Regio- and stereoselective lipase-catalysed acylation of methyl α-D-glycopyranosides with fluorinated β-lactams

Article abstract of DOI:10.1002/ejoc.201402800

Burkholderia cepacia lipase (lipase PS-D) catalysed acylation with 3,3-difluoro-4-phenyl-, -thiophen-3-yl- and -4-pyridylazetidin-2-ones was examined for the formation of N-Boc-protected 6-O-acylated sugar-β-amino acid conjugates from methyl α-D-galacto-, -gluco- and mannopyranosides and Boc₂O. The 6-O-acylated glycopyranoside-β-amino acid conjugates were isolated and characterized. The low solubility of the gluco- and mannopyranosides and the high reactivity of the pyridylazetidinone restricted product formation. Activation of the β-lactam ring by the presence of fluorine substituents was shown to be necessary for the enzymatic acylation reaction. The (S)-enantiomers of the racemic β-lactam substrates reacted with the sugars. The lipase-catalysed diastereoselective ring-opening of aromatic and hetero-aromatic β-lactams was explored for the formation of β-amino acid conjugates of methyl α-D-glycopyranosides. Potential hydrolytic side-reactions were minimized by a suitable choice of solvent, lipase preparation and the use of molecular sieves.

Full text of DOI:10.1002/ejoc.201402800

FULL PAPER
DOI: 10.1002/ejoc.201402800
Regio- and Stereoselective Lipase-Catalysed Acylation of Methyl
α- -Glycopyranosides with Fluorinated β-Lactams
D
Riku Sundell,[a] Elina Siirola,[a] and Liisa T. Kanerva*[a]
Keywords: Enzyme catalysis / Diastereoselectivity / Acylation / Carbohydrates / β-Lactams / Fluorine
Burkholderia cepacia lipase (lipase PS-D) catalysed acylation
with 3,3-difluoro-4-phenyl-, -thiophen-3-yl- and -4-pyridyl-
azetidin-2-ones was examined for the formation of N-Boc-
protected 6-O-acylated sugar–β-amino acid conjugates from
methyl α-D-galacto-, -gluco- and mannopyranosides and
Boc2O. The 6-O-acylated glycopyranoside–β-amino acid con-
of the gluco- and mannopyranosides and the high reactivity
of the pyridylazetidinone restricted product formation. Acti-
vation of the β-lactam ring by the presence of fluorine sub-
stituents was shown to be necessary for the enzymatic acyl-
ation reaction. The (S)-enantiomers of the racemic β-lactam
substrates reacted with the sugars.
jugates were isolated and characterized. The low solubility
been reported to be important for effective amide cleavages;
lipases (and esterases) lack this stabilization.[6] That lipases
cleave the amide bonds of β-lactams can be explained by
the fact that the amide functional group lacks the resonance
stabilization that typically stabilizes normal amide bonds.
A β-lactam serves as an irreversible acyl donor for lipase-
catalysed acylation reactions. In mechanistic terms, the ser-
ine residue of the catalytic triad at the active site of the
lipase enzyme first opens the β-lactam ring, leading to the
formation of an ester intermediate (an acyl–enzyme inter-
mediate). Typically, the formation of the intermediate
should release the first reaction product (glycerol in the nat-
ural reaction). However, with a β-lactam substrate, no prod-
uct is released in this mechanistic step. Reaction of the in-
termediate with a suitable nucleophile (NuH) then gives the
β-amino acid or its derivative as the only reaction product
(Scheme 1). For reactions in organic solvents, Pseudomonas
sp. lipases (lipase AK and P-30) were first reported to cata-
lyse the ring-opening of a N-benzoyl-activated β-lactam
with methanol in tert-butyl methyl ether (TBME).[7a] Later,
the value of lipase catalysis for the production of individual
β-amino acid enantiomers from racemic β-lactams indi-
cated that running the reaction in diisopropyl ether (DIPE)
with just traces of added water could give effective ring-
opening with Candida antarctica lipase B (CAL-B as a
Novozym 435 preparation).[7b] Studies with mixtures of a
4-benzyl-β-lactam, methanol, and Burkholderia cepacia
lipase (in the form of the lipase PS-D preparation) in dry
organic solvents revealed potential side-reactions resulting
from the presence of the so-called residual water in the en-
zyme preparation. These reactions led to the hydrolysis
both of the β-lactam ring itself (Scheme 1, route B) and of
the methyl ester (Scheme 1, route D) produced by ring-
opening by methanol (Scheme 1, route A).[7c] Accordingly,
studies where hydrolytic side-reactions are addressed and
Introduction
A four-membered β-lactam (azetidin-2-one) ring is a
structural element of natural products and a precursor of
β-amino acids, many of which are biologically and pharma-
cologically active compounds.[1,2] The ring-opening of β-
lactam antibiotics by β-lactamase-catalysed mechanisms
represents a threat to society in the form of growing bacte-
rial resistance, similar ring-opening reactions in the hands
of a chemist lead to intriguing synthetic possibilities.[2] En-
antiomeric and diastereomeric purity is generally expected
of pharmaceutically active ingredients. Enzymes are re-
markably effective among all known catalysts in the degree
of their acceleration of chemical reactions and their selectiv-
ity. Serine hydrolases such as lipases (EC 3.1.1.3) are valu-
able catalysts. Lipases accept a wide range of substrates in
the different types of reactions they catalyse (e.g., hydroly-
sis, alcoholysis, aminolysis, interesterification and Michael
addition),[1,3] and they also show a high stability and diver-
sity. They have also been used as cheap regio- and enantio-
selective ring-opening catalysts of β-lactams in the prepara-
tion of β-amino acids[1] and their esters[4a] and amides,
which could then be used, for example, in the synthesis of
β-dipeptides.[4b,4c] Instances of the catalytic cleavage of
amides by lipases are rare, and are mostly restricted to Can-
dida antarctica lipases; a number of serine proteases readily
fulfil this function.[5] The orientation of the nitrogen, which
can stabilize the transition state of serine proteases, has
[a] Institute of Biomedicine/Department of Pharmacology, Drug
Development and Therapeutics/Laboratory of Synthetic Drug
Chemistry, University of Turku,
20014 Turku, Finland
E-mail: lkanerva@utu.fi
http://www.utu.fi/en/units/med/units/pharmacology/Pages/
home.aspx
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201402800.
Eur. J. Org. Chem. 2014, 6753–6760
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6753

R. Sundell, E. Siirola, L. T. Kanerva
FULL PAPER
controlled are important for the enantioselective ring-open- tures rac-1a–1c were chosen as starting materials as it had
ing of β-lactams by nucleophiles other than water.
previously been shown that an electron-withdrawing fluor-
ine substituent favoured the ring-opening reaction by meth-
anol.[4a] Moreover, the presence of fluorine (an isostere of
hydrogen) may introduce various positive effects to biolo-
gically active molecules, such as increased metabolic sta-
bility, target binding, and lipophilicity.[9] β-Lactam rac-1d
was studied as a hydrogen-containing analogue of rac-1a.
The lipase-catalysed ring-opening of rac-1a–1c and that of
rac-1d with a primary alcohol group at the C-6 position of
the sugars was planned to be the regio- and stereoselective
step of the process. It was anticipated that the S (R for rac-
1d) enantiomer of the β-lactam would react selectively in
the ring-opening reaction, as observed in previous stud-
ies.[4,10] This selectivity was confirmed by testing enantio-
pure (R)- and (S)-1b in the enzymatic acylation of 2. For
reasons of cost, racemic β-lactams rather than pure
enantiomers were used as acyl donors in the acylation of
the unprotected methyl glycopyranosides. It is clear that the
use of pure enantiomers for acylation would have the ad-
vantage of ensuring the diastereomeric purity of the sugar
derivatives prepared.
Scheme 1. Possible reaction routes for the lipase-catalysed ring-
opening of β-lactams: acylation with nucleophile (NuH) by ring-
opening of the β-lactam (A), hydrolysis of the β-lactam (B), acyl-
ation with NuH by reaction with the free acid (C) and hydrolysis
of formed sugar–β-amino acid conjugate (D).
As a continuation of our work towards the chemoenzym-
atic preparation of β-amino acids[1a] and sugar deriva-
tives,[8] we aimed to combine our knowledge of these areas
for the preparation of sugar–β-amino acid conjugates.
Thus, we studied the regio- and stereoselective chemo-enzy-
matic formation by lipase catalysis of sugar esters 5–7,
starting from racemic β-lactams rac-1a–1d (for which the
electron-withdrawing ability of the substituents increases in
the order phenyl Ͻ thiophen-3-yl Ͻ pyridin-4-yl) and un-
protected methyl α-d-galacto-, α-d-gluco- and α-d-manno-
pyranosides (2–4, respectively; Scheme 2). β-Lactam struc-
Results and Discussion
Synthesis of β-Lactams
A slightly modified Reformatsky-type addition was used
to prepare racemic 3,3-difluoro-4-arylazetidin-2-ones (rac-
1a–1c).[4a,11] N-4-Methoxybenzyl-protected β-lactams rac-
13a–13c were obtained by heating the corresponding 4-
methoxybenzyl aldimines (i.e., 12a–12c) with ethyl bromo-
difluoroacetate (2 equiv.) in the presence of Zn dust in re-
fluxing 2-methyltetrahydrofuran (2-MeTHF) or tetra-
hydrofuran (THF). In the case of rac-13c, the isolated prod-
uct was an inseparable mixture of the β-lactam and the cor-
1
responding β-amino ester (in a 1:3 ratio, according to H
NMR spectroscopy). N-Deprotection by treatment with ce-
ric ammonium nitrate (CAN; 2 equiv.) gave rac-1a–1c. The
synthesis of 4-phenylazetidin-2-one (rac-1d) was ac-
complished by the cycloaddition of chlorosulfonyl isocyan-
ate (CSI) to styrene also by a known method.[4a,12] The syn-
thetic details are given in the Supporting Information
(Scheme S1).
For the preparation of the individual β-lactam enantio-
mers, rac-1b was first hydroxymethylated with paraformal-
dehyde under ultrasound conditions (Scheme S2, Support-
ing Information).[10] Thereafter, the lipase-catalysed enan-
tioselective O-acylation of N-hydroxymethylated rac-1b
with vinyl butanoate and lipase PS-D in toluene was per-
formed followed by the separation of the unreacted (R)-
alcohol (Ͼ99% ee) and the (S)-butanoate ester (74% ee).
Subsequent enantiomeric purification of the butanoate
product by enzymatic alcoholysis with 1-butanol in DIPE
allowed the preparation of N-hydroxymethylated (S)-1b.
Deprotection with KMnO4 gave the (R)- and (S)-1b
enantiomers with 99 and 98% ee, respectively. The same
Scheme 2. Chemoenzymatic route to sugar amino acid esters 5–7.
6754
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 6753–6760

Lipase-Catalysed Acylation of Methyl α-d-Glycopyranosides
routine procedure could also be used for the preparation of rac-1a had occurred after 24 h (62% in TBME, 45% in 2-
other (R)- and (S)-lactams, but this was not done.
MeTHF, and 35% in tAmOH). This shows that there is the
potential for substantial hydrolysis in all three solvents. The
hydrolysis tendency with CAL-B (Novozym 435) was
higher than with lipase PS-D, apparently because the
hydrophobic methacrylate carrier of Novozym 435 gives up
Optimization of Sugar–β-Amino Acid Conjugate Formation
The use of carbohydrates for the ring-opening of a β- the bound water more readily than the Celite carrier of li-
lactam by lipase catalysis has potential for the preparation pase PS-D (Figure S3, Supporting Information).[16] Accord-
of intriguing sugar–β-amino acid conjugates (Scheme 2). ingly, lipase PS-D was selected for further studies in dry
Optimization was performed using model substrates, i.e., tAmOH.
rac-1a as an acyl donor and methyl α-d-galactopyranoside
Table 1. Lipase screening for the diastereoselective ring opening of
(2) as a nucleophile. As the solubility of unprotected methyl
rac-3,3-difluoro-4-phenylazetidin-2-one (rac-1a; 50 mm) with
glycopyranosides in the organic solvents that are typically
methyl α-d-galactopyranoside (2; 1 equiv.) in the presence of a
used for lipase catalysed acylation might be limited, the sol- lipase preparation (30 mgmL–1) and molecular sieves (3 Å;
50 mgmL–1) in dried solvents; reaction time 24 h at 47 °C.
ubility of 2 (50 mm) in 11 organic solvents was tested (Fig-
ure S2, Supporting Information), and three of them were
selected for our studies on different grounds. tert-Amyl
alcohol (tAmOH) as a water-soluble alcohol was chosen as
it dissolves 2 well, and it was also used as a solvent in our
previous lipase-catalysed acylations of 2–4.[8c] TBME was
included as it is an excellent solvent for enantioselective β-
lactam ring cleavages with lipases.[1,4,7] 2-MeTHF was cho-
sen as a more stable variant of THF. THF is a common
solvent used for the enzymatic reactions of carbo-
hydrates,[13] and 2-MeTHF has more recently been used as
an alternative reaction solvent.[14] TBME and 2-MeTHF
are both water insoluble, and although they dissolve 2
poorly, we hoped that the presence of other reagents and
products would affect the solubility equilibrium in a posi-
tive way.[15]
Entry Lipase
TBME
2-MeTHF
tAmOH
Conv. ee1a
[%][a] [%][b]
Conv. ee1a
[%][a] [%][b]
Conv. ee1a
[%][a] [%][b]
1
2
3
4
CAL-B 57/70 Ͼ99/Ͼ99 26/42 37/75
PS-D 53/60 98/Ͼ99
RM IM 33/40 50/70
TL IM 26/36 18/27
43/47 79/Ͼ99
50/53 93/Ͼ99 44/54 78/98
10/14 12/17
11/18 10/18
7/9
5/7
25/35 4[c]/8[c]
[a] Conversion as the disappearance of rac-1a against internal stan-
dard after 24 h/48 h. [b] ee1a from the 24 h/48 h sample. [c] Oppo-
site diastereoselectivity observed.
For the lipase-screening, the model substrates [i.e., rac-
1a (50 mm) and 2 (1 equiv.)] were dissolved as well as pos-
sible in the selected solvents, and then a catalyst [CAL-B,
Another aspect of the reaction optimization concerns the
lipase PS-D, Rhizomucor miehei lipase (Lipozyme RM IM), amounts of the sugar and the β-lactam as well as their ratio
or Thermomyces lanuginosus lipase (Lipozyme TL IM); in the reaction mixture. As the chiral HPLC/UV method
30 mgmL–1] and molecular sieves (3 Å; 50 mgmL–1) were did not allow the detection of sugars 2–4 or conjugate prod-
added. To enhance the solubility of both the substrates and ucts 5, an achiral RP-HPLC/UV-CAD method[8c] was de-
the products, the reactions were carried out at 47 °C. A chi- veloped to quantify the amount of unreacted β-lactam by
ral HPLC/UV method[4a,4b] was used to follow the dia- UV detection, and the amounts of the unreacted sugar and
stereoselective ring-opening of rac-1a with 2. However, conjugate products by CAD detection simultaneously from
products 5a and (S)-14a were invisible to the chiral HPLC/ a single sample. The chiral HPLC-UV method was used
UV detection. All of the lipases catalysed the ring-opening, only to determine the enantiomeric excess values ee1 of un-
and the reactivity was always highest in TBME (Table 1, reacted β-lactam 1. Increasing the amount of 2 (1, 2, and
entries 1–4). The Lipozyme RM IM and TL IM prepara- 5 equiv.) relative to rac-1a did not affect the formation of
tions were dropped from further studies as the conversions 5a (53% in every case after 24 h) in the presence of lipase
were discouraging, even after 48 h (Table 1, entries 3 and 4). PS-D (30 mgmL–1) and molecular sieves in dry tAmOH.
The ring-opening reactions proceeded in a highly diastereo- As expected, doubling the amount of rac-1a (100 mm) and
selective manner with both CAL-B and lipase PS-D, as keeping the amount of 2 unchanged (50 mm) allowed 75%
indicated by the high enantiomeric excess (ee1a) values of of the sugar to be converted into 5a in 24 h. We continued
remaining 1a at the given conversions in the 24 h samples with our optimization using a 1:1 mixture of the reactants
(Table 1, entries 1 and 2). According to Scheme 1, ring- as using the commercially available methyl α-d-glycopyr-
opening of rac-1a can occur by both of the competing pro- anosides to drive the progress of the reaction is economi-
cesses of alcoholysis by the sugar to form the sugar–β- cally more sensible than using in-house prepared β-lactams.
amino acid conjugate (i.e., 5a; Scheme 1, route A) and the
There was still an analytical problem. Sugar conjugate
hydrolysis side-reaction leading to the corresponding β- 5a (Figure 1, dotted line) appeared as a broad peak with an
amino acid [i.e., (S)-14a; Scheme 1, route B]. When rac-1a indefinite shape, which made accurate quantification im-
(50 mm) was subjected to the reaction conditions with lipase possible. We concluded that the free amino group of the
PS-D in the absence of the sugar, significant hydrolysis of product should be protected for analysis. Di-tert-butyl di-
Eur. J. Org. Chem. 2014, 6753–6760
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6755

R. Sundell, E. Siirola, L. T. Kanerva
FULL PAPER
carbonate (Boc2O) was selected as a protecting reagent as it
reacts with amines (but not with alcohols) directly without
catalysts or coupling reagents.[17] First, we tested whether
or not the N-Boc-protected β-lactam could be formed un-
der the ring-opening reaction conditions. Thus, rac-1a and
2 were allowed to react in the presence of lipase PS-D and
molecular sieves in dry tAmOH for 48 h, and then the reac-
tion mixture was split in two portions. The lipase PS-D was
removed from one of the portions, and then Boc2O was
added to both portions. Subsequent analysis revealed that
the unreacted starting material (i.e., 1a) was unchanged,
while the product (i.e., 5a) was derivatized with a Boc
group; the presence or absence of lipase PS-D did not affect
this derivatization. Moreover, and as expected, the RP-
HPLC/UV-CAD peak for the N-Boc-protected compound
(i.e., 8a) appeared sharp (Figure 2, solid line), unlike that
of free 5a (dotted line), and the retention times of the two
compounds were different. We also found that the presence
of Boc2O in the enzymatic reaction mixture had no effect
on the formation of 8a in situ via 5a. Thus, it is possible to
choose either to prepare 5a and derivatize a sample for
analysis, or to prepare N-protected derivative 8a directly.
Figure 2. Effect of lipase PS-D (5–50 mgmL–1) on the consumption
of rac-1a (50 mm; filled signs) and the formation of 8a (open signs)
in the reaction with 2 (1 equiv.) and Boc2O (1 equiv.) in the pres-
ence of molecular sieves (3 Å; 50 mgmL–1) in tAmOH.
As already mentioned, 35% of rac-1a was hydrolysed in the
absence of the sugar (and Boc2O). A lipase PS-D content
of 30 mgmL–1 was chosen to be optimal; as well as consid-
erations about the hydrolysis side-reaction, the shaking of
the reaction mixture was more effective when less solid ma-
terial was present.
Substrate Scope Studies towards the Synthesis of Sugar–β-
Amino Acid Conjugates
The optimized conditions were used in substrate-scope
studies in which the structures of both the β-lactams and
the methyl glycopyranosides were varied. Thus, the reac-
tions between rac-1a–1d (50 mm) and 2–4 (1 equiv.) were
carried out in the presence of lipase PS-D (30 mgmL–1),
Boc2O (1 equiv.), and molecular sieves in dry tAmOH
(Table 2). Each β-lactam except rac-1d (Table 2, entry 6)
was reactive under the reaction conditions, as indicated by
the disappearance of 1 after 24 h. The result with rac-1d
was expected, as unfluorinated or otherwise unactivated
Figure 1. RP-HPLC/UV-CAD chromatograms: in the presence of
Boc2O (solid line) and without Boc2O (dashed line).
Finally, the amount of lipase PS-D (5–50 mgmL–1) was compounds were also unreactive in previous ring-opening
optimized for the acylation of rac-1a (50 mm) with 2 reactions.[4a] However, the yields of products 8–10 revealed
(1 equiv.) in the presence of Boc2O (1 equiv.) and molecular that the product formation was clear only between rac-1a
sieves in dry tAmOH (Figure 2). The reactivity progress- and sugars 2 and 3 and between rac-1b and 2 (Table 2, en-
ively increased with increasing amounts of catalyst, as de- tries 1, 7 and 2). Experiments with the pure enantiomers
termined by both the disappearance of rac-1a (solid lines revealed that it was the (S)-enantiomer of rac-1b that re-
with filled signs) and the formation of 8a (dashed lines with acted with 2 (Table 2, entries 2–4). Secondly, the conversion
open signs) in the 24 h sample. The difference between the of the β-lactam only gave a reasonable correlation with for-
solid curve and the corresponding dotted curve now more mation of the product (i.e., conjugates 8–10) for the reac-
or less indicates the proportion of the hydrolysis side-reac- tions of rac-1a with 2 and 3 (Table 2, entries 1 and 7). In
tion of the β-lactam ring under each of the acylation condi- other cases, the yields of the conjugate products (i.e., 8–10)
tions. In the 24 h samples, the amount of hydrolysis was were very low, even though the β-lactam had been con-
almost undetectable with low amounts of catalyst (5 and sumed to some extent, but also the amounts of recovered
10 mgmL–1), but it slightly increased when the amount of starting sugars 3 and 4 were low (Table 2, entries 8–12).
enzyme was increased, and reached ca. 5% with 50 mgmL–1 These results could be due to the low solubilities of starting
of lipase PS-D. The fact that so little hydrolysis was seen in sugars 3 and 4 in these reaction mixtures. Thirdly, the con-
the formation of 8a indicated that 2 effectively suppressed version values for pyridine-substituted β-lactam rac-1c were
the potential of the residual water to act as a nucleophile. around 80% for its reactions with sugars 2–4, but no sugar-
6756
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 6753–6760

Lipase-Catalysed Acylation of Methyl α-d-Glycopyranosides
conjugate formation was detected (Table 2, entries 5, 9, and 6-O-acylated products (i.e., 8–10; Table 3). For some
12). The high conversions might be due to the greater ten- reason, the yields of the products were much lower than in
dency of 1c to undergo hydrolysis as a result of the strongly the small-scale reactions (Table 2), and the sugar–β-
electron-withdrawing pyridine ring attached to the β-lactam amino acid conjugates (i.e., 8a–10a and 8b) were isolated in
ring. In addition, interactions between the basic pyridine 3–20% yield.
nitrogen and other reagents and materials in the reaction
In addition to the unprotected methyl glycopyranosides,
mixture may reduce the effective β-lactam concentration methyl 3,4-O-isopropylidene-α-d-galactopyranoside (15), a
under the reaction conditions. The appearance of several more soluble sugar, was prepared from 2 by a known
small unidentified peaks on the RP-HPLC/CAD chromato- method.[18] In the previously reported lipase-catalysed acyl-
gram is consistent with this observation.
ation of carbohydrates with glyceric acid enantiomers,
changing the substrate from 2 to 15 resulted in a significant
increase in both the reactivity and the eventual product
yield.[8c] Here, the results were less impressive (Table 2, en-
try 13). β-Lactam rac-1a reacted rather sluggishly with 15
in the presence of lipase PS-D, and the ring-opening pro-
ceeded with poor diastereoselectivity (E Ͻ 10 as estimated
using the ee1 value after the given conversion).
Table 2. Reaction between 1a–1d (50 mm) and methyl α-d-glycopyr-
anosides 2–4 (1 equiv.), and between rac-1a and 3,4-O-isopropyl-
idene methyl α-d-galactopyranoside (15; 1 equiv.), in the presence
of lipase PS-D (30 mgmL–1), Boc2O (1 equiv.), and molecular
sieves (3 Å; 50 mgmL–1) in tAmOH; reaction time 24 h at 47 °C.
Product Diastereoisomerism
The ring-opening of a racemic β-lactam by a methyl
glycopyranoside may lead to a mixture of two dia-
stereomers, each of which is enantiomerically pure. As
excellent enantioselectivity (E Ͼ 200) had previously been
observed in the lipase-PS-D-catalysed ring-opening of rac-
1a with methanol in TBME,[4a] a highly diastereoselective
ring-opening leading to the formation of highly dia-
stereomerically enriched sugar conjugates was expected also
in this work. Unfortunately, chromatographic means failed
to reveal whether more than one diastereomer was present
for any of the isolated products (i.e., 8a–10a and 8b). The
ee1 value 75% at 44% conversion (giving E of the order of
100) for the unreacted (R)-1a indicates good diastereomeric
purity for galactopyranoside derivative 8a, while the ee1
value 30% at 30% conversion (giving E of the order of 10)
tends to propose low diastereomeric purity for glucoside
derivative 9a (Table 2, entries 1 and 7). However, the dia-
stereoselective formation of the glucoconjugate 9a might be
better than approximated on the basis of the E value due
to the low solubity of 3. To further investigate the diastereo-
isomeric distribution of the products, we used enantiomer-
Entry
Sugar Lactam Conversion
[%][a]
ee1
[%]
Yield [%][b]
2–4 8–10
1
2
3
4
5
6
7
8
2
2
2
2
2
2
3
3
3
4
4
4
rac-1a
rac-1b
(R)-1b
(S)-1b
rac-1c
rac-1d
rac-1a
rac-1b
rac-1c
rac-1a
rac-1b
rac-1c
44
46
13
75
69
99
99
16
–
30
33
20
15
16
6
49
55
100
42
46
22
–
58
39
83[d]
0
64
–
[e]
100
70
47
52
22
–
30
24[c]
Ͻ1[c]
26
[e]
9
83[d]
15
–
10
11
12
13
Ͻ1[c]
Ͻ1[c]
22
20
76[d]
31
19
–
[e]
15[f] rac-1a
13
96[c]
4[c]
[a] Disappearance of rac-1a–1d against an internal standard. [b]
[2–4]t/[2–4]0 and [8–10]t/[2–4]0. [c] Relative amount. [d] According
to achiral analysis only. [e] Not determined due to the large number
of peaks (see text). [f] Methyl 3,4-O-isopropylidene-α-d-galactopyr-
anoside (15) was used instead of the unprotected glycopyranoside.
Next, preparative-scale reactions (with a 10 mL reaction ically pure β-lactams (R)- and (S)-1b. When the reaction
volume) were carried out for the reactions of rac-1a with between each of the individual enantiomers (R)- and (S)-1b
2–4 and that of 1b with 2. The reactions between 1 (50 mm) (50 mm) and 2 (1 equiv.) was examined in the presence of
and 2–4 (1 equiv.) were carried out in the presence of Boc2O lipase PS-D (30 mgmL–1), Boc2O (1 equiv.), and molecular
(1 equiv.), lipase PS-D (30 mgmL–1), and molecular sieves sieves in tAmOH, (R)-1b failed to give any of the product
in tAmOH, with the aim of isolating and characterizing the (Table 2, entry 3) although some of the lactam was con-
Table 3. Preparative-scale synthesis of 8–10.[a]
Entry
Lactam
Sugar
ee1
[%]
1a–1b
8–10
8–10
8–10
Yield [%][b]
Yield [%][c]
Yield [%][b]
[α]2D5[d]
1
2
3
4
rac-1a
rac-1b
rac-1a
rac-1a
2
2
3
4
Ͼ99
33
26
–
25
23
18
20
18
10
3
+78.1
+82.2
+48.2
n.d.
62
37
47
13
n.d.
[a] 1 (50 mm), 2–4 (1 equiv.), Boc2O (1 equiv.), lipase PS-D (30 mgmL–1), molecular sieves (3 Å; 50 mgmL–1), tAmOH, 47 °C. [b] Isolated
yield. [c] Amount ([8–10]t/[2–4]0) in the reaction mixture. [d] (c = 1.0, CH3CN).
Eur. J. Org. Chem. 2014, 6753–6760
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6757

R. Sundell, E. Siirola, L. T. Kanerva
FULL PAPER
sumed (31% conversion after 48 h; Figure 3). When the drolysis by the residual water in the lipase preparation was
same experiment was conducted in the absence of the sugar, shown to be negligible in the presence of molecular sieves.
the conversion of (R)-1b still reached 29% after 48 h. Com- By using pure enantiomers as acyl donors, it was shown
parative experiments without lipase or sugar showed similar that the (S)-enantiomer of rac-1b reacted with 2, while (R)-
behaviour, with an 8% conversion, suggesting that non-en- 1b was unreactive. β-Lactam rac-1c, with an electron-with-
zymatic ring-opening of 1b was occurring under the reac- drawing and basic pyridine substitent on the lactam ring
tion conditions in addition to possible enzyme-catalysed was consumed to a great extent in the reaction (ca. 80%),
hydrolysis. The (S)-enantiomer, on the other hand, gave a although conjugate formation was negligible or nil with
58% conversion with respect to the lactam under identical sugars 2–4. Unfluorinated β-lactam rac-1d did not react at
reaction conditions, and diastereomerically and enantio- all. Gluco- and mannopyranosides (3 and 4, respectively),
merically pure product 8b was formed in 39% yield with equatorial OH groups at C-4, were shown to be worse
(Table 2, entry 4). It can be concluded that highly dia- nucleophiles than the corresponding galactopyranoside,
stereomerically pure product 8a was also obtained in the with the axial OH group at C-4, in the lipase-PS-D-cata-
ring-opening reaction of rac-1b and 2 catalysed by lipase lysed reaction. In addition to any structural effect, the low
PS-D. The difference in the product yields [22% with rac- solubility of 3 and 4 may restrict the formation of the de-
1b (50 mm) and 39% with (S)-1b (50 mm)] is consistent with sired sugar–β-amino acid conjugates, as enzymatic hydro-
the fact that only half of the racemic lactam was the reac- lysis of the β-lactams by residual water is favoured at low
tive enantiomer.
sugar concentrations.
Experimental Section
General Remarks: All reagents and materials were used as received
from commercial sources unless otherwise stated. Solvents were
dried with molecular sieves (3 Å) before use. Powdered zinc was
acid-washed before use.[19] Lipase preparations from Amano (Burk-
holderia cepacia lipase as Lipase PS-D and Lipase PS-C II)
and Novozym (Candida antarctica lipase B as Novozym 435,
Rhizomucor miehei lipase as Lipozyme RM IM, and Thermomyces
lanuginosus lipase as Lipozyme TL IM) were used. HPLC analyses
were performed with
detector) equipped with
a
HP 1090 HPLC/DAD (diode-array
Daicel CHIRACEL-OD-H
a
(4.6 mmϫ250 mmϫ5 μm) column or a Waters 2690 HPLC/DAD-
CAD (charged-aerosol detector) equipped with an Agilent Tech-
nologies ZORBAX Eclipse XDB-C8 (4.6 mmϫ150 mmϫ5 mm)
column. GC analyses were performed with a HP 6850 GC/FID
Figure 3. Progression curves for the lipase-PS-d-catalysed ring-
opening of (R)-, (S)-, and rac-1b in tAmOH; consumption of 1b
(solid signs) and formation of 8b (open signs). For reaction condi-
tions, see text.
equipped with
a Chrompack CP-Chiralsil-DEX CB (25 m
ϫ0.25 mmϫ0.25 μm) column. Analytical thin-layer chromatog-
raphy (TLC) was carried out on Merck Kieselgel 60F254 sheets, and
the spots were visualized with UV light (254 nm) and by treatment
with sulfuric acid (10% in ethanol) and heating. Chromatographic
separations were carried out by column chromatography on Kiesel-
Conclusions
1
gel 60 (0.063–0.200 μm). H, 13C, and 19F NMR spectra were re-
corded with a Bruker Avance 500 spectrometer at 298 K in CDCl3,
[D6]DMSO, or [D4]methanol, and tetramethylsilane or 2,2,2-tri-
fluoroethanol was used as an internal standard. Mass spectra were
recorded with a Bruker Daltonics micrOTOF-Q (ESI-TOF) instru-
ment in positive mode. Specific rotations were measured with a
Perkin–Elmer 341 polarimeter using the sodium D line, and values
are presented as degcm–2 g–1. Melting points were measured with a
Gallenkamp device.
The regioselective formation of sugar–β-amino acid con-
jugates by lipase-PS-D-catalysed diastereoselective ring-
opening of three racemic 3,3-difluoro-4-arylazetidin-2-ones
(rac-1a–1c) and a non-fluorine-containing analogue (rac-
1d) with methyl α-d-glycopyranosides (2–4) in the presence
of Boc2O and molecular sieves in tAmOH was examined.
An achiral RP-HPLC/UV-CAD method was developed to
quantify the amounts of the products formed, and a chiral
HPLC/UV method was used to obtain the ee values of the
unreacted β-lactams.
Small-Scale Enzymatic Experiments
Reactions between Methyl α-D-Glycopyranosides 2–4 and rac-1a–1d:
In a typical small-scale reaction, a methyl α-d-glycopyranoside [ga-
lacto (2), gluco (3), or manno (4); 9.7 mg], a β-lactam (rac-1a–1d;
1–2 equiv.), Boc2O (1 equiv.), a lipase preparation (30 mgmL–1),
molecular sieves (3 Å; 50 mgmL–1), and a solvent (1 mL) were
shaken (170 rpm) at +47 °C. Samples (50 μL) were taken at inter-
vals, filtered, diluted with n-hexane (150 μL) or acetonitrile
(100 μL), and analysed by the developed HPLC methods (see Sup-
We have shown that the formation of sugar–β-
amino acid conjugates by regio- and stereoselective lipase
catalysis is possible. Conjugates 8a–10a and 8b were formed
by the diastereoselective ring-opening of phenyl- (rac-1a)
and thiophen-3-yl- (rac-1b) substituted 3,3-difluoro-4-aryl-
azetidin-2-ones by sugars 2–4, but the yield of 10a (isolated
yield 3%) was very low. The susceptibility of rac-1a to hy- porting Information).
6758
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 6753–6760

Lipase-Catalysed Acylation of Methyl α-d-Glycopyranosides
Preparative-Scale Reactions
Boc2O (109 mg, 1 equiv., 0.5 mmol), Lipase PS-D (30 mgmL–1;
300 mg), molecular sieves (3 Å; 50 mgmL–1; 500 mg), and tert-amyl
alcohol (tAmOH; 10 mL) were shaken (170 rpm) at 47 °C for 48 h.
Then the reaction mixture was filtered, and the solvents were evap-
orated. Column chromatography (ethyl acetate/petroleum ether,
1:9, and ethyl acetate/ethanol, 95:5) gave (R)-1a (34 mg, 0.2 mmol,
Methyl
aminopropanoyl]-α-
6-O-[(S)-2,2-Difluoro-3-phenyl-3-(tert-butoxycarbonyl)-
-galactopyranoside (8a): Methyl α-d-galacto-
D
pyranoside (2; 97 mg, 0.5 mmol), rac-1a (91 mg, 1 equiv.,
0.5 mmol), Boc2O (109 mg, 1 equiv., 0.5 mmol), Lipase PS-D
(30 mgmL–1; 300 mg), molecular sieves (3 Å; 50 mgmL–1; 500 mg),
and tert-amyl alcohol (tAmOH; 10 mL) were shaken (170 rpm) at
47 °C for 24 h. Then the reaction mixture was filtered, and the sol-
vents were evaporated. Column chromatography (ethyl acetate/eth-
anol, 95:5) gave (R)-1a as a mixture with Boc2O (101 mg, Ͼ99%
ee) and 8a (47 mg, 0.1 mmol, 20%). Data for 8a: [α]2D5 = +78.1 (c
= 1.0, CH3CN). 1H NMR (500 MHz, 298 K, [D6]DMSO): δ = 1.38
(s, 9 H, 3 CH3), 3.29 (s, 3 H, OCH3), 3.54 (ddd, J = 3.3, J = 5.6,
J = 9.5 Hz, 1 H, 3-H), 3.60 (ddd, J = 3.6, J = 6.6, J = 10.1 Hz, 1
H, 2-H), 3.70 (t, J = 3.4 Hz, 1 H, 4-H), 3.82 (dd, J = 2.8, J =
8.4 Hz, 1 H, 5-H), 4.22 (dd, J = 3.0, J = 11.2 Hz, 1 H, 6a-H), 4.33
(dd, J = 8.7, J = 11.1 Hz, 1 H, 6b-H), 4.61 (d, J = 3.6 Hz, 1 H, 1-
H), 4.63 (d, J = 6.6 Hz, 1 H, 2-OH), 4.69 (d, J = 5.7 Hz, 1 H, 3-
OH), 4.74 (d, J = 4.4 Hz, 1 H, 4-OH), 5.32 (ddd, J = 11.3, J =
18.2, J = 21.6 Hz, 1 H, CHPh), 7.37 (m, 3 H, arom.), 7.48 (m, 2
H, arom.), 8.22 (d, J = 10.2 Hz, 1 H, -NH) ppm. 13C NMR
(126 MHz, 298 K, [D6]DMSO): δ = 27.87 (3 CH3), 54.26 (OCH3),
56.05 (CHPh), 66.80 (C-6), 67.81 (C-5), 67.92 (C-2), 68.87 (C-3),
68.94 (C-4), 78.93 [C(CH3)3], 99.93 (C-1), 128.25 (arom.), 128.52
(arom.), 128.63 (arom.), 133.30 (CF2), 147.53 (arom.), 154.92 [C(O)
OtBu], 162.35 (C=O) ppm. 19F NMR (471 MHz, 298 K, [D6]-
DMSO): δ = –113.0 (dd, J = 11.2, J = 250.2 Hz), –116.2 (dd, J =
17.9, J = 250.7 Hz) ppm. HRMS: calcd. for C21H29F2NO9Na+ [M
+ Na]+ 500.17026; found 500.17064.
37%, 26% ee) and 9a (34 mg, 51 μmol, 10%). Data for 9a: [α]2D5
=
+48.4 (c = 1.0, CH3CN). 1H NMR (500 MHz, 298 K, [D6]DMSO):
δ = 1.38 (s, 9 H, 3 CH3), 3.07 (ddd, J = 6.3, J = 9.2, J = 15.4 Hz,
1 H, 4-H), 3.22 (ddd, J = 3.7, J = 6.5, J = 10.0 Hz, 1 H, 2-H), 3.28
(s, 3 H, OCH3), 3.40 (ddd, J = 4.9, J = 9.2, J = 14.1 Hz, 1 H, 3-
H), 3.60 (ddd, J = 1.5, J = 7.4, J = 9.3 Hz, 1 H, 5-H), 4.23 (dd, J
= 7.5, J = 11.7 Hz, 1 H, 6a-H), 4.44 (dd, J = 1.4, J = 11.4 Hz, 1
H, 6b-H), 4.56 (d, J = 3.4 Hz, 1 H, 1-H), 4.83 (d, J = 6.5 Hz, 1 H,
2-OH), 4.95 (d, J = 4.9 Hz, 1 H, 3-OH), 5.24 (J = 5.7 Hz, 1 H, 4-
OH), 5.31 (ddd, 1 H, J = 10.2 Hz, J = 17.1 Hz, J = 23.2 Hz,
CHPh), 7.37 (m, 3 H, arom.), 7.47 (m, 2 H, arom.), 8.20 (d, 1 H,
J = 10.2 Hz, NH) ppm. 13C NMR (126 MHz, 298 K, [D6]DMSO):
δ = 27.95 (3 CH3), 54.25 (OCH3), 56.06 (CHPh), 66.34 (C-6), 69.05
(C-5), 70.12 (C-4), 71.58 (C-2), 72.99 (C-3), 78.96 [C(CH3)3], 99.60
(C-1), 114.12 (CF2), 128.30 (arom.), 128.55 (arom.), 128.64 (arom.),
133.41 (arom.), 154.93 [C(O)OtBu], 162.35 (C=O) ppm. HRMS:
calcd. for C21H29F2NO9Na+ [M
500.17040.
+
Na]+ 500.17026; found
Methyl
aminopropanoyl]-α-
6-O-[(S)-2,2-Difluoro-3-phenyl-3-(tert-butoxycarbonyl)-
-mannopyranoside (10a): Methyl α-d-manno-
D
pyranoside (4; 97 mg, 0.5 mmol), rac-1a (91 mg, 1 equiv.,
0.5 mmol), Boc2O (109 mg, 1 equiv., 0.5 mmol), Lipase PS-D
(30 mgmL–1; 300 mg), molecular sieves (3 Å; 50 mgmL–1; 500 mg),
and tert-amyl alcohol (tAmOH, 10 mL) were shaken (170 rpm) at
47 °C for 48 h. Then the reaction mixture was filtered, and the sol-
vents were evaporated. Column chromatography (ethyl acetate/pe-
troleum ether, 1:9, and ethyl acetate/ethanol, 95:5) gave (R)-1a
(43 mg, 0.2 mmol, 47%, 13% ee) and 10a (8 mg, 17 μmol, 3%).
Methyl 6-O-[(S)-2,2-Difluoro-3-(thiophen-3-yl)-3-(tert-butoxycarb-
onyl)aminopropanoyl]-α-D-galactopyranoside (8b): Methyl α-d-ga-
lactopyranoside (2; 97 mg, 0.5 mmol), rac-1b (95 mg, 1 equiv.,
0.5 mmol), Boc2O (109 mg, 1 equiv., 0.5 mmol), Lipase PS-D
(30 mgmL–1; 300 mg), molecular sieves (3 Å; 50 mgmL–1; 500 mg),
and tert-amyl alcohol (tAmOH; 10 mL) were shaken (170 rpm) at
47 °C for 24 h. Then the reaction mixture was filtered, and the sol-
vents were evaporated. Column chromatography (ethyl acetate/pe-
troleum ether, 1:9, and ethyl acetate/ethanol, 95:5) gave (R)-1b
(57 mg, 0.3 mmol, 62%, 33% ee) as a colourless semisolid, and 8b
(43 mg, 91 μmol, 18%) as a colourless semisolid. Data for 8b:
[α]2D5 = +82.2 (c = 1.0, CH3CN). 1H NMR (500 MHz, 298 K, [D6]-
DMSO): δ = 1.39 (s, 9 H, 3 CH3), 3.28 (s, 3 H, OCH3), 3.54 (ddd,
J = 3.2, J = 5.6, J = 9.5 Hz, 1 H, 3-H), 3.59 (ddd, J = 3.6, J = 6.5,
J = 10.1 Hz, 1 H, 2-H), 3.70 (dd, J = 3.4, J = 3.6 Hz, 1 H, 4-H),
3.84 (dd, J = 2.8, J = 8.5 Hz, 1 H, 5-H), 4.21 (dd, J = 3.1, J =
11.2 Hz, 1 H, 6a-H), 4.33 (dd, J = 8.8, J = 11.0 Hz, 1 H, 6b-H),
4.60 (d, J = 3.5 Hz, 1 H, 1-H), 4.62 (d, J = 6.6 Hz, 1 H, 2-OH),
4.68 (d, J = 5.7 Hz, 1 H, 3-OH), 4.74 (d, J = 4.4 Hz, 1 H, 4-OH),
5.44 [ddd, J = 12.3, J = 15.8, J = 23.5 Hz, 1 H, CH(thiophenyl)],
7.25 (d, J = 4.9 Hz, 1 H, arom.), 7.55 (dd, J = 3.0, J = 4.9 Hz, 1
H, arom.), 7.65 (d, J = 2.2 Hz, 1 H, arom.), 8.09 (J = 10.1 Hz, 1
H, NH) ppm. 13C NMR (126 MHz, 298 K, [D6]DMSO): δ = 27.93
(3 CH3), 54.32 (OCH3), 55.72 [CH(thiophenyl)], 66.80 (C-6), 67.84
(C-5), 67.94 (C-2), 68.89 (C-3), 68.97 (C-4), 78.97 [C(CH3)3], 99.97
(C-1), 125.35 (arom.), 126.48 (arom.), 127.65 (arom.), 133.58
(arom.), 154.89 [C(O)OtBu], 162.32 (C=O) ppm. 19F NMR
(471 MHz, 298 K, [D6]DMSO): δ = –114.3 (dd, J = 11.6, J =
248.6 Hz), –116.6 (dd, J = 16.3, J = 248.6 Hz) ppm. HRMS: calcd.
for C19H27F2NO9SNa+ [M + Na]+ 506.12668; found 506.12123.
1
Data for 10a: H NMR (500 MHz, 298 K, [D6]DMSO): δ = 1.38
(s, 9 H, 3 CH3), 3.26 (s, 3 H, OCH3), 3.41 (m, 1 H, 4-H), 3.46 (m,
1 H, 3-H), 3.54 (m, 1 H, 5-H), 3.62 (ddd, J = 1.7, J = 3.2, J =
4.7 Hz, 1 H, 2-H), 4.24 (dd, J = 7.6, J = 11.6 Hz, 1 H, 6a-H), 4.46
(ddd, J = 1.7, J = 11.4, J = 13.1 Hz, 1 H, 6b-H), 4.53 (d, J =
1.3 Hz, 1 H, 1-H), 4.76 (d, J = 5.9 Hz, 1 H, 3-OH), 4.87 (d, J =
4.5 Hz, 1 H, 2-OH), 5.09 (J = 5.5 Hz, 1 H, 4-OH), 5.32 (ddd, J =
11.8, J = 17.5, J = 22.6 Hz, 1 H, CHPh), 7.37 (m, 3 H, arom.),
7.47 (m, 2 H, arom.), 8.22 (d, J = 10.2 Hz, 1 H, NH) ppm. 13C
NMR (126 MHz, 298 K, [D6]DMSO): δ = 27.95 (3 CH3), 53.92
(OCH3), 55.72 (CHPh), 66.60 (C-4, C-6), 69.96 (C-2), 70.30 (C-5),
70.59 (C-3), 78.97 [C(CH3)3], 100.97 (C-1), 128.30 (arom.), 128.55
(arom.), 128.65 (arom.), 133.43 (arom.), 154.94 [COC(CH3)3],
162.14 (C=O) ppm. 19F NMR (471 MHz, 298 K, [D6]DMSO): δ =
–113.3 (dd, J = 11.6, J = 250.7 Hz), –116.2 (dd, J = 17.3, J =
250.5 Hz) ppm. HRMS: calcd. for C21H29F2NO9Na+ [M + Na]+
500.17026; found 500.17040.
Supporting Information (see footnote on the first page of this arti-
cle): Synthesis of rac-1a–1d; retention times for compounds during
HPLC analysis; solubility estimation for 2; estimation of enzymatic
hydrolysis by residual water; chemoenzymatic preparation of both
enantiomers of 1b; 1H and 13C NMR spectra for products and
intermediates.
Acknowledgments
Methyl
aminopropanoyl]-α-
6-O-[(S)-2,2-Difluoro-3-phenyl-3-(tert-butoxycarbonyl)-
-glucopyranoside (9a): Methyl α-d-glucopyr-
D
The Instrumentarium Science Foundation, Finland is generously
anoside (3; 97 mg, 0.5 mmol), rac-1a (91 mg, 1 equiv., 0.5 mmol), acknowledged by R. S. for financial support.
Eur. J. Org. Chem. 2014, 6753–6760
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6759

R. Sundell, E. Siirola, L. T. Kanerva
FULL PAPER
1212; c) X.-G. Li, L. T. Kanerva, Adv. Synth. Catal. 2006, 348,
197–205.
[8] a) A. Hietanen, F. S. Ekholm, R. Leino, L. T. Kanerva, Eur. J.
Org. Chem. 2010, 6974–6980; b) A. Hietanen, L. T. Kanerva,
Eur. J. Org. Chem. 2012, 2729–2737; c) R. Sundell, L. T.
Kanerva, Eur. J. Org. Chem. 2013, 4971–4978.
[1] a) A. Liljeblad, L. T. Kanerva, Tetrahedron 2006, 62, 5831–
5854; b) E. Busto, V. Gotor-Fenández, V. Gotor, Chem. Soc.
Rev. 2010, 39, 4504–4523; c) E. Busto, V. Gotor-Fernández, V.
Gotor, Chem. Rev. 2011, 111, 3998–4035; d) E. Forro, F. Fülop,
Curr. Med. Chem. 2012, 19, 6178–6187.
[9] W. K. Hagmann, J. Med. Chem. 2008, 51, 4359–4369.
[10] a) P. Csomós, L. T. Kanerva, G. Bernáth, F. Fülöp, Tetrahe-
dron: Asymmetry 1996, 7, 1789–1796; b) E. Forró, F. Fülöp,
Tetrahedron: Asymmetry 2001, 12, 2351–2358; c) X.-G. Li, L. T.
Kanerva, Tetrahedron: Asymmetry 2007, 18, 2468–2472.
[2] a) C. Palomo, J. M. Aizpurua, I. Ganboa, M. Oiarbide, Curr.
Med. Chem. 2004, 11, 1837–1872; b) B. Alcaide, P. Almendros,
C. Aragoncillo, Chem. Rev. 2007, 107, 4437–4492; c) T. Heck,
H.-P. E. Kohler, M. Limbach, O. Flogel, D. Seebach, B.
Geueke, Chem. Biodiversity 2007, 4, 2016–2030.
[3] See, for instance: L. T. Kanerva, A. Liljeblad, Transesterifica-
tion – Biological, in: Encyclopedia of Catalysis, Wiley Online
Library, 2010, DOI: 10.1002/0471227617.eoc197.
[4] a) X.-G. Li, M. Lähitie, M. Päiviö, L. T. Kanerva, Tetrahedron:
Asymmetry 2007, 18, 1567–1573; b) X.-G. Li, M. Lähitie, L. T.
Kanerva, Tetrahedron: Asymmetry 2008, 19, 1857–1861; c) P.
D’Arrigo, L. T. Kanerva, X.-G. Li, C. Saraceno, S. Servi, D.
Tessaro, Tetrahedron: Asymmetry 2009, 20, 1641–1645.
[5] a) H. Smidt, A. Fischer, P. Fischer, R. D. Schmid, Biotechnol.
Techniques. 1996, 10, 335–338; b) T. Wagegg, M. M. Enzel-
berger, U. T. Bornscheuer, R. D. Schimd, J. Biotechnol. 1998,
61, 75–78; c) D. R. Duarte, E. L. Castillo, E. Bárzana, A.
López-Munguía, Biotechnol. Lett. 2000, 22, 1811–1814; d)
M. V. Sergeeva, V. V. Mozhaev, J. O. Rich, Y. L. Khmelnitsky,
Biotechnol. Lett. 2000, 22, 1419–1422; e) A. Torres-Gavilán, J.
Escalante, I. Regla, A. López-Munguía, E. Castillo, Tetrahe-
dron: Asymmetry 2007, 18, 2621–2624; f) H. Ismail, R. M. Lau,
F. van Rantwijk, R. A. Sheldon, Adv. Synth. Catal. 2008, 350,
1511–1516; g) A. Liljeblad, P. Kallio, M. Vainio, J. Niemi, L. T.
Kanerva, Org. Biomol. Chem. 2010, 8, 886–895; h) J. Brem, L.-
C. Nencze, A. Liljeblad, M. C. Turcu, C. Paizs, F.-D. Irimie,
L. T. Kanerva, Eur. J. Org. Chem. 2012, 3288–3294.
[11]
a) S. Marcotte, X. Pannecoucke, C. Feasson, J.-C. Quirion, J.
Org. Chem. 1999, 64, 8461–8464; b) N. Boyer, P. Gloanec, G.
De Nanteuil, P. Jubault, J.-C. Quirion, Tetrahedron 2007, 63,
12352–12366; c) N. Boyer, P. Gloanec, G. De Nanteuil, P. Ju-
bault, J.-C. Quirion, Eur. J. Org. Chem. 2008, 4277–4295.
C. Palomo, M. Diarbide, S. Bindi, J. Org. Chem. 1998, 63,
2469–2474.
[12]
[13]
B. Danieli, M. Luisetti, G. Sampognaro, G. Carrea, S. Riva, J.
Mol. Catal. B 1997, 3, 193–201.
[14] a) Y. Simeó, J. V. Sinisterra, A. R. Alcántar, Green Chem. 2009,
11, 855–862; b) Z.-G. Chen, D.-N. Zhang, L. Cao, Y.-B. Han,
Bioresource Technol. 2013, 133, 82–86.
[15] a) P. Tsavas, S. Polydorou, I. Faflia, E. C. Voutsas, D. Tassios,
M. V. Flores, K. Naraghi, P. J. Halling, F. Chamouleau, M.
Ghoul, J.-M. Engasser, M. Ferrer, F. J. Plou, J. Chem. Eng.
Data 2002, 47, 807–810; b) E. C. Voutsas, P. Tsavas, K. Mag-
oulas, D. Tassios, M. Ferrer, F. J. Plou, A. Ballesteros, J. Chem.
Eng. Data 2002, 47, 1517–1520.
[16] L. Veum, L. T. Kanerva, P. J. Halling, T. Maschmeyer, U.
Hanefeld, Adv. Synth. Catal. 2005, 347, 1015–1021.
[17] Y. Basel, A. Hassner, J. Org. Chem. 2000, 65, 6368–6380.
[18] A. Rashid, W. Mackie, I. J. Colquhoun, D. Lamba, Can. J.
Chem. 1990, 68, 1122–1127.
[19] A. I. Vogel, B. S. Furniss, A. J. Hannaford, P. W. G. Smith,
A. R. Tatchell, Vogel’s Textbook of Practical Organic Chemis-
try, 5th ed., Pearson Education Ltd., Harlow, UK, 1989, p. 467.
Received: June 23, 2014
[6] a) P.-O. Syrén, P. Hendil-Forssell, L. Aumailley, W. Besen-
matter, F. Gounine, A. Svendsen, M. Martinelle, K. Hult,
ChemCatChem 2012, 13, 645–648; b) P.-O. Syrén, FEBS J.
2013, 280, 3069–3083.
[7] a) R. Brieva, J. Z. Crich, C. J. Sih, J. Org. Chem. 1993, 58,
1068–1075; b) E. Forró, F. Fülöp, Org. Lett. 2003, 5, 1209–
Published Online: September 5, 2014
6760
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 6753–6760