(Trans)esterification of mannose catalyzed by lipase B from candida antarctica in an improved reaction medium using co-solvents and molecular sieve

Article abstract of DOI:10.1080/10826068.2011.622330

Four co-solvents (dimethylformamide [DMF], formamide, dimethyl sulfoxide [DMSO], and pyridine) were tested with tert-butanol (tBut) to optimize the initial rate (v₀) and yield of mannosyl myristate synthesis by esterification catalyzed by immobilized lipase B from Candida antarctica. Ten percent by volume of DMSO resulted in the best improvement of v₀ and 48-hr yield (respectively 115% and 13% relative gain compared to pure tBut). Use of molecular sieve (5% w/v) enhances the 48-hr yield (55% in tBut/DMSO [9:1, v/v]). Transesterification in tBut/DMSO (9:1, v/v) with vinyl myristate leads to further improvement of v₀ and 48-hr yield: a relative gain of 85% and 65%, respectively, without sieve and 25% and 10%, respectively, with sieve, compared to esterification. No difference in v₀ and 48-hr yield is observed when transesterification is carried out with or without sieve. Copyright Taylor & Francis Group, LLC.

Full text of DOI:10.1080/10826068.2011.622330

This article was downloaded by: [Fondren Library, Rice University ]
On: 22 April 2013, At: 09:57
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Preparative Biochemistry and
Biotechnology
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lpbb20
(TRANS)ESTERIFICATION OF MANNOSE
CATALYZED BY LIPASE B FROM Candida
antarctica IN AN IMPROVED REACTION
MEDIUM USING CO-SOLVENTS AND
MOLECULAR SIEVE
Katherine Nott a b , Alison Brognaux a b , Gaëtan Richard a b , Pascal
Laurent a b , Audrey Favrelle d , Christine Jérôme d , Christophe
Blecker c , Jean-Paul Wathelet b , Michel Paquot a & Magali Deleu a
a Department of Industrial Biological Chemistry, University of Liège,
Gembloux, Belgium
b Department of General and Organic Chemistry, University of Liège,
Gembloux, Belgium
c Department of Food Technology, University of Liège, Gembloux,
Belgium
d Center for Education and Research on Macromolecules, University
of Liège, Chemistry Institute, Liège, Belgium
Accepted author version posted online: 06 Feb 2012.Version of
record first published: 18 Jun 2012.
To cite this article: Katherine Nott , Alison Brognaux , Gaëtan Richard , Pascal Laurent , Audrey
Favrelle , Christine Jérôme , Christophe Blecker , Jean-Paul Wathelet , Michel Paquot & Magali Deleu
(2012): (TRANS)ESTERIFICATION OF MANNOSE CATALYZED BY LIPASE B FROM Candida antarctica IN AN
IMPROVED REACTION MEDIUM USING CO-SOLVENTS AND MOLECULAR SIEVE, Preparative Biochemistry
and Biotechnology, 42:4, 348-363
To link to this article: http://dx.doi.org/10.1080/10826068.2011.622330
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
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.

The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.

Preparative Biochemistry & Biotechnology, 42:348–363, 2012
Copyright # Taylor & Francis Group, LLC
ISSN: 1082-6068 print/1532-2297 online
DOI: 10.1080/10826068.2011.622330
(TRANS)ESTERIFICATION OF MANNOSE CATALYZED BY LIPASE B
FROM Candida antarctica IN AN IMPROVED REACTION MEDIUM
USING CO-SOLVENTS AND MOLECULAR SIEVE
Katherine Nott,1,2 Alison Brognaux,1,2 Gae¨tan Richard,1,2
4
4
Pascal Laurent,1,2 Audrey Favrelle, Christine Jerome,
´ ˆ
Christophe Blecker,3 Jean-Paul Wathelet,2 Michel Paquot,1
and Magali Deleu1
1Department of Industrial Biological Chemistry, University of Lie`ge, Gembloux, Belgium
2Department of General and Organic Chemistry, University of Lie`ge, Gembloux, Belgium
3Department of Food Technology, University of Lie`ge, Gembloux, Belgium
4Center for Education and Research on Macromolecules, University of Lie`ge, Chemistry
Institute, Lie`ge, Belgium
& Four co-solvents (dimethylformamide [DMF], formamide, dimethyl sulfoxide [DMSO], and
pyridine) were tested with tert-butanol (tBut) to optimize the initial rate (v0) and yield of mannosyl
myristate synthesis by esterification catalyzed by immobilized lipase B from Candida antarctica.
Ten percent by volume of DMSO resulted in the best improvement of v0 and 48-hr yield (respectively
115% and 13% relative gain compared to pure tBut). Use of molecular sieve (5% w=v) enhances
the 48-hr yield (55% in tBut=DMSO [9:1, v=v]). Transesterification in tBut=DMSO (9:1, v=v)
with vinyl myristate leads to further improvement of v0 and 48-hr yield: a relative gain of 85% and
65%, respectively, without sieve and 25% and 10%, respectively, with sieve, compared to esterifi-
cation. No difference in v0 and 48-hr yield is observed when transesterification is carried out with
or without sieve.
Keywords acylation, initial rate, myristic acid, novozym 435, sugar ester, vinyl ester
INTRODUCTION
Sugar esters are nonionic surfactants with multiple uses in the cosmetic,
food, and pharmaceutical industries. Their enzymatic synthesis is generally
preferred to chemical synthesis which is less selective toward the various
K. Nott, A. Brognaux, and Gaetan Richard have contributed equally to this work.
¨
Address correspondence to Magali Deleu, Department of Industrial Biological Chemistry,
University of Lie`ge, Gembloux Agro-Bio Tech (GxABT), Passage des De´porte´s 2, B-5030 Gembloux,
Belgium. E-mail: magali.deleu@ulg.ac.be

CAL-B Assisted (Trans)Esterification of Mannose
349
hydroxyl groups and requires fastidious protection and deprotection steps.
Due to the drastic conditions generally used (pH and high temperature),
chemical synthesis often leads to generation of side products, rendering
the recovery of the ester difficult. The range of substrates accessible to enzy-
matic synthesis is more restricted due to the selectivity of the enzymes.
Numerous articles report the use of lipases to catalyze the synthesis of sugar
esters.[1–3] Glucose is the main substrate exploited in these studies. Other
sugars such as mannose (Man) have an interest, mainly in the medical field.
Mannose-based vaccine or drug carrier can be interesting for increasing
immunogenicity[4,5] and tumor targeting.[6]
For glucose-based esters synthesis, co-solvents and molecular sieve are
frequently used to improve the product yield. In the case of mannose-based
esters, pure water-miscible solvents such as acetonitrile, acetone, 2-methyl-2-
propanol, and 2-methyl-2-butanol, without any co-solvent, are rather
used.[7–9]
In the present study, the influence of the presence of various co-solvents
and the effect of their proportion on the esterification of Man by myristic
acid (C14Ac), catalyzed by the immobilized lipase B of Candida antarctica, is
investigated. Dimethylformamide (DMF), formamide, dimethyl sulfoxide
(DMSO), or pyridine is added to tert-butanol (tBut) and their effect on
the initial rate (v0) and yield of the reaction is examined. The effect of
DMSO and pyridine on the sugar solubility and on the denaturation of
the lipase is also studied to better understand the impact of these
co-solvents on the reaction’s v0 and yield. The influence on these two para-
meters of the use of molecular sieve to remove water from the esterification
medium is also investigated. The esterification is compared to transesterifi-
cation of Man with vinyl myristate.
EXPERIMENTAL
Materials
Immobilized lipase B from Candida antarctica (Novozym 435) was kindly
supplied by Novozymes (Denmark). Its activity measured by hydrolysis of
p-nitrophenyl butyrate (pNPB test), is 69 U=g (1 unit corresponds to the
amount of enzyme required to hydrolyze 1 mmol of pNPB=min). D-(þ)-
Mannose (>99%), myristic acid (>99%), pyridine (99%), formamide
(99%), dimethylformamide (99.5%), dimethyl sulfoxide (99.5%), formic
˚
acid (FA, puriss p.a. for mass spectrometry), and molecular sieve 3 A
(8–12 mesh) were purchased from Sigma Aldrich (USA). Vinyl myristate
(>99%, stabilized with MEHQ) was obtained from TCI Europe. Chloro-
form (stabilised with ꢀ0.5% of ethanol), the HPLC-grade acetonitrile
(ACN), and methanol (MeOH) were from Scharlau (Spain). Liquid

350
K. Nott et al.
chromatography–mass spectroscopy (LC-MS) grade ACN and MeOH were
from Biosolve (Netherlands).
Optimization of the Sugar and Fatty Acid Concentrations for
the Enzymatic Esterification
The esterification reaction is shown in Figure 1a. tBut was added to
Man (apparent concentrations from 0.05 to 0.60 M) and C14Ac (0.25 M)
and the mixture was magnetically stirred and heated at 60ꢁC in a water
bath. After 30 min, the reaction was started by adding 20 mg of Novozym
435. The reactions were carried out for 48 hr. Aliquots were withdrawn over
time, diluted in ACN=MeOH (1:1, v=v), centrifuged (5 min at 13,000 rpm),
and analyzed by HPLC. In a second set of experiments, reactions were car-
ried out with the optimal Man concentration found earlier (0.10 M) and
C14Ac concentrations varying from 0.05 M to 1.20 M. The optimal fatty
acid concentration was determined. These optimal reagent concentrations
were established as the reference conditions for the esterification reaction
(0.10 M Man and 0.60 M C14Ac).
Enzymatic Synthesis of Mannosyl Myristate by
Transesterification
The procedure used was the same as described earlier for the reference
esterification reaction but with vinyl myristate (0.60 M) used as acyl donor
instead of C14Ac.
Influence of Co-Solvents on the Synthesis of Mannosyl
Myristate by (Trans)Esterification
The co-solvent (formamide, DMF, DMSO, or pyridine) percentage in
tBut varied from 0 (reference reaction) to 40% v=v. Reactions and analysis
FIGURE 1 Synthesis of mannosyl myristate catalyzed by immobilized lipase B from Candida antarctica
(Novozym 435) by (a) esterification or (b) transesterification.

CAL-B Assisted (Trans)Esterification of Mannose
351
were carried out in the same manner as described for the optimization of
the Man concentration.
Influence of the Co-Solvents on the Sugar Solubility and
Enzyme Activity
The solubility of Man in various solvent mixtures (tBut with different
percentages by volume of DMSO or pyridine) was determined using the
following procedure: solutions of 0.10 M Man (apparent concentration)
and 0.60 M C14Ac were magnetically stirred at 60ꢁC for 30 min. After cen-
trifugation, the supernatant was diluted five times with ACN=MeOH (1:1,
v=v), analyzed by HPLC–evaporative light-scattering detector (ELSD), and
quantified by external calibration.
To study the influence of the co-solvents on the enzyme activity, Novo-
zym 435 was submitted to pure tBut (blank) or various co-solvents (DMSO
or pyridine) percentages by volume in tBut for 5 hr. The enzyme was then
recovered by filtration, rinsed with tBut, dried, and used for esterification
under the reference conditions (in pure tBut). The v0 and 48-hr yield
obtained for Novozym 435 pretreated with the co-solvents were compared
to those obtained with the blank.
Influence of Molecular Sieve on the Synthesis of Mannosyl
Myristate by (Trans)Esterification
The (trans)esterification reactions were repeated twice under the refer-
ence conditions and in tBut with 10% v=v DMSO or 40% v=v pyriꢁdine with
˚
5% w=v of molecular sieve 3 A (preactivated by drying 4 hr at 250 C before
use). The amount of sieve used was more than 50 times the amount neces-
sary to absorb the water produced by the reaction if its yield was 100%.
High-Performance Liquid Chromatography (HPLC)
The HPLC analyses were performed on an Agilent Technologies 1200
series HPLC coupled to an evaporative light-scattering detector (ELSD).
The column, a Zorbax 300 SB C18 (3.5 mm, 4.6 ꢂ 150 mm, Agilent), was
thermostatted at 30ꢁC. The flow rate was 0.8 mL=min and a linear gradient
of milliQ water (0.1% FA) and ACN (0.1% FA) starting at 30% ACN and
increasing to 100% ACN in 5 min was used. Then 100% ACN was main-
tained during 5 min. The ELSD parameters were 40ꢁC and 3.5 bars N2.
Standards for external calibration consisted of commercial Man and puri-
fied mannosyl myristate (see later discussion). The HPLC-ELSD quantifi-
cation allowed determination of the reactions 48-hr yield (%) and v0

352
K. Nott et al.
(mMester
L
ꢃ1 hꢃ1 gꢃEn1z). v0 corresponds to the slope of the graph represent-
ing the ester concentration as a function of reaction time within the first
5 hr (or less if nonlinear). Esterification and transesterification reactions
were duplicated and determined the standard deviations (SD) were always
smaller than 10% and 3% for v0 and the yields, respectively.
LC-MS (negative mode) analyses of the reaction medium were per-
formed on an Agilent 1100 system coupled to a HCT mass spectrometer
(Bruker Daltonics, Bremen, Germany) equipped with an electrospray ioni-
zation (ESI) source. The sample was diluted in MeOH with 0.1% FA in order
to obtain a total concentration of 5 mg=mL. Analyses were performed on an
Agilent Zorbax Eclipse XDB-C18 column (150 mm ꢂ 2.1 mm; 3.5 mm). The
elution (flow of 0.2 mL=min) was achieved with a linear gradient of ACN
(0.1% FA) and milliQ water (0.1% FA) starting at 50% ACN and ending at
100% ACN in 20 min; 100% of ACN was maintained for 5 min. The ESI para-
meters were 40 psi (N2), 9 L=min (N2), and 365ꢁC. The ionic trap was set to
scan from m=z ¼ 160 to m=z ¼ 1000 with a target mass of 300 m=z. The trap
was emptied after 200 ms or as soon as a total ionic current of 100 000
was attained. Results obtained (retention time [min], observed m=z, attri-
bution): (1.7 min, 179 m=z, [Man-H]ꢃ); (11.7 min, 389 m=z and 435 m=z,
[mannosyl myristate–H]ꢃ and [mannosyl myristate þ HCOOH–H]ꢃ);
(19.3 min, 227 m=z, [C14Ac-H]ꢃ). The LC-MS analyses of the reference reac-
tion medium after 48 hr showed that only Man monoester was produced; no
peaks or ions were detected for higher esters such as di- or triesters.
Purification and Chemical Characterization of the Mannosyl
Myristate
After synthesis under the reference conditions, the ester was purified
˚
by flash chromatography on a silica gel (60 A=40–63 mm) column using a
mixture of chloroform=MeOH=water (65:15:1.5, v=v=v). The purity of the
collected fractions was checked by HPLC-ELSD.
MS and MS-MS spectra (negative mode) were acquired on the Bruker
HCT mass spectrometer (see earlier description): ESI parameters (10 psi
(N2), 4 L=min (N2), and 300ꢁC), respectively. The scan range was adjusted
to m=z 160–460 and the target mass was set to m=z 180, 228, and 390 for
Man, C14Ac, and mannosyl myristate, respectively. Solutions of mannose,
myristic acid, and mannosyl myristate at 5 mg=mL (diluted in water or
MeOH with 0.1% formic acid [FA]) were infused into the electrospray ioni-
zation (ESI) source at a flow rate of 240 mL=min. For C14Ac, no FA was
used. The fragmentation of mannosyl myristate was examined in the nega-
tive mode by recording of MS-MS experiments on ions at m=z 389 [M-H]ꢃ
and m=z 435 [M þHCOOH-H]ꢃ. The MS and MS-MS data (Table 1)

CAL-B Assisted (Trans)Esterification of Mannose
353
TABLE 1 ESI MS Results for C14Ac, Man, and Mannosyl Myristate and MS-MS Experiments for
Mannosyl Myristate
MS data, Molecule
C14Ac
Detected ions (m=z)
Attribution
227
455
179
225
389
435
779
[M-H]ꢃ
[2 M-H]ꢃ
Man
[M-H]ꢃ
[M þ HCOOH-H]ꢃ
Mannosyl myristate
[M-H]ꢃ
[M þ HCOOH-H]ꢃ
[2 M-H]
Mannosyl myristate MS-MS data, parent ion (m=z)
Fragment ions
Proposed fragmentation
389 [M-H]ꢃ
161
227
269
371
161
227
269
329
389
[M-C14H28O2-H]ꢃ
[C14H27O2]ꢃ
[M-C4H8O4-H]ꢃ
[M-H2O-H]ꢃ
435 [MþHCOOH-H]ꢃ
[M-C14H28O2-H]ꢃ
[C14H27O2]ꢃ
[M-C4H8O4-H]ꢃ
[M-C2H4O2-H]ꢃ
[M-H]
confirmed the molecular mass of the mannosyl myristate and also contrib-
uted to confirm its structure.
The infrared spectrum of the KBr pellet of the purified mannosyl myr-
istate was recorded on a Bruker IFS 25 spectrometer (Karlsruhe, Germany)
and revealed the expected characteristic bands (3200–3600 cmꢃ1 (-OH),
2850–2920 cmꢃ1 (-CH3 and -CH2), and 1735 cmꢃ1 (ester)). The nuclear
magnetic resonance (NMR) spectra (1H, 13C, 1H=1H COSY, HSQC, HMBC)
were recorded in a 6:1 (v=v) mixture of DMSO-d6=D2O at 600 MHz (1H)
and 150 MHz (13C) with a Varian instrument. Data are reported as follows:
chemical shift in ppm [multiplicity (bs: broad singlet, dd: double doublet,
t: triplet, m: multiplet), number of H, coupling constants in Hertz, attri-
bution].
1
6-O-Tetradecanoyl-D-mannose: H RMN (DMSO-d6=D2O) 4.81 (bs, 1H,
H-1), 4.21 and 3.96 (dd, 2H, J ¼ 11.4 Hz, J ¼ 6.6 Hz, H-6), 3.67 (t, 1H,
J ¼ 7.5 Hz, H-6), 3.52 (bs, 1H, H-2), 3.48 (dd, 1H, J ¼ 9 Hz, J ¼ 2.4 Hz,
H-3), 3.34 (t, 1H, J ¼ 9.6 Hz, H-4), 2.25 (t, 2H, J ¼ 7.2 Hz, H-20), 1.48 (m,
2H, H-30), 1.48 (m, 20H, H-40 to H-130), 0.82 (t, 3H, J ¼ 6.6 Hz, H-140) ;
13C RMN (DMSO-d6 = D2O) 173.4 (C-10), 94.5 (C-1), 71.9 (C-2), 70.8
(C-3), 70.8 (C-5), 67.6 (C-4), 64.6 (C-6), 33.8 (C-20), 24.9 (C-30), 22.5-33;9
(C-40 to C-130), 14.4 (C-140).
The relative configurations at C-1 (a and b anomers) were established
1
from their H and 13C chemical shifts and from the examination of their
1
1H multiplicity. The anomeric ratio (a=b) was determined by H-NMR,

354
K. Nott et al.
which showed that the a anomer is the most important (90%) in the con-
ditions of the analysis. The NMR analyses have demonstrated that only 6-
O-tetradecanoyl-D-mannose was obtained. Two spin systems were revealed
1
in the H=1H COSY spectrum: CH-1 to CH-6 and CH2-20 to CH3-140. The
connectivity between those two spin systems was established by key HMBC
correlation, the most noteworthy being between H2-6 at dH 4.21 and 3.96
and C-10(dC 173.4), and between H2-20 at dH 2.25 and C-6 (dC 64.6). Our
result is in accordance with many other studies that have demonstrated
the high regioselectivity of Candida antarctica lipase B for primary hydroxyl
group of osidic molecules.[10]
RESULTS AND DISCUSSION
Optimization of the Sugar and Fatty Acid Concentrations for
the Enzymatic Esterification
The solubility of Man in tBut containing 0.60 M C14Ac at 60ꢁC deter-
mined by HPLC-ELSD is 0.060 M (SD ¼ 10%, n ¼ 4). As Man is consumed
during the reaction, more may dissolve; the apparent Man concentration
range tested is 0.05–0.60 M. The v0 given in Figure 2, estimated thanks to
the HPLC-ELSD peak area obtained for the ester, shows that v0 is not sig-
nificantly influenced by the Man concentration. The area of the ester
observed after 48 hr is constant as of 0.10 M Man. That concentration
was chosen for the further experiments.
Figure 3 indicates that for the concentration range of C14Ac (0.10–1.20
M) the maximum v0 and ester concentration at 48 hr are attained at 0.60 M,
FIGURE 2 Optimization of the Man concentration for the esterification (0.25 M C14Ac, 0.2% w=v
ꢁ
&
Novozym 435, 60 C, means of two independent experiments): ( ) v0 and (ꢁ) 48-hr ELSD Areaester.

CAL-B Assisted (Trans)Esterification of Mannose
355
FIGURE 3 Optimization of the C14Ac concentration for the esterification (0.10 M Man, 0.2% w=v
ꢁ
&
Novozym 435, 60 C, means of two independent experiments): ( ) v0 and (ꢁ) 48-hr ELSD Areaester.
acid and this was chosen for the rest of the study. This corresponds to a
molar excess of six of the acyl donor compared to the acceptor.
Influence of Co-Solvent Addition on the V0 and Yield of the
Enzymatic Esterification
The choice of solvent is critical for the enzymatic synthesis of sugar fatty
esters. Indeed, the solvent must allow the lipase to maintain its catalytic
activity and must be able to dissolve both sugar and lipid, two substrates dif-
fering highly in polarity. A majority of authors claim that the activity of the
lipase is best correlated with the solvent’s log P,[11–13] but some have found
the best relation with its empirical Dimroth–Reichardt parameter (ET),[14]
[15]
others with the dielectric constant[7,8] or with the log Sw=o
,
and some
could not find any relation with any parameter. It thus seems that until
now no universal parameter has been found that allows selection of the best
solvent for a synthesis. Enzymatic catalyzed ester synthesis is often done in a
pure solvent. In this work, we aimed at optimizing the v0 and yield of the
esterification of Man by C14Ac catalyzed by Novozym 435 by adding a
co-solvent to tBut in which the lipase B from Candida antarctica shows good
catalytic activity and tBut is also too sterically hindered to be used as a sub-
strate by the enzyme. The four co-solvents (DMF, formamide, DMSO, and
pyridine) were chosen as they allow good solubilization of sugars but
generally cannot be used pure for enzymatic catalysis as they tend to inac-
tivate the lipase.[16]
Man esterification without co-solvent (reference reaction) gave an 48-hr
yield of 45% and a v0 of 1.7 mMester Lꢃ1 hꢃ1
g
ꢃ1
(Figures 4a and 4b). A
Enz
small addition (5% v=v) of any of the co-solvents, except pyridine, increased

356
K. Nott et al.
FIGURE 4 Influence of co-solvent percentage by volume of tBut on the v0 (a) and 48-hr yield (b) of the
esterification (0.10 M Man, 0.60 M myristic acid, 0.2% w=v Novozym 435, 60ꢁC, means of two inde-
&
~
pendent experiments): (ꢁ) DMF, ( ) DMSO, ( ) pyridine, and (^) formamide.
the v0 (Figure 4a). The 48-hr yield (Figure 4b) was slightly improved only
with 10% v=v DMSO (48-hr yield of 51%, relative gain of 13%). However,
added at high proportions, all of the co-solvents have a negative impact
on the v0 and yield. The v0 as a function of the % v=v of co-solvent added
(Figure 4a) presents a maximum at 5% v=v formamide, 10% v=v DMSO,
and between 10 and 30% v=v DMF. For these maxima, the relative v0 gains
are of 55% for formamide and 115% for DMSO and DMF, with DMSO
being the most interesting as the optimum is reached at only 10% v=v.
The v0 and 48-hr yield diminish with percentages of DMSO higher than
10% v=v and reach zero at 30% v=v. At 40% v=v of DMF, the v0 diminishes
by 40% compared to the reference. These results are in accordance with
the observations of Watanabe et al.,[8] who reported no synthesis of manno-
syl laurate by Novozym 435-catalyzed esterification of Man in pure DMSO
or DMF medium. Pyridine’s relative gain is the lowest of all the solvents
tested, while it was shown to be the most effective of the four for the
Novozym 435-catalyzed synthesis of myristyl glucuronate from myristyl alco-
hol and glucuronic acid in tBut (unpublished results). The HPLC-ELSD
chromatogram showed no formation of higher (di-, tri, etc.) esters, which-
ever the two-solvent mixture used.
Influence of the Co-Solvents on the Mannose Solubility and the
Enzyme Denaturation
To better understand the effect of pyridine and DMSO on v0 and yield,
their impact on the solubility of Man and on the denaturation of Novozym
435 was studied. The dissolved Man concentration was measured for tBut

CAL-B Assisted (Trans)Esterification of Mannose
357
containing 0.60 M of C14Ac and various percentages (0 to 40% v=v) of
DMSO or pyridine (Figure 5). The solubility increased with the percentage
of co-solvent and reached a constant maximum value from 15% to 40% v=v
of co-solvent corresponding to solubilization of approximately 85% and
75% of the Man introduced in the medium, respectively, for DMSO and
pyridine. The v0 of the mannosyl ester production was already optimal at
10% v=v of DMSO, slightly lower than the 15% v=v minimum DMSO per-
centage allowing the maximum solubilization of the sugar. Pyridine also
improved the solubility of the Man in the medium, but, contrary to DMSO,
it had no positive effect on the reaction’s v0. This confirms that the sugar
solubility is not the only factor that explains the positive effect of the
co-solvents.
The possible denaturing effect of the co-solvents on Novozym 435 was
studied. For all the percentages by volume of pyridine, no significant effect
of the preincubation of Novozym 435 was observed on the v0 and the yield
(results not shown). This indicates that pyridine does not denature the
Candida antarctica lipase B. The results are in accordance with Degn and
Zimmermann,[12] who report the use of Novozyme 435 in reaction medium
consisting of tBut and pyridine in proportions 55:45 (v=v). On the contrary,
the preincubation with tBut=DMSO (Figure 6) led to a decrease of both
48-hr yield and v0. The higher the DMSO percentage is, the lower is the
enzyme efficiency. At 30% v=v and above, the catalytic activity is comple-
tely lost (no ester synthesized after 48 hr). This is probably due to the
FIGURE 5 Influence of co-solvent percentage by volume of tBut on the Man solubility (0.10 M Man
ꢁ
&
)
apparent concentration, 0.60 M C14Ac, 60 C, 30 min, means of two independent experiments): (
~
DMSO and ( ) pyridine.

358
K. Nott et al.
FIGURE 6 Influence on v0 (black) and 48-hr yields (white) of the incubation of Novozym 435 in mix-
tures of tBut=DMSO before the esterification performed under the reference conditions (0.10 M Man,
0.60 M C14Ac, 0.2% w=v Novozym 435, 60ꢁC, means of two independent experiments).
inactivation of the lipase by DMSO, which can cause unfolding of proteins
and disrupt the hydration shell of the enzyme.[17]
Influence of Molecular Sieve on the Enzymatic Esterification
The reactants, the enzyme, and the solvents initially contain a small
amount of water, which is also a by-product of the esterification. Water will
affect the equilibrium position of the reaction and its elimination must
favor synthesis rather than hydrolysis. Many methods have been employed
to remove water, such as synthesizing under reduced pressure, under
reflux, in membrane reactors, in solvents forming low-boiling-point azeo-
tropes with water, and with use of desiccants. In this work, desiccants were
chosen and a molecular sieve used, as it has already been studied by many
researchers for lipase catalyzed synthesis,[8,12,18] and as Cauglia and
Canepa[19] have shown, it gives better results than CaSO4, CaCl2, and
MgSO4 for Novozym 435-catalyzed esterification of glucose.
Blanks with molecular sieve but no lipase showed that the desiccant
used was not able to catalyze the esterification of Man by C14AC in the
three media tested (tBut, tBut=DMSO 9:1 v=v, tBut=pyridine 6:4 v=v), as
no peak corresponding to the ester was detected by HPLC-ELSD (data
not shown). Furthermore, the addition of the sieve to the enzyme did
not lead to a qualitative change of the HPLC-ELSD chromatogram of the
reaction media whatever the time of synthesis, and no higher esters were
detected. Recently, some researchers have found entirely different results
with the same enzyme whilst studying the transesterification with vinyl laur-
ate of fructo-oligosaccharides.[20] They used a 4-A molecular sieve, and
˚

CAL-B Assisted (Trans)Esterification of Mannose
359
while they observed only monosubstitution in the absence of the sieve, they
detected many multiple substituted products when using the desiccant
alone. Furthermore, the product distribution depended on the percentage
of DMSO in the tBut. They hypothesized that the catalytic activity was due
to the strong acidic sites present on the zeolites constituting the sieve.
Although the acidic sites catalytic activity is low, products can be obtained
as high temperature and long reaction times were used.
Figure 7a shows that the v0 of esterification is not influenced by the
presence of molecular sieve except for the reaction medium containing
10% v=v of DMSO as co-solvent, for which a 50% v0 increase is observed with
the sieve. We can assume that at the beginning of the reaction, the amount
of water produced (that would be trapped by the molecular sieve) is too
small to have an influence on v0. With DMSO, the increase is probably
due to the faster reaction (producing more water) or to the higher initial
content of free water. The 48-hr yield for the esterification (Figure 7b) is
enhanced by the use of molecular sieve (ꢀ55% in tBut=DMSO 9:1 v=v
and in tBut=pyridine 6:4 v=v and ꢀ80% in pure tBut). It traps the water pro-
duced and displaces the equilibrium of the reaction toward synthesis rather
than hydrolysis.
Comparison of Esterification and Transesterification of
Mannose for the Mannosyl Myristate Synthesis
Lipase can catalyze ester synthesis by esterification or transesterifica-
tion. The latter avoids the major drawback of esterification, namely, the
FIGURE 7 Influence of molecular sieve on the v0 and 48-hr yield of the (trans)esterification (0.10 M
Man, 0.60 M C14Ac or vinyl myristate, 0.2% w=v Novozym 435, 60ꢁC, means of two independent experi-
˚
ments): esterification (black: without sieve; white: with 5% w=v molecular sieve 3 A); transesterification
(squared: without sieve).

360
K. Nott et al.
production of water, which causes ester hydrolysis. In this study, it was
decided to compare the results obtained by esterification with C14Ac to
those obtained by transesterification with vinyl myristate. A vinyl ester was
preferred to alkyl (methyl, ethyl) esters, as Ferrer et al.[21] demonstrated
that the rate of transesterification with the former was about 20 to 100 times
faster than with the alkyl esters. Vinyl esters release vinyl alcohol, which is
nearly irreversibly converted by tautomerization into acetaldehyde
(Figure 1b). Some lipases are inactivated by acetaldehyde, which plays
the role of alkylating agent in Maillard type reactions leading to Schiff
bases. Candida antarctica lipase B is remarkably stable to this product.[22]
In pure tBut without a molecular sieve, the transesterification leads to
better results than the esterification with C14Ac, with the former leading
to respectively 110% and 85% higher v0 and 48-hr yield (Figures 7a and
7b). This can be due to the water produced during esterification and to
the fact that the tautomerization of the vinyl alcohol to acetaldehyde favors
the ester production for the transesterification reaction. Figure 8 shows
that the influence of the DMSO percentage on the transesterification v0
and 48-hr yield follows the same trend as for esterification, with the opti-
mum also being at 10% of DMSO. At that percentage, the transesterifica-
tion still gives better results than the esterification but to a slightly
smaller extent than in pure tBut: the v0 and 48-hr yield are respectively
enhanced by 85% and 65%.
The transesterification was also carried out with 5% w=v of molecular
˚
sieve 3 A in pure tBut or with 10% v=v of DMSO and contrary to the results
for the esterification of Man with C14Ac, the desiccant has no significant
FIGURE 8 Influence of DMSO percentage by volume of tBut on the v0 and 48-hr yield of the transes-
terification (0.10 M Man, 0.60 M vinyl myristate, 0.2% w=v Novozym 435, 60ꢁC, means of two inde-
&
pendent experiments): (ꢁ) v0 and ( ) 48-hr yield.

CAL-B Assisted (Trans)Esterification of Mannose
361
effect on v0 and 48-hr yield (results not shown). This can be explained by
the fact that no water is produced during the reaction and that the water
initially present in the media does not cause any significant sugar ester
hydrolysis.
With 5% w=v of molecular sieve in pure tBut, transesterification leads to
respectively 80% higher v0 and to the same 48-hr yield as those obtained by
esterification. With 5% w=v of molecular sieve in tBut=DMSO (9:1, v=v), the
transesterification leads to respectively 25% and 10% higher v0 and 48-hr
yield than those obtained by esterification. In both cases, the transesterifica-
tion v0 is greater than that of the esterification but to a lesser extent than
without the sieve. The effect on the 48-hr yield is very small and is much
less marked than in the absence of sieve.
CONCLUSIONS
Of the four co-solvents (DMF, formamide, DMSO, and pyridine) tested
in combination with tBut for mannosyl myristate synthesis by esterification
catalyzed by immobilized lipase B from Candida antarctica, DMSO allowed
the best improvement of v0 and 48-hr yield (respectively 115% and 13%
relative gain compared to the reference), and pyridine never significantly
improved the v0 and even had a negative effect on 48-hr yield with percen-
tages as low as 5% v=v. Solubility measurements demonstrated that both
DMSO and pyridine help to solubilize Man. Experiments also showed that
DMSO has a denaturing effect on Novozyme 435, and as a consequence, its
percentage in the medium must be kept relatively low (<15% v=v). Con-
trary to DMSO, pyridine does not denature the enzyme, but it does not
improve the reaction’s v0 and 48-hr yield. One hypothesis would be that
pyridine could deprotonate the carboxylic function of myristic acid and
thus reduce its availability. By trapping the water, use of molecular sieve
in the esterification medium allows enhancement of the 48-hr yield (55%
in tBut=DMSO [9:1, v=v]) by displacing the equilibrium of the reaction
toward synthesis rather than hydrolysis. Mannosyl myristate was also
obtained by transesterification with vinyl myristate. Without a molecular
sieve, the transesterification leads to higher v0 and yield than the esterifi-
cation (no water production and tautomerization of the vinyl alcohol to
acetaldehyde), but it requires an additional step for the vinyl ester pro-
duction and increases the cost of the reagent compared to the fatty acid.
This step can be avoided by using a molecular sieve and the solvent mixture
(tBut=DMSO 9:1 v=v), as the esterification under these conditions gives v0
and yield close to those of the transesterification. However, this option also
presents disadvantages. For example, stirring of the medium becomes more
difficult and the mass transfer can be limited. The use of the sieve can also

362
K. Nott et al.
require bigger reactors, and an excess of sieve may remove the water in the
enzyme vicinity essential for its activity.[23] Other authors have reported a
change in the selectivity of the reaction,[20] and some have noted the possi-
bility of adsorption and degradation of their ester (6-unsaturated acyl-L-
ascorbates) by the sieve.[24] In the future, it will be interesting to optimize
the concentration of the sieve and to study its effect on the mannosyl
myristate (adsorption, degradation and ease of purification).
ACKNOWLEDGMENTS
The research was funded through an ARC grant, financed by the
French Community of Belgium, which is gratefully acknowledged for its
financial support. The authors thank Novozymes, which provided the
Novozyme 435, Thomas Bertrand for his technical support, and Antoine
Debuigne for his scientific support. Magali Deleu thanks the Fonds
National de la Recherche Scientifique (FNRS) for her Research Associate
position.
REFERENCES
1. Chang, S.W.; Shaw, J.F. Biocatalysis for the Production of Carbohydrate Esters. New Biotechnol. 2009,
26(3=4), 109–116.
2. Gandhi, N. Applications of Lipases. J. AOCS 1997, 74(6), 621–634.
3. Piccicuto, S.; Blecker, C.; Brohe´e, J.-C.; Mbampara, A.; Logany, G.; Deroanne, C.; Paquot, M.;
Marlier, M. Les esters de sucres: voies de synthe`se et potentialite´s d’utilisation [Sugar esters: Syn-
thesis routes and potential applications]. Biotechnol. Agron. Soc. Environ. 2001, 5(4), 209–219.
4. Jiang, H.L.; Kang, M.L.; Quan, J.S.; Kang, S.G.; Akaike, T.; Yoo, H.S.; Cho, C.S. The Potential of
Mannosylated Chitosan Microspheres to Target Macrophage Mannose Receptors in an
Adjuvant-Delivery System for Intranasal Immunization. Biomaterials 2008, 29, 1931–1939.
5. Zhou, X.; Liu, B.; Yu, X.; Zha, X.; Zhang, X.; Wang, X.; Chen, Yu; Chen, Yan; Chen, Yue; Shan, Y.;
Jin, Y.; Wu, Y.; Liu, J.; Kong, W.; Shen, J. Enhance Immune Response to DNA Vaccine Based on a
Novel Multicomponent Supramolecular Assembly. Biomaterials 2007, 28, 4684–4692.
6. Prakash, J.; Beljaars, C.; Harapanahalli, A.K.; Zeinstra-Smith, H.; de Jager-Krikken, A.; Hessing, M.;
Steen, H.; Poelstra, K. Tumor-Targeted Intracellular Delivery of Anticancer Drugs Through the
Mannose-6-Phosphate=Insulin-Like Growth Factor II Receptor. Int. J. Cancer 2010, 126, 1966–1981.
7. Adachi, S.; Kobayashi, T. Synthesis of Esters by Immobilized-Lipase-Catalyzed Condensation Reac-
tion of Sugars and Fatty Acids in Water-Miscible Organic Solvents. J. Biosci. Bioeng. 2005, 99(2),
87–94.
8. Watanabe, Y.; Miyawaki, Y.; Adachi, S.; Nakanishi, K.; Matsuno, R. Equilibrium Constant for Lipase-
Catalyzed Condensation of Mannose and Lauric Acid in Water-Miscible Organic Solvents. Enzyme
Microbial Technol. 2001, 29, 494–498.
9. Watanabe, Y.; Miyawaki, Y.; Adachi, S.; Nakanishi, K.; Matsuno, R. Synthesis of Lauroyl Saccharides
Through Lipase-Catalyzed Condensation in Microaqueous Water-Miscible Solvents. J. Mol. Catal. B
Enzymatic 2000, 10, 241–247.
10. Bousquet, M.-P.; Willemot, R.-M.; Monsan, P.; Bourres, E. Enzymatic Synthesis of Unsaturated Fatty
Acid Glucoside Esters for Dermo-Cosmetic Applications. Biotechnol. Bioeng. 1999, 63, 730–736.
11. Pederson, N.R.; Wimmer, R.; Emmersen, J.; Degn, P.; Pedersen, L.H. Effect of Fatty Acid Chain
Length on Initial Rates and Regioselectivity of Lipase-Catalysed Esterification of Dicaccharides.
Carbohyd. Res. 2002, 337, 1179–1184.

CAL-B Assisted (Trans)Esterification of Mannose
363
12. Degn, P.; Zimmermann, W. Optimization of Carbohydrate Fatty Acid Ester Synthesis in Organic
Media by a Lipase From Candida antarctica. Biotechnol. Bioeng. 2001, 74(6), 483–491.
13. Cao, L.; Bornscheuer, U.; Schmid, R.D. Lipase-Catalyzed Solid-Phase Synthesis of Sugar Esters.
Influence of Immobilization on Productivity and Stability of the Enzyme. J. Mol. Catal. B Enzymatic
1999, 6(3), 279–285.
14. Castillo, E.; Pezzotti, F.; Navarro, A.; Lopez-Munguia, A. Lipase-Catalyzed Synthesis of Xylitol
Monoesters: Solvent Engineering Approach. J. Biotechnol. 2003, 102, 251–259.
15. Valivety, R.H.; Johnston, G.A.; Suckling, C.J.; Halling, P.J. Solvent Effects on Biocatalysis in Organic
Systems: Equilibrium Position and Rates of Lipase Catalyzed Esterification. Biotechnol. Bioeng. 1991,
38, 1137–1143.
16. Walsh, M.K.; Bombyk, R.A.; Wagh, A.; Bingham, A.; Berreau, L.M. Synthesis of Lactose Monolaurate
as Influenced by Various Lipases and Solvents. J. Mol. Catal. B Enzymatic 2009, 60, 171–177.
17. Plou, F.J.; Cruces, M.A.; Ferrer, M.; Fuentes, G.; Pastor, E.; Bernabe´, M.; Christensen, M.; Comelles,
F.; Parra, J.L.; Ballesteros, A. Enzymatic Acylation of Di- and Trisaccharides With Fatty Acids: Choos-
ing the Appropriate Enzyme, Support and Solvent. J. Biotechnol. 2002, 96, 55–66.
18. Moreau, B.; Lognay, G.C.; Blecker, C.; Brohe´e, J.-C.; Che´ry, F.; Rollin, P.; Paquot, M.; Marlier, M.
Synthesis of Novel D-Glucuronic Acid Fatty Esters Using Candida antartica Lipase in Tert-Butanol.
Biotechnol. Lett. 2004, 26, 419–424.
19. Cauglia, F.; Canepa, P. The Enzymatic Synthesis of Glucosylmyristate as a Reaction Model for
General Considerations on ‘Sugar Esters’ Production. Bioresource Technol. 2008, 99, 4065–4072.
20. ter Haar, R.; Schols, H.A.; van den Broek, L.A.M.; Saglam, D.; Frissen, A.E.; Boeriu, C.G.; Gruppen,
H. Molecular Sieves Provoke Multiple Substitutions in the Enzymatic Synthesis of Fructose
Oligosaccharide-Lauryl Esters. J. Mol. Catal. B Enzymatic 2010, 62, 183–189.
21. Ferrer, M.; Cruces, M.A.; Bernabe´, M.; Ballesteros, A.; Plou, F. Lipase-Catalyzed Regioselective Acyla-
tion of Sucrose in Two-Solvents Mixtures. Biotechnol. Bioeng. 1999, 65, 10–16.
22. Weber, H.K.; Zuegg, J.; Faber, K.; Pleissb, J. Molecular Reasons for Lipase-Sensitivity against Acet-
aldehyde. J. Mol. Catal. B Enzymatic 1997, 3, 131–138.
23. Sabeder, S.; Habulin, M.; Knez, Z. The Lipase-Catalysed Synthesis of Fatty Acid Fructose Esters in
Organic Media and in Supercritical Carbon Dioxide. Chem. Ind. Chem. Eng. Q. 2006, 12, 147–151.
24. Kuwabara, K.; Watanabe, Y.; Adachi, S.; Nakanishi, K.; Matsuno, R. Synthesis of 6-O-Unsaturated
Acyl-Ascorbates by Immobilized Lipase in Acetone in the Presence of Molecular Sieve. Biochem.
Eng. J. 2003, 16, 17–22.