Application of Site-Directed Lipase Mutants on Regioselective Acylation of Monosaccharides

Article abstract of DOI:10.1081/CAR-120026464

Mutants F344V and F345V of Candida rugosa lipasel (CRL1) were tested in acylation reactions of monosaccharide derivatives 1-8, in order to study the regioselectivity, and the substrate specificity of lipase variants towards unnatural substrates, such as carbohydrates. Mutant F344V showed a better reaction kinetics and/or regioselectivity then the wild type enzyme with several substrates while mutant F345V was inefficient in most cases. With the aim of correlating experimental data with the structural features of the enzyme and substrates, the interaction of substrates methyl 4,6-O-benzylidene-α- D-glucopyranoside (5a) and 4,6-O-benzylidene-α-D-galactopyranoside (6a) with the wild type enzyme and the mutant F344V was investigated, using a molecular modelling approach.

Full text of DOI:10.1081/CAR-120026464

This article was downloaded by: [New York University]
On: 28 July 2015, At: 22:28
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: 5 Howick Place,
London, SW1P 1WG
Journal of Carbohydrate Chemistry
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/lcar20
Application of Site‐Directed Lipase Mutants on
Regioselective Acylation of Monosaccharides
Laura Cipolla a , Marina Lotti a , Luca De Gioia a & Francesco Nicotra a
a Department of Biotechnology and Biosciences , University of Milano‐Bicocca , Piazza della
Scienza 2, 20126, Milano, Italy
Published online: 23 Feb 2007.
To cite this article: Laura Cipolla , Marina Lotti , Luca De Gioia & Francesco Nicotra (2003) Application of Site‐Directed Lipase
Mutants on Regioselective Acylation of Monosaccharides , Journal of Carbohydrate Chemistry, 22:7-8, 631-644, DOI: 10.1081/
CAR-120026464
To link to this article: http://dx.doi.org/10.1081/CAR-120026464
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

JOURNAL OF CARBOHYDRATE CHEMISTRY
Vol. 22, Nos. 7 & 8, pp. 631–644, 2003
Application of Site-Directed Lipase Mutants on
Regioselective Acylation of Monosaccharidesy
*
Laura Cipolla, Marina Lotti, Luca De Gioia, and Francesco Nicotra
Department of Biotechnology and Biosciences, University of
Milano-Bicocca, Milano, Italy
ABSTRACT
Mutants F344V and F345V of Candida rugosa lipase1 (CRL1) were tested in
acylation reactions of monosaccharide derivatives 1–8, in order to study the
regioselectivity, and the substrate specificity of lipase variants towards unnatural
substrates, such as carbohydrates. Mutant F344V showed a better reaction kinetics
and/or regioselectivity then the wild type enzyme with several substrates while mutant
F345V was inefficient in most cases. With the aim of correlating experimental data
with the structural features of the enzyme and substrates, the interaction of substrates
methyl 4,6-O-benzylidene-a-D-glucopyranoside (5a) and 4,6-O-benzylidene-a-D-
galactopyranoside (6a) with the wild type enzyme and the mutant F344V was
investigated, using a molecular modelling approach.
Key Words: Biocatalysis; Lipases; Site-directed mutagenesis; Monosaccharides.
yThis paper is dedicated to Professor Ge´rard Descotes on the occasion of his 70th birthday.
*
Correspondence: Professor F. Nicotra, Department of Biotechnology and Biosciences, University
of Milano-Bicocca, Piazza della Scienza 2, 20126 Milano, Italy; Fax: +39-02-64483565; E-mail:
francesco.nicotra@unimib.it.
631
DOI: 10.1081/CAR-120026464
0732-8303 (Print); 1532-2327 (Online)
www.dekker.com
Copyright D 2003 by Marcel Dekker, Inc.

632
Cipolla et al.
INTRODUCTION
Lipases (triacylglycerol ester hydrolases, EC 3.1.1.3) are ubiquitous enzymes that
catalyse the hydrolysis of fats and oils.[1–4] In recent years lipases have received a
great deal of attention as biocatalysts in numerous industrial processes.[5,6] They are the
most widely used enzymes in synthetic organic chemistry, as catalyst of the chemo-,
regio-, and/or stereoselective hydrolysis of carboxylic acid esters or of the reverse
reaction in organic solvents.[6–8]
Recently, genetic engineering approaches have become powerful tools to improve
enzyme performance,[9–12] in general. They have been largely applied in the search for
more efficient enantioselective lipases.[13–16]
Lipases can acylate or deacylate hydroxyl groups present on carbohydrates (See
Scheme 1). Their ability to discriminate among the available secondary hydroxyl
groups in saccharides,[17] is particularly useful in organic synthesis. In general, lipases
exhibit a preference for primary hydroxyl groups, and regioselective acylation at O-6 of
glycopyranoses may be easily achieved.[18,19] In this context, we are interested in the
search for lipase variants with modified regioselectivity properties towards monosac-
charide derivatives. To this end, engineered forms of Candida rugosa lipase1 (CRL1)
were tested on different monosaccharide derivatives.
RESULTS AND DISCUSSION
A synthetic gene coding for CRL1 was obtained by mutually priming long
overlapping oligonucleotides. CUG codons were substituted with triplets coding for
serine in order to enable its heterologous expression because of the nonuniversal use of
CUG (serine instead of leucine) characteristic of C. rugosa genes.[20] The CRL1 active
site is located inside a long hydrophobic substrate binding tunnel and may be either
shielded or accessible to substrates according to the conformation of a mobile surface
loop named the lid.[21] The substrate-binding site has been mapped by co-crystallization
with substrate analogs.[22] Two bulky phenylalanine side chains F344 and F345 have
been shown to protrude into the tunnel, reducing its size. Therefore, the wild-type
Scheme 1. Reaction mechanism of lipases; in the present study R = vinyl and R’ =
monosaccharide.

Application of Site-Directed Lipase Mutants
633
recombinant lip1 gene was used as the template for PCR site-directed mutagenesis to
replace phenylalanine residues 344 and 345 by valine.[23] Mutants were expressed in
Pichia pastoris under the control of the methanol-inducible alcohol oxidase (AOX1)
promoter. Fermentation in sorbitol-based minimal medium gave good yields of the
recombinant proteins (0.6 mg/mL for the mutant F344V and 0.2 mg/mL for the mutant
F345V) in the culture supernatants, from which they could be easily recovered,
concentrated and lyophilised. The use of the minimal medium reduced the secretion of
proteins other than the recombinant product, thus allowing the direct use of lyophilized
culture supernatant in the reactions of interest. The variant F344V was secreted to
levels comparable to the wild type recombinant protein and displayed similar activity in
standard hydrolysis reactions (not shown), whereas both the secretion and the activity
of mutant F345V were reduced. The effect of introducing substitutions in recombinant
proteins is still largely unpredictable, since it can result in misfolding or impaired
transport through the secretory pathway of the host’s organism.[24] In fact, the sen-
sitivity of the P. pastoris expression system to the amino acid sequence of heterologous
proteins, and in particular rCRL, has been often noticed in our laboratory. Indeed, the
replacement of residues that do not play any obvious role in the folding or activity of
this protein can strongly decrease or even abolish its production.[25]
In order to study the different factors that can influence the regioselectivity of the
acylation the lipase mutants F344V, F345V and the wild-type lipase (CRL), were tested
for their ability to acylate a set of monosaccharides derivatised at their anomeric and
primary hydroxyl groups.
Figure 1. Substrates acylated by the WT-CRL, F344V and F345V mutants.

634
Cipolla et al.
Several parameters were investigated such as 1) the nature and orientation of the
anomeric substituent, 2) the bulkiness of the O-6 substituent, 3) and the stereochemistry
at C-4 as well as the blocking of the corresponding hydroxyl group (Figure 1). All
reactions were carried out in vinyl acetate, which by acting both as the solvent and the
acyl donor rendered the reaction completely irreversible.[26]
Initially, acylation of methyl 6-O-tritylglycosides 1[27] and 3[27] was studied using
CRL and the mutated enzymes F344V and F345V (Table 1). The regioselectivity of the
1
enzymatic acylation was determined from H NMR and COSY experiments (Table 2).
Analysis of the reaction kinetics indicated that a-anomers are better substrates of both
CRL and the F344V mutant than the b-ones; in both cases, acylation took place at
Table 1. Enzymatic acetylation of partially protected monosaccharide derivatives using CRL, and
mutants F344V and F345V.
React. time
(% conv.)a
% acylation in
position 2
Compd.
Enzyme
1a
CRL
36 h
14 h
> 98
> 98
> 98
90
F344V
F345V
CRL
1 month (10%)
10 days
10 days
1 month (10%)
14 h
1b
3a
3b
5a
5b
6a
6b
7a
7b
F344V
F345V
CRL
> 98
> 98
> 98
> 98
n.d.
92
F344V
F345V
CRL
14 h
1 month (10%)
10 days
10 days
1 month (10%)
3 days
F344V
F345V
CRL
95
n.d.
> 98
> 98
n.d.
70
F344V
F345V
CRL
3 days
40 days (10%)
10 days
10 days
40 days (10%)
10 days
5 days
F344V
F345V
CRL
> 98
n.d.
90
F344V
F345V
CRL
> 98
n.d.
–
40 days (10%)
n.r.
F344V
F345V
CRL
10 days
40 days (10%)
5 days
> 98
n.d.
> 98
–
F344V
F345V
CRL
n.r.
n.r.
–
5 days
n.r.
n.r.
> 98
–
F344V
F345V
–
n.r. = no reaction; n.d. = not determined.
aThe percentage of conversion is 100% when not otherwise stated.

Application of Site-Directed Lipase Mutants
635

636
Cipolla et al.
position 2 (Table 1); it’s worth noting that the regioselectivity of b-glucosides acylation
was slightly improved when using mutant F344V, instead of the wild-type enzyme, but
reaction rate was conserved. No improvement of the regioselectivity was observed in
the case of mutated lipase F345V. Besides, the reaction time was extremely long, hence
useless for synthetic purposes.
Acylation of the p-nitrophenyl glycosides 2 and 4 was studied in order to get better
insight into the influence of the bulkiness of the anomeric aglycon on the lipase activity.
Data (Table 1) showed that the concomitant presence of both the 6-O-trityl group and the
anomeric p-nitrophenyl group prevented acylation, presumably interfering with the
binding of the substrate. When the known 4,6-O-benzylidene glycosides 5[28] and 6[28]
were tested as substrates, the a-anomers were strongly preferred by the wild-type lipase
and its mutant F344V. Regarding mutant F345V, reaction times were, again, too long in
order to be of synthetic utility (5–10% of conversion after 40 days). Comparing the CRL
and the mutant F344V, it was observed that acylation with the wild-type enzyme was
totally regioselective for position 2 only when methyl 4,6-O-benzylidene-a-D-glucopyr-
anoside was the substrate. Again, the reaction rate was favored in the case of a-glycosides
(3 and 10 days reaction time for 5a and 6a, respectively, versus 10 days for 5b and no
reaction for 6b). On the contrary, although maintaining a strong preference for a-
glycosides (3 and 5 days reaction time for 5a and 6a, respectively, versus 10 days for 5b
and 6b), F344V showed improved reaction rate and regioselectivity towards all
benzylidene glycosides 5–6 (Table 1). In other words, F344V is less substrate-specific
compared to the wild-type enzyme (substrate 6b could be acylated with the mutated
lipase), but, at the same time, catalyses the acylation reaction with higher regioselectivity.
Finally, in order to elucidate if the minor substrate specificity, thus observed, could
be ascribed to the enlargement of the catalytic pocket, or if polarity effects had to be
taken into account, compounds 7 and 8 were synthesized. In this case, only the glucose
derivatives were acylated by the wild-type enzyme, with 1) total regioselection for
position 2, and 2) no apparent discrimination between anomers. The discriminating
factor seemed to be the configuration at the 4-OH. In that case, the mutants failed to
acylate the potential substrates.
With the aim of correlating some experimental data with the structural features
of enzyme and substrates, we have investigated the interaction of substrates 5a and
6a with the wild type enzyme and the mutant F344V, using a molecular model-
ling approach.
To find acceptable starting configurations of the sugar-protein adducts, different
binding modes of the substrates have been investigated by: 1) manually docking the
substrates in the hydrophobic funnel leading to the active site, 2) monitoring the
intermolecular energy as defined by Van der Waals and Coulomb terms, and 3) taking
into account that the anionic oxygen of the tetrahedral intermediate has to be directed
toward the oxyanion hole formed by the backbone NH atoms of Gly124 and Ala 210.
In fact, following the computational approach proposed by Manetti et al.,[23] we have
assumed that the different catalytic behaviours of the wild type enzyme and the mutant
F344V depend mainly on the stability and structural characteristics of the tetrahedral
intermediate formed along the reaction path. Remarkably, docking modes derived
clustering results obtained for each protein-adduct are extremely similar and the only
small differences are associated to rotation of the phenyl ring of the substrate (data not
shown). The increase of the maximum temperature in the simulated annealing pro-

Application of Site-Directed Lipase Mutants
637
cedure to 1000 K did not affect results, indicating that the observed limited con-
formational freedom of the substrate does not depend on the adopted methodology but
reflects the tight steric demand of the active site. However, significant differences are
observed when results for different protein-substrate adducts are compared. The most
stable structures of the enzyme-substrate adducts, as obtained by the simulated
annealing approach described in Methods, are shown in Figure 2.
Figure 2. The most stable structures of the enzyme-substrate adducts, as obtained by the
simulated annealing approach described in Methods. (A) CRL-5a; (B) CRL-6a; (C) F344V-5a; (D)
F344V-6a. The figure was prepared using the software package InsightII [Biosym/MSI 9685,
Scranton Road, San Diego, CA, USA]. (Go to www.dekker.com to view this figure in color.)

638
Cipolla et al.
Considering the structure of the active site, the substitution of phenylalanine 344
with valine results, as expected, in a larger pocket, decreasing the steric demands
imposed by the enzyme.
Investigations of Wild Type Lipase-Substrate 5a Adduct
Surprisingly, in the optimized structure of this adduct (Figure 2A) the negatively
charged oxygen atom of the tetrahedral intermediate weakly interacts with the oxyanion
cavity of the enzyme defined by the backbone NH groups of residues 123, 124 and
210. The sugar ring and the phenyl group are largely solvent exposed and affect only
slightly the orientation of Phe344 and Phe345. Residues Phe296 and Phe345 stabilize
the complex by hydrophobic interactions with the sugar ring.
Investigations of Wild Type Lipase-Substrate 6a Adduct
In this adduct, the oxyanion cavity strongly interacts with the substrate, with the
˚
NH groups of residues 123, 124 and 210 at a distance ranging from 2.9 to 3.5 A from
the charged oxygen. The sugar ring induces a large conformational rearrangement of
Phe344 and Phe345, to accommodate the substrate. Moreover, Phe296, Leu297, Phe344
and Phe345 form a hydrophobic cluster that interacts with galactose through stacking
interactions (Figure 2B). This is the only case where the enzyme structure is strongly
affected by the presence of the substrates.
Investigations of Mutant F344V-Substrate 5a Adduct
In this adduct, the negatively charged oxygen of the tetrahedral intermediate
strongly interacts with the oxyanion cavity. Moreover, the substitution of Val for
Phe344 enlarges the enzyme binding pocket and, as a consequence, the substrate can
interact with the hydrophobic pocket formed by Phe296, Leu297, Val344 and Phe345
(Figure 2C).
Investigations of Mutant F344V-Substrate 6a Adduct
In this adduct, only weak hydrogen bonds between the substrate and the oxyanion
cavity can be formed. The local structure of the enzyme is similar to that observed for
the F344V—5a adduct. However, the substrate cannot fit properly in the hydrophobic
pocket formed by Phe296, Leu297, Val344 and Phe345 (Figure 2D), interacting only
with residues 296 and 345.
Remarkably, the structural analysis of these adducts reveals that the interaction
between the substrate and the protein involves mainly hydrophobic and van der Waals
interactions. This suggests that the very different reactivity of the fluoro-derivatives 7
and 8, with respect to 5 and 6, may depend on electronic interactions, such as hydrogen
bonds, taking place during the initial recognition of the substrate by the enzyme.
Several considerations can be drawn from these studies: both the commercial, and
mutated F344V enzymes generally showed a sharp preference for a-glycosides; this
substrate specificity accounts for the different reaction kinetic, rendering the acylation
faster for a-substrates compared to their b-counterparts. The larger binding pocket of

Application of Site-Directed Lipase Mutants
639
the mutated enzyme F344V accounts for the minor substrate specificity, as confirmed
also by molecular modelling. In addition, different interactions between the relevant
residues in the oxyanion hole with the substrates play a fundamental role in the
determination of the regioselection; this effect is clearly evidenced for the enzymatic
reaction of 6a with wild-type enzyme and F344V mutant. No discrimination of the
anomers could be observed for compounds 7a and 7b, since the enzymes accepted both
the a- and b-pentafluorobenzylidene glucosides; on the contrary neither anomer of
galactosides (substrates 8a and 8b) could be acylated. In this case, the hydrophobic and
van der Waals interactions are probably responsible for the discrimination between
glucosides and galactosides.
Concerning mutant F345V, all enzymatic reactions were extremely slow, making
this mutant unsuitable for biocatalyzed acylation.
Finally, it is worth noting that the glycoside anomeric configuration didn’t
influence the regiochemistry of the acylation reaction, contrary to similar experiments
reported with lipase from Pseudomonas cepacia.[27]
EXPERIMENTAL
General methods. Solvents were dried over molecular sieves, for at least 24
h prior to use. Anhydrous reactions were performed under an Ar atmosphere. Thin-
layer chromatography (TLC) was performed on Silica Gel 60 F254 plates (Merck) with
detection with UV light when possible, or charring with a solution containing conc.
H2SO4/EtOH/H2O in a ratio of 5/45/45. Flash column chromatography was performed
on silica gel 230–400 mesh (Merck). Optical rotations were measured at room
temperature with a Perkin–Elmer 241 polarimeter. NMR spectra were recorded at 400
MHz on a Varian Mercury instrument using CDCl3 as solvent unless otherwise stated.
Chemical shifts are reported in ppm downfield from TMS as an internal standard;
aromatic signals are omitted. CRL was purchased from SIGMA.
Synthesis of substrates. Substrates 1, 3, 5, 6 were synthesised according to
standard procedures,[27,28] starting from the commercially available glycosides.
General procedure for the synthesis of compounds 2, 4. p-Nitrophenyl
glycosides were dissolved in dry pyridine, then trityl chloride was added. The reaction
mixtures were allowed to stir for 48 h, then the reaction mixture was concentrated to
dryness under reduced pressure. The crude material was purified by flash chromatography.
General procedure for the synthesis of compounds 7–8. Methyl glycosides (1
g, 5.15 mmol) were dissolved in dry acetonitrile (20 mL), then pentafluorobenzalde-
hyde (1.51 g, 7.73 mmol) and a catalytic amount of camphorsulfphonic acid (CSA, 24
mg, 0.10 mmol) were added. The reaction mixtures were allowed to stir overnight, then
the reaction mixture was diluted with AcOEt, and washed with a sat. aq. NaHCO3
solution. The organic layer was dried over Na2SO4, filtered and concentrated to dryness
under reduced pressure. The crude was purified by flash chromatography.
Enzymatic acetylation reaction. The enzymatic reaction was performed by
dissolving the substrate (10 mg) in vinyl acetate (0.5 mL), and incubating with the

640
Cipolla et al.
different lipases (10 mg) with shaking at rt. The conversion was monitored by TLC.
The enzyme was removed by filtration, and the solvent evaporated to dryness under
1
reduced pressure. The regioselectivity was then determined by H NMR.
p-Nitrophenyl 6-O-trityl-a-D-glucopyranoside (2a). p-Nitrophenyl a-D-gluco-
pyranoside (50 mg, 0.166 mmol) was reacted with trityl chloride (93 mg, 0.33 mmol),
in dry pyridine (1 mL) following the general procedure. After purification by flash
chromatography (eluent: pure AcOEt) afforded 76 mg of 2a (84%) as a colourless oil.
1
[a]D + 31.9 (c 1.5, CHCl3); H NMR: d 5.60 (d, 1 H, J = 3.4 Hz, H-1), 3.95 (t, 1 H,
J = 9.0 Hz, H-3), 3.74 (dd, 1 H, J = 9.0, 3.4 Hz, H-2), 3.72–3.68 (m 1 H, H-5), 3.54 (t,
1 H, J = 9.0 Hz, H-4), 3.37 (dd, 1 H, J = 10.4, 3.2 Hz, H-6a), 3.27 (dd, 1 H, J = 10.4,
5.7 Hz, H-6b), 2.30 (bs, 2 H, OH), 1.70 (bs, 1 H, OH).
Anal. Calcd for C31H29NO8: C, 64.50; H, 5.38; N, 2.58. Found: C, 64.56; H, 5.41;
N, 2.55.
p-Nitrophenyl 6-O-trityl-b-D-glucopyranoside (2b). p-Nitrophenyl b-D-gluco-
pyranoside (100 mg, 0.33 mmol) was reacted with trityl chloride (186 mg, 0.66 mmol),
in dry pyridine (1.5 mL) following the general procedure. After purification by flash
chromatography (eluent: petroleum ether:AcOEt 2:8) afforded 140 mg of 2b (78%) as a
1
colourless oil. [a]D ꢀ 22.2 (c 2, CHCl3); H NMR: d 4.86 (d, 1 H, J = 9.5 Hz, H-1),
3.61 (t, 1 H, J = 9.5 Hz, H-2), 3.57 (t, 1 H, J = 9.5 Hz, H-3), 3.50 (t, 1 H, J = 9.5 Hz,
H-4), 3.47–3.44 (m, 1 H, H-5), 3.39 (t, 1 H, J = 9.5 Hz, H-6a), 3.23 (dd, 1 H, J = 9.5,
6.2 Hz, H-6b), 2.40 (bs, 1 H, OH), 2.18 (bs, 2 H, OH).
Anal. Calcd for C31H29NO8: C, 64.50; H, 5.38; N, 2.58. Found: C, 64.47; H, 5.37;
N, 2.60.
p-Nitrophenyl 6-O-trityl-a-D-galactopyranoside (4a). p-Nitrophenyl a-D-galac-
topyranoside (37 mg, 0.123 mmol) was reacted with trityl chloride (68 mg, 0.25 mmol),
in dry pyridine (0.5 mL) following the general procedure. After purification by flash
chromatography (eluent: petroleum ether:AcOEt 2:8) afforded 50 mg of 4a (76%) as a
1
colourless oil. [a]D + 43.3 (c 0.75, CHCl3); H NMR: d 5.65 (d, 1 H, J = 3.4 Hz, H-1),
4.00 (dd, 1 H, J = 9.4, 3.4 Hz, H-2), 3.94 (bs, 1 H, H-4), 3.88 (dd, 1 H, J = 9.4, 3.3
Hz, H-3), 3.74 (t, 1 H, J = 5.3 Hz, H-5), 3.39 (dd, 1 H, J = 10.2, 5.3 Hz, H-6a), 3.26
(dd, 1 H, J = 10.2, 5.3 Hz, H-6b), 2.75 (bs, 3 H, OH).
Anal. Calcd for C31H29NO8: C, 64.50; H, 5.38; N, 2.58. Found: C, 64.53; H, 5.39;
N, 2.57.
p-Nitrophenyl 6-O-trityl-b-D-galactopyranoside (4b). p-Nitrophenyl b-D-galac-
topyranoside (100 mg, 0.33 mmol) was reacted with trityl chloride (186 mg, 0.66
mmol), in dry pyridine (1.5 mL) following the general procedure. After purification by
flash chromatography (eluent: petroleum ether:AcOEt 2:8) afforded 150 mg of 4b
1
(84%) as a colourless oil. [a]D + 33.2 (c 0.25, CHCl3); H NMR: d 4.85 (d, 1 H,
J = 7.6 Hz, H-1), 3.93–3.89 (m, 2 H, H-2, H-4), 3.59–3.55 (m, 2 H, H-3, H-5), 3.48
(dd, 1 H, J = 10.1, 6.6 Hz, H-6a), 3.31 (dd, 1 H, J = 10.1, 4.7 Hz, H-6b), 2.80 (bs, 1 H,
OH), 1.50 (bs, 2 H, OH).
Anal. Calcd for C31H29NO8: C, 64.50; H, 5.38; N, 2.58. Found: C, 64.52; H, 5.36;
N, 2.57.

Application of Site-Directed Lipase Mutants
641
Methyl 4,6-O-pentafluorobenzylidene-a-D-glucopyranoside (7a). The crude
material was purified by flash chromatography, eluting with pure AcOEt. Compound
7a (1.72 g) was obtained as a white solid (90%): mp 59–61°C, [a]D + 74.6 (c 0.5,
1
CHCl3); H NMR: d 5.88 (s, 1 H, H-benzylidene), 4.78 (d, 1 H, J = 3.9 Hz, H-1),
4.27 (dd, 1 H, J = 9.8, 4.8 Hz, H-6eq.), 3.92 (t, 1 H, J = 9.2 Hz, H-4), 3.84 (dt, 1 H,
J = 9.8, 4.8 Hz, H-5), 3.70 (t, 1 H, J = 9.8 Hz, H-6ax.), 3.50 (dd, 1 H, J = 9.8, 3.9
Hz, H-2), 3.48 (s, 3 H, OCH3), 3.46 (t, 1 H, J = 9.8 Hz, H-3), 2.9 (bs, 1 H, OH), 2.4
(bs, 1 H, OH).
Anal. Calcd for C14F5H13O6: C, 45.17; F, 25.52; H, 3.52. Found, C, 45.21; F,
25.55; H, 3.51.
Methyl 4,6-O-pentafluorobenzylidene-b-D-glucopyranoside (7b). Following
the general procedure previously described, 7b (1.67 g, 87%) was obtained as a white
solid, after purification by flash chromatography (eluent: petroleum ether:AcOEt 2:8):
1
mp 120–121°C, [a]D ꢀ 45.8 (c 0.5, CHCl3); H NMR d, 5.86 (s, 1 H, H-benzylidene),
4.32 (dd, 1 H, J = 9.9, 4.1 Hz, H-6eq), 4.30 (d, 1 H, J = 7.5 Hz, H-1), 3.80 (t, 1 H,
J = 9.9 Hz, H-3), 3.72 (t, 1 H, J = 9.9 Hz, H-6ax), 3.55 (s, 3 H, OCH3), 3.52–3.48 (m,
2 H, H-4 and H-5), 3.42 (dd, 1 H, J = 9.9, 7.5 Hz, H-2), 3.02 (bs, 2 H, OH).
Anal. Calcd for C14F5H13O6: C, 45.17; F, 25.52; H, 3.52. Found, C, 45.14; F,
25.54; H, 3.49.
Methyl 4,6-O-pentafluorobenzylidene-a-D-galactopyranoside (8a). Compound
8a (1.57 g, 82%) was obtained as a white solid after purification by flash chro-
matography (petroleum ether:AcOEt 2:8): mp 95–97°C, [a]D + 102.0 (c 0.65, CHCl3);
1H NMR d, 5.94 (s, 1 H, H-benzylidene), 4.98 (d, 1 H, J = 3.1 Hz, H-1), 4.30 (dd, 1 H,
J = 12.5, 1.5 Hz, H-6a), 4.25 (bd, 1 H, J = 2.2 Hz, H-4), 4.05 (dd, 1 H, J = 12.5, 1.5
Hz, H-6b), 3.94–3.92 (m, 2 H, H-2 and H-3), 3.72 (bd, 1 H, J = 7.5, 1.5 Hz, H-5), 3.47
(s, 3 H, OCH3), 2.50 (bs, 1 H, OH), 2.22 (bs, 1 H, OH).
Anal. Calcd for C14F5H13O6: C, 45.17; F, 25.52; H, 3.52. Found, C, 45.19; F,
25.50; H, 3.54.
Methyl 4,6-O-pentafluorobenzylidene-b-D galactopyranoside (8b). Compound
8b (1.74 g, 91%) was obtained as a white solid after purification by flash chro-
matography (petroleum ether:AcOEt 2:8): mp 108–110°C, [a]D ꢀ 20.2 (c 0.5, CHCl3);
1H NMR d, 5.96 (s, 1 H, H-benzylidene), 4.34(bd, 1 H, J = 12.5 Hz, H-6a), 4.22 (d, 1
H, J = 8.2 Hz, H-1), 4.19 (bd, 1 H, J = 2.6 Hz, H-4), 4.05 (bd, 1 H, J = 12.5 Hz, h-6b),
3.77 (t, 1 H, J = 8.2 Hz, H-2), 3.71 (dd, 1 H, J = 8.2, 2.6 Hz, H-3), 3.59 (s, 3 H,
OCH3), 3.51 (bs, 1 H, H-5), 2.81 (bs, 1 H, OH), 1.98 (bs, 1 H, OH).
Anal. Calcd for C14F5H13O6: C, 45.17; F, 25.52; H, 3.52. Found, C, 45.15; F,
25.48; H, 3.53.
Enzyme production. Lipase variants were produced through fermentation of P.
pastoris strain X-33 transformed with the expression plasmid pPICZaB (Invitrogen
Corporation, San Diego, CA) carrying the mutated lipase genes.[24] Cells were grown in
shaking flasks at 30°C in minimal medium containing 2% sorbitol, 1.34 g/L yeast
nitrogen base in phosphate buffer, pH 6.0. Cells were grown until a density of 108
cells/mL was reached, and then induced by daily addition of 1% v/v methanol. After 5

642
Cipolla et al.
days, induction cells were harvested by centrifugation at 11.000 rpm for 45 min. The
supernatant was filtered through 0.45 mm cut-off filters (Millipore) and concentrated by
tangential flow filtration against 4–5 volumes of 5 mM Tris-HCl pH 7.5 with a
Minitan system (Millipore Co., Bedford, MA, USA) using filter plates at 60 kDa
exclusion. Mutant enzymes were identified by electrophoresis on 10% SDS-poly-
acrylamide gel,[29] and quantified by comparison with standard BSA at known
concentrations. Determination of total proteins was according to Bradford.[30] Lipase
was quantified by densitometry analysis.
Molecular modelling. Calculations and graphical manipulations were carried out
on a Silicon Graphics Indigo workstation using the software package InsightII [Biosym/
MSI 9685 Scranton Road, San Diego, CA, USA]. Molecular mechanics and dynamics
calculations were performed using the Consistent Valence Force Field.[31]
The structure of the enzyme used in this study (CRL) was derived from the
Brookhaven Protein Data Bank. The potential energy surface has been scanned, with the
aim of locating low energy structures, using molecular dynamics and simulated
annealing approaches. Due to the extremely large number of atoms, some degrees of
freedom have been frozen during the simulations. In particular, final structures have
been obtained according to the following procedure. The structure of the starting
enzyme-substrate adduct was obtained by freezing all atoms but the substrate and
the amino acids directly interacting with it (residues 296, 297, 344, 345); then the
˚
optimization was carried out by unfreezing all residue within 8 A from the substrate and
finally performing 25 cycles of simulated annealing. Each simulated annealing cycle is
made by 25 picoseconds (ps) of molecular dynamics where the temperature is raised
from 50 to 500 K, 100 ps of molecular dynamics at 500 K and 150 ps of molecular
dynamics where the temperature is decreased to 50 K. The time step was set to 0.001 ps.
The structures obtained from the simulated annealing procedure have been optimized
and clustered according to structure similarity (root mean square displacement).
ACKNOWLEDGMENTS
This work was partly supported by a F.A.R grant (ex 60%) to M. L. , by the
Project COFIN 2002 of the Italian Ministry for Education, University and Research,
and by the European Community’s Human Potential Programme under contract HPRN-
CT-2002-00173, [GLYCIDIC SCAFFOLDS].
REFERENCES
1. Schmidt, R.D.; Verger, R. Lipases: interfacial enzymes with attractive applications.
Angew. Chem., Int. Ed. Engl. 1998, 37, 1608–1633.
2. Jaeger, K.-E.; Dijkstra, B.W.; Reetz, M.T. Bacterial biocatalysts: molecular biology,
three-dimensional structures, and biotechnological applications of lipases. Annu.
Rev. Microbiol. 1999, 53, 315–351.
3. Villeneuve, P.; Muderhwa, J.M.; Graille, J.; Haas, M.J. Customizing lipases for

Application of Site-Directed Lipase Mutants
643
biocatalysis: a survey of chemical, physical and molecular biological approaches. J.
Mol. Catal., B Enzym. 2000, 9, 113–148.
4. Rich, J.O.; Michels, P.C.; Khmelnitsky, Y.L. Combinatorial biocatalysis. Curr.
Opin. Chem. Biol. 2002, 6, 161–167.
5. Vulfson, E.N. Lipases, Their Structure, Biochemistry and Application; Woolley, P.,
Petersen, S.B., Eds.; Cambridge University Press, 1994; 271–287.
6. Schmid, A.; Dordick, J.; Hauer, B.; Kiener, A.; Wubbolts, M.; Witholt, B. Industrial
biocatalysis today and tomorrow. Nature 2001, 409, 258–268.
7. Koeller, K.M.; Wong, C.-H. Enzymes for chemical synthesis. Nature 2001, 409,
232–240.
8. Klibanov, A.M. Improving enzymes by using them in organic solvents. Nature
2001, 409, 241–246.
9. Arnold, F.H. Combinatorial and computational challenges for biocatalyst design.
Nature 2001, 409, 253–257.
10. Zhao, H.; Chockalingam, K.; Chen, Z. Directed evolution of enzymes and pathways
for industrial biocatalysis. Curr. Opin. Biotechnol. 2002, 13, 104–110.
11. Farinas, E.T.; Bulter, T.; Arnold, F.H. Directed enzyme evolution. Curr. Opin.
Biotechnol. 2001, 12, 545–551.
12. Jaeger, K.-E.; Reetz, M.T. Directed evolution of enantioselective enzymes for
organic chemistry. Curr. Opin. Chem. Biol. 2000, 4, 68–73.
13. Jaeger, K.-E.; Reetz, M.T. Microbial lipases form versatile tools for biotechnology.
Trends Biotechnol. 1998, 16, 396–403.
14. Magnusson, A.; Hult, K.; Holmquist, M. Creation of an enantioselective hydrolase
by engineered substrate-assisted catalysis. J. Am. Chem. Soc. 2001, 123, 4354–
4355.
15. Svendsen, A. Lipase protein engineering. Biochim. Biophys. Acta 2000, 1543,
223–238.
16. Kovac, A.; Scheib, H.; Pleiss, J.; Schmid, R.D.; Paltauf, F. Molecular basis of lipase
stereoselectivity. Eur. J. Lipid Sci. Technol. 2000, 102, 61–77.
17. Therisod, M.; Klibanov, A.M. Regioselective acylation of secondary hydroxyl
groups in sugars catalyzed by lipases in organic solvents. J. Am. Chem. Soc. 1987,
109, 3977–3981.
18. Henne, W.J.; Sweers, H.M.; Wang, Y.-F.; Wong, C.-H. Enzymes in carbohydrate
synthesis. Lipase-catalyzed selective acylation and deacylation of furanose and
pyranose derivatives. J. Org. Chem. 1988, 53, 4939–4945.
19. Therisod, M.; Klibanov, A.M. Facile enzymatic preparation of monoacylated sugars
in pyridine. J. Am. Chem. Soc. 1986, 108, 5638–5640.
20. Brocca, S.; Schmidt-Dannert, C.; Lotti, M.; Alberghina, L.; Schmid, R.D. Design,
total synthesis and functional overexpression of the Candida rugosa lip1 gene
coding for a major industrial lipase. Protein Sci. 1998, 7, 1415–1422.
21. Grochulski, P.; Li, Y.; Schrag, J.D.; Cygler, M. Two conformational states of
Candida rugosa lipase. Protein Sci. 1994, 3, 82–91.
22. Grochulski, P.; Bouthilier, F.; Kazlauskas, R.J.; Serreqi, A.N.; Schrag, J.D.;
Ziomek, E.; Cygler, M. Analogs of reaction intermediates identify a unique
substrate binding site in Candida rugosa lipase. Biochemistry 1994, 33, 3494–
3500.

644
Cipolla et al.
23. Manetti, F.; Mileto, D.; Corelli, F.; Soro, S.; Palocci, C.; Cernia, E.; D’Acquarica,
I.; Lotti, M.; Alberghina, L.; Botta, M. Design and realization of a tailor-made
enzyme to modify the molecular recognition of 2-arylpropionic esters by Candida
rugosa lipase. Biochim. Biophys. Acta 2000, 1543/1, 146–158.
24. Sagt, C.M.J.; Mueller, W.H.; Boonstra, J.; Verkleij, A.J.; Verrips, C.T. Impaired
secretion of a hydrophobic cutinase by Saccharomyces cerevisiae cells correlates
with an increased association with immunoglobulin heavy-chain binding protein
(BiP). Appl. Environ. Microbiol. 1998, 64, 316–324.
25. Brocca, S.; Secundo, F.; Ossola, M.; Alberghina, L.; Carrera, G.; Lotti, M.
Sequence of the lid affects activity and specificity of Candida rugosa lipase iso-
enzymes. Prot. Sci., in press.
26. Degueil-Castaing, M.; DeJesco, B.; Drouillard, S.; Maillard, B. Enzyme reactions
in organic synthesis: ester interchange of vinyl esters. Tetrahedron Lett. 1987, 28,
953–954.
27. McManus, D.A.; Vulfson, E.N. Substituent effects on the regioselectivity of
enzymatic acylation of 6-O-alkylglycopyranosides using Pseudomonas cepacia
lipase. Carbohydr. Res. 1995, 279, 281–291.
28. Panza, L.; Luisetti, M.; Crociati, E.; Riva, S. Selective acylation of 4,6-O-
benzylideneglycopyranosides by enzymatic catalysis. J. Carbohydr. Chem. 1993,
12, 125–130.
29. Laemmli, U.K. Cleavage of structural proteins during the assembly of the head of
the bacteriophage T4. Nature 1970, 227, 680–685.
30. Bradford, M. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem.
1976, 72, 248–254.
31. Douber-Osguthorpe, P.; Roberts, V.A.; Osguthorpe, D.J.; Wolff, J.; Genest, A.T.;
Hagler, A.T. Structure and energetics of ligand binding to proteins: E. coli
dihydrofolate reductase-trimethoprim, a drug-receptor system. Prot. Struct. Funct.
Genet. 1988, 4, 31.
Received March 4, 2003
Accepted July 30, 2003