc-
2.4.1. Optimum pH
FSAB optimum pH was calculated assaying its retroal-
dol activity at pH values between 5.0 and 9.0. Activity was
determined by measuring the formation of dihydroxyacetone
(DHA) and d-glyceraldehyde-3-phosphate (d-G3P) from 40 mM
d-fructose-6-phosphate (F6P) by spectrophotometric monitoring
in reaction mixtures of 2 mL. Each reaction contained 50 mM
glycylglycine buffer and 0.18 U pure FSAB. pH was adjusted
before enzyme addition and it was checked at the end of
the reaction. Standard spectrophotometric assays for DHA/d-
G3P detection were performed in 1 mL of 50 mM glycylglycine
buffer containing NADH 0.4 mM, glycerol dehydrogenase (1.5 U)/␣-
glycerophosphate dehydrogenase (1.3 U) plus triose phosphate
isomerase (12.6 U), and reaction aliquots of 30–60 ␮L. Variations of
A340 nm were proportional to DHA/d-G3P concentration in reaction
(εNADH = 6220 cm−1 M−1).
3ꢀ;
ccatgaactgtatctggacacc-3ꢀ
PfsaB 5PRIM:
5ꢀ-aaagaaggagataggatcatgcatcatcaccatca
PfsaB 3PRIM
5ꢀ-gtgtaatggatagtga
tcttagagatgagtagtgccaaatg-3ꢀ). The amplified sequences were
inserted into an expression vector pET22b(+) modified [9]. The
forward primers introduced a hexahistidine sequence in the
proteins after the initial methionine for purification purposes.
The sequences of the resulting plasmids, named pET22-fsaA and
pET22-fsaB, were verified.
For expression of an untagged version of FSAB, the gene
was amplified by PCR from chromosomal DNA of E. coli
K12 (strain W3110) using specific primers (fsaB up 5ꢀ-
ttttgctagcgaactgtatctggacaccgc-3ꢀ; fsaB down 5ꢀ-ttttaagcttttaga
gatgagtagtgccaaatgcg-3ꢀ). The amplified sequence was cloned
into an expression vector pET11a using the restriction sites NheI
and HindIII. The pET16-fsaA construct was used for expression of
untagged FSAA as described previously [2,7].
2.4.2. Enzyme thermostability
Thermal stability of FSA enzymes was determined by mon-
itoring the remaining retroaldol activity of purified enzymes
(5–10 mg mL−1 in glycylglycine buffer 50 mM pH 8.5) along the
time at three different incubation temperatures: 50 ◦C, 60 ◦C and
70 ◦C. Retroaldol activity was assayed using the previously reported
FSA activity assay from 45 mM F6P [2].
2.3. Expression and purification
E. coli colonies (expression strain) containing the plasmids
pET22-fsaA and pET22-fsaB were cultured in Luria–Bertani (LB)
broth in presence of selection antibiotic at 37 ◦C with shaking.
When the culture reached an OD600 nm of 0.5, protein expression
was induced with IPTG (0.5 mM) and the temperature was dropped
2.4.3. d,l-G3P stability
Stability of d,l-G3P was assayed at different reaction conditions,
such as in presence of FSA’s purified by IMAC or heat treatment. All

I. Sánchez-Moreno et al. / Journal of Molecular Catalysis B: Enzymatic 84 (2012) 9–14
11
the experiments were performed at pH 7.5 to avoid stronger basic
conditions that could affect the chemical stability of d,l-G3P. Incu-
bation buffer was selected in a previous experiment where d,l-G3P
stability was tested in presence of three different buffers: HEPES,
glycylglycine, and borate (50 mM, pH 7.5), and water as control
assay (pH adjusted to 7.5). In stability assays, 6 mM of d,l-G3P was
incubated in borate buffer (50 mM, pH 7.5) in presence of three
different purified FSA samples: (i) purified by heat treatment, (ii)
purified by IMAC, and (iii) purified by heat treatment followed
by IMAC (extra-pure enzyme). Each reaction contained 4 U of the
corresponding enzyme and a control lacking enzyme was run in
parallel.
2.5. Activity assay and kinetic constant studies
One unit of FSA activity was defined as the amount that cleaves
1 ␮mol of d-fructose-6-phosphate per min to afford d-G3P and DHA
at pH 8.5 (glycylglycine buffer 50 mM) and 25 ◦C.
Kinetics of FSAB for general donor and acceptor substrates were
measured following the corresponding previously described pro-
tocol [3c]. Kinetics of FSAA and B for retroaldol reaction from
d-fructose and l-sorbose were calculated by measuring the amount
of DHA formed using a glycerol dehydrogenase from Cellulomonas
sp. (GDHcell). One mL of each kinetic assay contained glycyl-
glycine buffer 50 mM pH 8.5, 1.5 U FSA, substrate concentrations
(d-fructose or l-sorbose) from 20 mM to 800 mM, 10 U GDHcell
and 0.5 mM NADH. Decrease of A340 nm was proportional to the
Fig. 1. SDS-PAGE analysis of the FSAB expression and purification. Equal amounts
of protein (10 ␮g) were loaded in each lane. Lane 1: cell free extract; lane 2: heat
treatment purification fraction (75 ◦C for 40 min); lane 3: IMAC purification fraction.
63% and the purification fold was increased to 2.3. The production
of this expression system was 256 U/L of culture. Similar results
were obtained for His-tagged FSAA (357 U/L of culture).
SDS-PAGE analysis of FSAB expression has shown an IPTG
inducible protein matching the expected molecular weight of the
protein plus the His-tag (24,380 Da) that represented 40% of the
total protein in soluble fraction (CFE) analysed by densitometry
(Fig. 1). This analysis of IMAC fractions has also shown final high
purity protein fractions of 95%.
amount of DHA produced in a reaction (εNADH = 6220 cm−1 M−1
)
where 1 mmol of NADH oxidized was equivalent to 1 mmol of sub-
strate cleaved.
2.6. Software and hardware
In addition, enzymes purified by IMAC could be freeze-dried
and stored at 4 ◦C as powder without any loss of activity for at
least one month. It has to be mentioned that several attempts from
Sprenger’s group to add a His-tag by the mean of another plas-
mid construction on FSAA at either its C- or N-terminus failed [14].
On the one hand, the N-terminal His-tagged protein could not be
retained by an IMAC column and on the other hand, the C-terminal
His-tagged protein has led to 85% loss of activity. C-terminal His-
tagged FSAA was also described by Wong’s group [3b] but with no
data about the efficiency or success of this construction. Remark-
ably, the different approach described here where the six histidines
are directly branched to the protein without a linker has allowed
conserving the whole activity for both FSAA and B.
Automated homology modelling was performed by MODELLER
9.10 [11] in chimera [12] and confirmed with SWISS-MODEL [13]
website.
Sequence retrieval alignment and template selection
Sequence of FSAB (P32669.1) was retrieved from Uniprot
database in FASTA format. Template FSAA was retrieved from RCSB
˚
database, PDB code 1L6W, crystallographic structure with 1.93 A
resolution. This structure of FSAA is decameric, homology was per-
formed only on the chain A from crystallographic structure.
In MODELLER 9.10, modelling was based on the alignment
between sequences of FSAA and FSAB, and crystallographic struc-
ture (1L6W chain A) was used as template for structuration of FSAB.
In SWISS-MODEL website three different ways of modelling
were performed: (i) the template used was 1L6W (decamer), (ii)
the template used was just chain A of 1L6W, and (iii) no template
was used.
3.2. Thermostability
One of the characteristics of FSAA illustrated by Sprenger’s
group [2] is a high thermostability. This property is very useful
from a biocatalytical point of view, and heat treatment has been
mostly used as the sole method for direct purification of wt or
mutant FSAA from CFE. In order to determine if FSAB also display
this interesting property, the standard purification protocol usually
used with FSAA including a heat treatment was applied. Thus,
CFE containing the recombinant enzyme FSAB was incubated
at 75 ◦C for 40 min. The final fraction has lost 85.5% of its initial
activity (Table 1) and the purity degree determined by SDS-PAGE
densitometry raised only to 60% (Fig. 1). However, the protein
that survived the heat treatment and stayed in solution showed
an increase in specific activity (Table 1) compared to CFE. In con-
clusion, the His-tag fusion enzyme FSAB is not as thermostable as
previously described for FSAA and will not permit heat treatment
purification. Previously, it was proposed that the thermostability
of FSA might be due to the structural properties i.e. tight packing
of this enzyme [15]. To exclude that the lack of thermal stability is
3. Results and discussion
3.1. Cloning, overexpression and purification of FSAA and B from
E. coli
FSAA and B from E. coli were cloned as N-terminal His-tagged
recombinant proteins in order to simplify their purification pro-
cedure. A hexahistidine sequence was introduced after the initial
methionine. Thus, His-tagged FSA could be purified in only one
step: an IMAC affinity chromatography that was followed by a
desalting procedure (dialysis) for removing imidazole from elu-
tion buffer. Starting from 0.5 L of culture of FSAB expressing cells
(178 mg of protein in CFE with a specific activity of 0.72 U mg−1),
50 mg of pure protein was obtained with a specific activity of
1.62 U mg−1 (Table 1). Final yield of this purification method was

12
I. Sánchez-Moreno et al. / Journal of Molecular Catalysis B: Enzymatic 84 (2012) 9–14
Table 1
Purification methods of recombinant FSAB (from 0.5 L of cell culture, retroaldol activity measured at RT for d-fructose-6-P).
Sample
Activity (U)
Protein (mg/ml)
Volume (ml)
Specific activity (U/mg)
Fold purification
Recovery (%)
1 Cell free extract (CFE)
2 Heat treatment
3 IMAC purification
128.0
18.6
80.3
9.36
0.85
4.05
19.0
19.0
12.3
0.72
1.15
1.62
–
1.6
2.3
100
14.5
62.7
50ºC
60ºC
70ºC
120
100
80
120
100
80
60
40
20
0
120
100
80
60
40
20
0
60
40
20
0
0
50
100
150
200
0
50
100
150
200
0
50
100
150
200
Time (min)
Time (min)
Time (min)
Fig. 2. Retroaldol activity monitored in His6-tag FSAA (᭹), wt FSAA (᭹), His6-tag FSAB (ꢀ) and wt FSAB (ꢁ). Plots are triplicate mean values.
caused by the addition of the His-tag, the thermal stability of FSAB
variants with and without tag was explored. In comparison tagged
and untagged versions of FSAA were investigated. The retroaldol
activity on F6P with purified enzymes was monitored along the
time at three different incubation temperatures: 50 ◦C, 60 ◦C and
70 ◦C (Fig. 2). Significant differences were observed at 60 ◦C and
70 ◦C, showing that wt FSAB is less thermostable (half-life lower
than 25 min at both temperatures) than wt FSAA (half-life around
600 min at 70 ◦C). The His-tag is clearly not involved in the loss
of thermostability for both enzymes as illustrated by the curve
profiles. His6-tagFSAB has a half-life of 160 min at 60 ◦C and below
25 min at 70 ◦C, whilst His6-tagFSAA displays half-life of 700 min
at 70 ◦C, and it is as stable as wt FSAA at 60 ◦C. Why FSAB is less
thermostable than FSAA is not clear at the moment.
3.4. 3D structure simulation of FSAB
FSAB has not been crystallized and thus its 3D structure remains
unknown. Nevertheless, FSAA and FSAB show high sequence iden-
tity of about 70%, this allows modelling the three-dimensional
structure of FSAB by sequence homology towards FSAA. In this
work, we used two different programs, MODELLER 9.10 [11] and
the website of SWISS-MODEL [12].
The superimposition (Fig. 4) obtained from MODELLER 9.10
shows that all results are similar until residue 195 which is the
beginning of the C-terminal alpha helix. The last alpha helix has a
flexible point with residue 196 and thus could be positioned differ-
ently.
The FSAB sequence was confronted to model 1L6W (SWISS-
MODEL website). A decameric model of FSAB was obtained based
on the decameric model of FSAA described in the 1L6W structure.
The study of the decameric structure of FSAB generated by
SWISS-MODEL did not reveal any clashes in regions were sequences
are different. This has been corroborated by the experimental
From this thermostability study, we can conclude that FSAB can-
not be efficiently purified through heat treatment.
To finally conclude on the purification process, another advan-
tage of the IMAC technique has been illustrated by studying the
stability of the best acceptor substrate d,l-G3P as a model. Indeed,
d,l-G3P and phosphorylated sugars in general are natural metabo-
lites and contaminant enzymes like isomerases or phosphatases
could react with them. As depicted in Fig. 3, d,l-G3P is less stable
in the presence of heat treated FSAA than in the presence of other
FSAA preparations of higher purity. This result illustrates that for
use as tool in organic synthesis, the purification through IMAC or
heat treatment should be decided depending on the acceptor sub-
strate targeted. This choice is only possible for FSAA as FSAB cannot
be purified by heat treatment.
8
6
4
2
0
3.3. Enzyme characterization
The molecular mass of the purified FSAB, as estimated by
SDS-PAGE and Coomassie Blue staining, was ∼24 kDa. This mass
was consistent with the calculated molecular weights of 23.6 kDa.
According to gel filtration experiments, FSAB is functional as a
decamer exactly like FSAA. The effect of pH on FSAB activity was
assayed. Optimal activity was found at pH 8.5, as described previ-
ously for FSAA [2].
0
1
2
3
4
Time (h)
Fig. 3. Stability of d,l-G3P (initial concentration of 6 mM) in presence of His6-tag
FSAA purified by IMAC (ꢂ), heat treatment (᭹), IMAC + heat treatment (ꢃ), and in
absence of enzyme (ꢄ). All assays were carried out in borate buffer 50 mM and final
pH was 7.5.

I. Sánchez-Moreno et al. / Journal of Molecular Catalysis B: Enzymatic 84 (2012) 9–14
13
Table 2
Kinetic constants of recombinant FSAA and FSAB.
Entry
Substrate
FSAA
FSAB
KM (mM)
Keat (s−1
)
Keat/KM (s−1 mM−1
)
KM (mM)
Keat (s−1
)
Keat/KM (s−1 mM−1
)
1
2
3
4
5
6
7
8
DHAc
32a
17.4a
0.2a
116a
2527a
16.5a
16.5a
173a
27a
4a
145a
83a
27b
15b
41b
650b
11b
1.52b
43.3b
39b
HAc
GA donord
GA acceptord
d,l-G3Pe
0.28b
65b
62.8a
0.80a
9a
0.3a
11b
0.17b
62b
216a
3a
0.65b
6.70b
210b
271b
40b
d-Fructose-6-phosphatef
d-Fructose g
l-Sorboseg
7.3b
1.1b
128b
283b
0.4b
0.003b
0.0003b
0.064b
0.081b
0.0003b
0.0003b
0.086b
a
Data taken from Ref. [3c].
b
c
d
e
f
Data determined in this work.
d,l-G3P as acceptor.
Self-addition.
DHA as donor.
Cleavage assay by following d-G3P.
Cleavage assay by following DHA.
g
determination of the FSAB oligomeric state that was found as a
decamer (see Section 3.3 above). However, this model did not allow
any evaluation of parameters involved in the observed difference
in thermostability between FSAA and B.
Whatever the methods used for generating a 3D structure of
FSAB, all have shown that the active site is strongly conserved.
˚
When the analyses were located on a sphere of 10 A diameter
around the Lys 85 (Fig. 5) a perfect superimposition was observed,
which was not surprising if we consider that in this sphere the
sequence identity is higher than 95% (Fig. 6).
3.5. Kinetic analysis
In order to evaluate the catalytic properties of FSAB compared
to FSAA, kinetic parameters of FSAB were calculated for the most
commonly reported FSAA substrates [3c]. Thus the donors were
dihydroxyacetone (DHA), hydroxyacetone (HA), glycolaldehyde
(GA), whereas the acceptors were d,l-G3P and GA. We also studied
the retroaldol kinetics for d-F6P, d-fructose and l-sorbose. These
two latter have not yet been evaluated with kinetic parameters. To
be sure that the His-tag does not affect the kinetic properties of
FSAB the kinetic constants for d-F6P were performed in parallel for
the untagged and tagged version of FSAB. Both enzymes behaved
similar.
Fig. 5. FSAA (red), FSAB (blue) active sites superimposition. (For interpretation of
the references to color in figure legend, the reader is referred to the web version of
the article.)
catalytic efficiency kcat/KM three times lower than FSAA (Table 2,
entry 6). Moreover, like for FSAA, HA and GA (entries 2 and 3) were
determined far better donor substrates than DHA for FSAB (entry
1). However, it has to be noticed that FSAA is more efficient than
FSAB for the three donor substrates reported. These similar kinetic
parameters could result from the significant similarities (79%) and
identities (70%) between the FSAA and FSAB protein sequences.
As illustrated by the molecular modelling, both FSA active sites
are very similar in sequence and positioning of the active site
residues, and this could explain the same level of recognition
As previously reported by Sprenger’s group, retroaldol activity
for d-F6P was higher for FSAA than for FSAB [2b]. FSAB displays a
Fig. 4. Superimposition of MODELLER results for FSAB structure (blue) with FSAA
(red) as template. (For interpretation of the references to color in figure legend, the
reader is referred to the web version of the article.)
˚
Fig. 6. Sequence alignment of the residues included in the 10 A sphere around Lys
85.

14
I. Sánchez-Moreno et al. / Journal of Molecular Catalysis B: Enzymatic 84 (2012) 9–14
towards the substrates and similarities in kinetic parameters for
the two enzymes.
constant support and critical reading of the manuscript. We thank
N. Trachtmann for her help with cloning of the untagged FSAB.
Concerning acceptors kinetic studies, the same conclusion can
be raised. Remarkably, FSAB was also able to catalyse the autoaldol
reaction of glycolaldehyde, discovered for the first time with FSAA
[3c].
References
[1] (a) P. Clapés, W.-D. Fessner, G.A. Sprenger, A.K. Samland, Curr. Opin. Chem. Biol.
14 (2010) 154–167;
Regarding substrates d-fructose and l-sorbose in the retroaldol
reaction, kinetic analysis has shown that FSAA was able to sig-
nificantly differentiate between d-fructose and l-sorbose (10 fold
difference of respective kcat/KM values). A difference by a factor of
2 in the KM values and by a factor of 4.5 in kcat values have revealed
for the first time that the enzyme was stereoselective towards the
C5 position. On the other hand, FSAB has not displayed different
kcat/KM values for l-sorbose and d-fructose. Notably, this result was
the most significant difference in the catalytic properties between
the two enzymes. We have no rational explanation for those kinetic
differences.
(b) A.K. Samland, M. Rale, G.A. Sprenger, W.-D. Fessner, ChemBioChem 12
(2011) 1454–1474.
[2] (a) M. Schürmann, G.A. Sprenger, J. Biol. Chem. 276 (2001) 11055–11061;
(b) M. Schürmann Dissertation, University of Düsseldorf (2001).
[3] (a) M. Schürmann, M. Schürmann, G.A. Sprenger, J. Mol. Catal. B: Enzyme 19–20
(2002) 247–252;
(b) M. Sugiyama, Z. Hong, P.-H. Liang, S.M. Dean, L.J. Whalen, W.A. Greenberg,
C.-H. Wong, J. Am. Chem. Soc. 129 (2007) 14811–14817;
(c) X. Garrabou, J.A. Castillo, C. Guérard-Hélaine, T. Parella, J. Joglar, M. Lemaire,
P. Clapés, Angew. Chem. 121 (2009) 5629–5633;
(d) X. Garrabou, J.A. Castillo, C. Guérard-Hélaine, T. Parella, J. Joglar, M. Lemaire,
P. Clapés, Angew. Chem. Int. Ed. 48 (2009) 5521–5525.
[4] (a) A.L. Concia, C. Lozano, J.A. Castillo, T. Parella, J. Joglar, P. Clapés, Chem. Eur.
J. 15 (2009) 3808–3816;
(b) C. Guérard-Hélaine, B. Légeret, C. Fernandes, V. Prévot, C. Forano, M. Lemaire,
New J. Chem. 35 (2011) 776–779;
4. Conclusions
(c) J.A. Castillo, J. Calveras, J. Casas, M. Mitjans, M.P. Vinardell, T. Parella, T. Inoue,
G.A. Sprenger, J. Joglar, P. Clapés, Org. Lett. 8 (2006) 6067–6070.
[5] J.A. Castillo, C. Guérard-Hélaine, M. Gutierrez, X. Garrabou, M. Sancelme, M.
Schuermann, T. Inoue, V. Hélaine, F. Charmantray, T. Gefflaut, L. Hecquet, J.
Joglar, P. Clapés, G.A. Sprenger, M. Lemaire, Adv. Synth. Catal. 352 (2010)
1039–1046.
[6] M. Gutierrez, T. Parella, J. Joglar, J. Bujons, P. Clapés, Chem. Commun. 47 (2011)
5762–5764.
[7] S. Schneider, T. Sandalova, G. Schneider, G.A. Sprenger, A.K. Samland, J. Biol.
Chem. 283 (2008) 30064–30072.
In this work we have presented His-tagged FSAA and for the
first time His-tagged and untagged FSAB that could be obtained in
high purity, avoiding the presence of contaminant enzymes respon-
sible for side-reaction with sensible substrates such as d,l-G3P.
This His6-tag system allowed the production and characteriza-
tion of the new enzyme FSAB from E. coli, which did not display
enough thermostability to be purified by heat treatment. We have
demonstrated that FSAB is very similar to the previously described
FSAA also from E. coli. FSAA and B displaying a high sequence
identity have given comparable results such as kinetic behaviour,
oligomeric form and optimal pH value. Slight kinetic differences
were detected generally in favour of a better catalytic efficiency
for FSAA. This non exhaustive comparative study was performed
on seven substrates and eventually leaves an open frame for the
finding of different catalytic performances on other substrates that
would be in favour of FSAB.
[8] (a) C. Aslanidis, P.J. De Jong, Nucleic Acids Res. 18 (1990) 6069–6074;
(b) Unpublished results.
[9] A. Kreimeyer, A. Perret, C. Lechaplais, D. Vallenet, C. Médigue, M. Salanoubat, J.
Weissenbach, J. Biol. Chem. 282 (2007) 7191–7197.
[10] U. Laemmli, Nature 227 (1970) 680–685.
[11] A. Sali, T.L. Blundell, J. Mol. Biol. 234 (1993) 779–815.
[12] (a) K. Arnold, L. Bordoli, J. Kopp, T. Schwede, Bioinformatics 22 (2006) 195–201;
(b) F. Kiefer, K. Arnold, M. Künzli, L. Bordoli, T. Schwede, Nucleic Acids Res. 37
(2009) D387–D392;
(c) M.C. Peitsch, Nat. Biotechnol. 13 (1995) 658–660.
[13] E.F. Pettersen, T.D. Goddard, C.C. Huang, G.S. Couch, D.M. Greenblatt, E.C. Meng,
T.E. Ferrin, J. Comput. Chem. 25 (2004) 1605–1612.
[14] T. Inoue, Dissertation, University of Stuttgart (2006).
[15] (a) S. Thorell, M. Schürmann, G.A. Sprenger, G. Schneider, J. Mol. Biol. 319 (2002)
161–171;
Acknowledgements
(b) A.K. Samland, S. Baier, M. Schürmann, T. Inoue, S. Huf, G. Schneider, G.A.
Sprenger, T. Sandalova, FEBS J. 279 (2012) 766–778.
We gratefully acknowledge the Auvergne council for its finan-
cial support. We gratefully acknowledge Georg A. Sprenger for