Semisynthetic peptide-lipase conjugates for improved biotransformations

Article abstract of DOI:10.1039/c2cc34816k

An efficient chemoselective method for the creation of semisynthetic lipases by site-specific incorporation of tailor-made peptides on the lipase-lid site was developed. These new enzymes showed excellent improved specificity and regio- or enantioselectivity in different biotransformations.

Full text of DOI:10.1039/c2cc34816k

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 9053–9055
www.rsc.org/chemcomm
COMMUNICATION
Semisynthetic peptide–lipase conjugates for improved biotransformationsw
a
a
b
c
d
Oscar Romero, Marco Filice, Blanca de las Rivas, Cesar Carrasco-Lopez, Javier Klett,
e
d
c
a
a
Antonio Morreale, Juan A. Hermoso, Jose M. Guisan, Olga Abian and Jose M. Palomo*
Received 5th July 2012, Accepted 23rd July 2012
DOI: 10.1039/c2cc34816k
An efficient chemoselective method for the creation of semi-
synthetic lipases by site-specific incorporation of tailor-made
peptides on the lipase-lid site was developed. These new enzymes
showed excellent improved specificity and regio- or enantio-
selectivity in different biotransformations.
Chemical modification is a fascinating approach for altering
the protein function by the introduction of non-natural frag-
1
ments into proteins. The addition of unnatural moieties
2
such as fluorophores, sugars, peptides, etc.) to proteins has
(
also proven useful for a variety of processes and applications
both in vivo and in vitro. Furthermore this specific modifica-
tion on proteins has permitted the alteration of catalytic
properties of enzymes for creation of novel active selective
Scheme 1 Lipase-conjugates preparation.
This introduces the advantages of solid-state chemistry: using an
excess of peptides, quantitative transformations, or easy purifica-
9
tion. The catalytic activity of the new semisynthetic enzymes in
3
,4
biocatalysts. Therefore the combination of molecular biology
methods and efficient synthetic approaches has made possible
successful preparation of semisynthetic proteins in large
different asymmetric biotransformations was evaluated.
The incorporation of molecules on the oligopeptide lid
represents an elegant strategy to control its movement and
therefore the lipase catalysis. As far as we know, no example of
the improvement of lipase catalytic efficiency by this metho-
dology has been described.
5
,6
amounts. Modification of the nucleophilic thiol of a unique
cysteine is a widely employed strategy for site-selective bio-
7
conjugation. A cysteine can be introduced at virtually any
position within a protein structure by site-directed muta-
genesis and then selectively modified using for example
disulfide compounds. Furthermore, the disulfide can be effi-
ciently transformed into a thioether by a desulfurization
First we focused on the protein engineering of this lipase to
replace the two cysteines (Cys65, Cys296) in the wild type
enzyme (BTL wt) by two serines. The new-engineered enzyme
8
method.
(
BTL C65S/C296S) was expressed in E. coli without detriment
Herein, we describe the design of new semisynthetic enzymes
by the site-specific incorporation of a set of tailor-made
cysteine-containing peptides on successfully engineered thermo-
philic lipase from Geobacillus thermocatenulatus (BTL) with a
single cysteine at different positions on the lid-site of the
enzyme (Scheme 1). The chemical modifications were performed
on the lipase already immobilized on CNBr-activated Sepharose.
to the enzyme activity. Considering the complexity of this
lipase with a tricky lid formed by two different loops, the idea
was to introduce a unique cysteine at different positions of this
engineered enzyme. This approach was first confirmed by a
bioinformatics study using the crystallographic open confor-
1
0
mation of the enzyme we have recently solved.
Three different positions were selected to be mutated:
Ala193 (in the internal loop), Leu230 (in the external loop)
and Ser196 (in the middle of both loops). These three variations
represent a conservative change. The three new mutants
BTL-C65S/C296S/A193C (BTL*-A193C), BTL-C65S/C296S/
L230C (BTL*-L230C) and BTL-C65S/C296S/S196C (BTL*-
S196C) were obtained by site-directed mutagenesis with a
good production, similar to BTL wt or BTL C65S/C296S.
All BTL variants were efficiently purified by hydrophobic
a
Departamento de Biocata´lisis, Instituto de Cata´lisis (CSIC),
Marie Curie 2. Cantoblanco. CampusUAM, 28049 Madrid, Spain.
E-mail: josempalomo@icp.csic.es; Fax: +34 915854760
Departamento de Biotecnologı´a Microbiana, Instituto de Ciencia y
Tecnologı´a de alimentos y Nutricio´n, (ICTAN-CSIC), Madrid,
Spain
Departamento de Cristalografı´a y Biologı ´a Estructural,
Instituto de Quı´mica-Fı´sica Rocasolano (CSIC), Madrid, Spain
Unidad de Bioinforma´tica, Centro de Biologı´a Molecular Severo
Ochoa (CSIC-UAM), Madrid, Spain
Instituto de Biocomputacio´n y Fı´sica de Sistemas Complejos (BIFI),
Universidad de Zaragoza, Zaragoza, Spain
b
c
d
e
10
chromatography and characterized by circular dichroism
and fluorescence (see ESIw, Fig. S1 and S2).
For the strategic chemical incorporation of tailor-made
w Electronic supplementary information (ESI) available: Experimental
section, characterization data of the different BTL variants, additional
tables and figures and movie files. See DOI: 10.1039/c2cc34816k
peptides including a cysteine on the peptide sequence, the
This journal is c The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 9053–9055 9053

View Article Online
0
protein cysteine was previously protected using 2,2 -dithiodi-
4
pyridine disulfide (2-PDS) to avoid oxidation (Scheme 1). The
different protected BTL variants were specifically modified by
disulfide exchange. Considering the amino acid residues on the
lid loops two different peptides Ac-Cys-Phe-Gly-Phe-Gly-Phe-
CONH (p1) and Ac-Cys-(Asp) -Asp-COOH (p2) were selected
2
4
(
Fig. S3, ESIw). The BTL*-A193C was modified with the
complementary p1, BTL*-L230C with p2 whereas BTL*-S196C
was modified with p1 or p2. The coupling reaction in all cases
was performed in aqueous solution at pH 8 and 25 1C (see
8
ESIw). After coupling the desulfurization was performed to
get a thioether irreversible bond between the enzyme and the
peptide. A quantitative conversion into the corresponding
peptide-modified enzymes was observed by mass spectroscopy
(
MALDI-TOF; Fig. 1B; Fig. S4 and S5, ESIw). The excess of
peptide was easily removed by simple filtration of the immo-
bilized semisynthetic lipases. No enzyme modification was
observed after treatment of the BTL-C65S/C296S variant with
the peptides. The circular dichroism (CD) spectra confirmed
that the incorporation of the peptide affected the secondary
structure of the enzyme with increase in alpha-helix structure
after modification (Fig. 1C; Fig. S4 and S5, ESIw). The
fluorescence results (Fig. 1D; Fig. S4 and S5, ESIw) show a
decrease in signal intensity by the peptide coupling. The
irreversible inhibition experiments (Fig. S6, ESIw) showed that
the BTL variants exhibited a more open conformation with
the peptide incorporation.
Fig. 2 Comparison between crystal structures of active BTL and a
model active BTL*-A193C–p1 variant. (A) Molecular surface of active
BTL with Triton X-100 moieties in the active site (sticks 1 and 2). Lids
L1 and L2 are labeled. (B) Molecular surface of BTL*–A193C–p1.
Detergent moieties from the crystal structure are superimposed for
comparison. As observed, there is no room for moiety 2. Arrows
indicate the position of the catalytic Ser114. The position of p1 is
highlighted in red, and the aromatic residues (F , F , and F ) are
A
B
C
labeled.
variants, computer simulations indicate a smaller active site of
the modified enzyme than the wild type enzyme. The model of
the active BTL*-A193C–p1 variant reveals that the p1 moiety
lies on a hydrophobic patch at the active site (Fig. 2 and Fig. S8,
ESIw). The active site of the BTL*-S196C–p2 variant presents
strong modifications compared with that of BTL. This is mainly
In order to elucidate structural changes caused by peptide
incorporation, molecular dynamics simulations were carried
out for BTL*-A193C–p1 and BTL*-S196C–p2 variants. As
observed in crystal structures, BTL activation involves large
structural rearrangements and the concerted movement of
1
0
two lids, L
1
and L
2
(Fig. 2), unmasking the active site.
1 2
due to the insertion of the acidic p2 moiety between L and L
Conformational transitions from the closed to the open state
of BTL were simulated by the Targeted Molecular Dynamics
lids (Fig. S9, ESIw) that establishes strong salt bridge inter-
actions with basic residues of both lids (Fig. S10, ESIw).
The method was extended to peptides p3–p8 [Ac-Phe-Cys-
Phe-Gly-Phe-CONH (p3), Ac-Gly-Gly-Cys-Gly-Gly-CONH
(
(
TMD) method, based on the crystal structures of both states
see ESIw, Fig. S7). TMD coupled with energy minimizations
2
2
clearly indicate the resulting active sites that could be reached
in BTL variants depending on both the nature of the peptide
stem and the position at which they are attached. In both
(p4), Ac-Cys-(Arg) -CONH (p5), Ac-Cys-(Phe-Gly-) -Asp-
7 2 2
Asp-CONH2 (p6), Ac-Asp-Gly-Asp-Cys-Asp-CONH2 (p7),
Ac-Lys-Gly-Lys-Cys-Lys-CONH (p8)] (Fig. S11 and Table S2,
2
ESIw) with amino acids with different properties affording the
desire conjugates with high selectivity in yields >95%.
In order to investigate the catalytic effect of the peptide
incorporation on the BTL variants, these new biocatalysts
were applied in three different biotransformations. BTL wt
and BTL C65S/C296S variants showed similar catalytic results
in all biotransformations tested (see ESIw).
In the regioselective deprotection of per-O-acetylated
thymidine (1) (Table 1), the BTL*-A193C and BTL*-L230C
variants (mutants on both loops) hydrolyzed mainly at the C-5
position of ribose with slightly higher regioselectivity than
BTL C65S/C296S. The incorporation of p1 in the BTL*-
A193C or p2 in the BTL*-L230C variant improved the
regioselectivity of the enzyme (around 88% of 2) (Table 1).
However, the incorporation of other peptide sequences did not
produce a better result on regioselectivity (Tables S3 and S4,
ESIw).
Fig. 1 Characterization of the BTL*-A193C–p1 conjugate. (A)
MALDI-MS spectra of BTL*-A193C, [M + Na]; (B) MALDI-MS
spectra of BTL*-A193C–p1, [M + Na]; (C) far-UV CD spectra; (D)
fluorescent spectra. BTL*-A193C (blue), BTL*-A193C–p1 (red).
The introduction of the cysteine at 196 (BTL*-S196C)
caused an inversion and good regioselectivity of the lipase,
9
054 Chem. Commun., 2012, 48, 9053–9055
This journal is c The Royal Society of Chemistry 2012

View Article Online
Table 1 Regioselective hydrolysis of peracetylated tymidine 1 cata-
lyzed by peptide-modified immobilized BTL2 variants
Table 3 Desymmetrization of phenylglutaric acid dimethyl ester
catalyzed by different BTL-variants at pH 7.0
a
Peptide Time [h] C [%] Yield [%] ee [%]
b
Enzyme
a
Peptide pH Time [h] Yield 2 [%] Yield 3 [%] T
a
Enzyme
BTL-C65S/C296S
BTL*-A193C
BTL*-A193C
196
186
96
28
23
54
23
15
47
64
78
>99
—
p1
BTL C65S/C296S —
7.0 70
7.0 93
7.0 48
5.0 50
5.0 100
7.0 57
7.0 93
52
79
88
0
0
61
86
5
8
4
86
97
14
4
43
13
8
14
3
BTL*-A193C
BTL*-A193C
BTL*-S196C
BTL*-S196C
BTL*-L230C
BTL*-L230C
—
p1
—
p2
—
p2
a
Yield of the monoester 7. The rest of conversion corresponds to the
b
dicarboxylic acid. Determined by HPLC.
25
10
BTL variants showed higher enantiomeric excess than BTL wt
and BTL C65S/C296S (e.g. from 64% to 78%) (Table 3 and
Table S8, ESIw). The best result was found after the chemical
modification of BTL*-A193C with p1. This semisynthetic
lipase showed higher activity (4 fold) and excellent enantio-
meric excess (>99%). The peptide modification also increased
the selectivity of BTL*-S196C (with p1 or p2 ee >99%)
although a negative effect was observed for BTL*-L230C
a
Yield of the corresponding product at 100% conversion. T:
thymidine.
Table 2 Regioselective hydrolysis of peracetylated glucal 4 catalyzed
by peptide-modified immobilized BTL2 variants
(
Table S6, ESIw).
In conclusion, we report for the first time an efficient
chemoselective, and potentially general method, for the creation
of semisynthetic lipases by site-specific chemical incorporation
of tailor-made peptides on the lid-site of three different
cysteine-BTL variants based on a fast thiol-disulfide exchange
ligation.
a
b
Yield 5[%]
Enzyme
Peptide
Time [h]
Activity
BTLC65S/C296S
BTL*-A193C
BTL*-A193C
BTL*-A193C
—
—
p1
p3
24
1
1
5
183
214
40
79
70
73
94
5
These modifications generated new enzyme variants showing
different structural changes in the lid region yielding important
changes in the catalytic properties.
a
ꢀ1
lipꢀ1
ꢀ3
b
Specific activity is defined as: mmol min mg
ꢁ 10
.
the monodeprotected 5 at 100% conversion. The rest of yield corre-
Yield of
sponds to the bihydrolyzed product.
This work has been sponsored by the Spanish National
Research Council (CSIC) and by grants BFU2011-25326,
BFU2011-24595. O. Romero is grateful to CONICYT and
Programa Bicentenario Becas-Chile for financial support.
hydrolyzing at the C-3 position of 1 (86% of 3) without any
trace of 2 (Table 1). The incorporation of p2 enhanced the
regioselectivity of this BTL variant up to an excellent yield
(
97% of 3) with no trace of 2. This is the best value in
Notes and references
enzymatic regioselectivity ever reported in the literature so
far. The BTL*-A193C and BTL*-L230C peptide-conjugates
showed lower specificity when the reaction was performed at
pH 5.0 (Table S3, ESIw) whereas the BTL*-S196C variant
showed very low specificity at pH 7.0 (Table S4, ESIw).
Secondly, the regioselective deprotection of per-O-acetylated
glucal (4) was successfully catalysed by the semisynthetic
enzymes (Table 2 and Table S5, ESIw). All BTL variants were
regioselective to the monodeprotection at C-3. Surprisingly, an
increase of more than 36 fold of the enzyme activity was
observed with the introduction of the cysteine at 193. The
incorporation of p1 increased the activity but not the regio-
selectivity whereas p3 in BTL*-A193C improved the yield of 5
from 70 to 94%. For the other two mutants, the regio-
selectivity was not improved after the enzyme–peptide conjugates
formation (Table S5, ESIw).
1
E. M. Sletten and C. R. Bertozzi, Angew. Chem., Int. Ed., 2009, 48,
974–6998.
2 J. M. Palomo, Eur. J. Org. Chem., 2010, 6303–6314.
3 A. Dıaz-Rodrıguez and B. G. Davis, Curr. Opin. Chem. Biol., 2011,
5, 211–219.
C. A. Godoy, B. de Las Rivas, M. Filice, G. Ferna
J. M. Guisan and J. M. Palomo, Process Biochem., 2010, 45, 534–541.
6
´
´
1
4
´
ndez-Lorente,
5 J. M. Palomo, M. Lumbierres and H. Waldmann, Angew. Chem.,
Int. Ed., 2006, 45, 477–481.
6
L. Brunsveld, J. Kuhlmann, K. Alexandrov, A. Wittinghofer,
R. Goody and H. Waldmann, Angew. Chem., Int. Ed., 2006, 45,
6622–6646.
7 J. M. Chalker, G. J. L Bernardes, Y. A. Lin and B. G. Davis,
Chem.–Asian J., 2009, 4, 630–640.
8
G. J. L. Bernardes, E. J. Grayson, S. Thompson, J. M. Chalker,
J. C. Errey, F. El Oualid, T. D. W. Claridge and B. G. Davis,
Angew. Chem., Int. Ed., 2008, 47, 2244–2247.
9
M. J. Kurth, J. Am. Chem. Soc., 2010, 132, 10615.
1
0 C. Carrasco-Lo
Lorente, J. M. Palomo, J. M. Guisa
M. Martınez-Ripoll and J. A. Hermoso, J. Biol. Chem., 2009, 284,
4365–4372.
´
pez, C. Godoy, B. de Las Rivas, G. Ferna
´
ndez-
´
n, R. Fernandez-Lafuente,
´
Finally, the catalysis of the desymmetrization of dimethylphenyl-
´
glutaric acid diester (6) was studied. The three-monocysteine
This journal is c The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 9053–9055 9055