One-pot oxidation-hydrocyanation sequence coupled to lipase-catalyzed diastereoresolution in the chemoenzymatic synthesis of sugar cyanohydrin esters

Article abstract of DOI:10.1002/ejoc.201200086

A three-step, one-pot synthesis and diastereoresolution sequence is described in anhydrous toluene starting from methyl α-D-2,3,4-tri-O- acetylgalacto- (1a), -manno- (1b) and -glucopyranosides (1c). The reaction sequence, including consecutive transformations through the aldehyde [PhI(OAc)₂, 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)] and cyanohydrin [basic resin or (R)-oxynitrilase] into the (6R)-cyanohydrin ester (lipase) is shown to proceed in a one-pot cascade, except that the basic resin (when used) should be removed before the addition of the enzymatic acylation reagents. We have shown that the effective transformation of 1a (75 % reaction yield) through labile intermediates gives the stable (6R)-cyanohydrin butanoate (85 % de). Further diastereomeric purification by chromatography is possible, although the product is already of high diastereopurity. (6R)-Cyanohydrin esters are obtained through acylation with Burkholderia cepacia lipase. The (6S)-ester (de 99 %) is produced by Candida rugosa lipase when the sequence is started from 1c whereas the other sugar derivatives are less suited to the reaction with lipase. Copyright

Full text of DOI:10.1002/ejoc.201200086

FULL PAPER
DOI: 10.1002/ejoc.201200086
One-Pot Oxidation–Hydrocyanation Sequence Coupled to Lipase-Catalyzed
Diastereoresolution in the Chemoenzymatic Synthesis of Sugar Cyanohydrin
Esters
Ari Hietanen[a] and Liisa T. Kanerva*[a]
Keywords: Carbohydrates / Enzyme catalysis / Heterogeneous catalysis / Hydrolases / Chiral resolution / Oxidation
A three-step, one-pot synthesis and diastereoresolution se-
quence is described in anhydrous toluene starting from
methyl α-D-2,3,4-tri-O-acetylgalacto- (1a), -manno- (1b)
and -glucopyranosides (1c). The reaction sequence, includ-
ing consecutive transformations through the aldehyde
[PhI(OAc)2, 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)]
and cyanohydrin [basic resin or (R)-oxynitrilase] into the
(6R)-cyanohydrin ester (lipase) is shown to proceed in a one-
pot cascade, except that the basic resin (when used) should
be removed before the addition of the enzymatic acylation
reagents. We have shown that the effective transformation of
1a (75% reaction yield) through labile intermediates gives
the stable (6R)-cyanohydrin butanoate (85% de). Further
diastereomeric purification by chromatography is possible,
although the product is already of high diastereopurity. (6R)-
Cyanohydrin esters are obtained through acylation with Bur-
kholderia cepacia lipase. The (6S)-ester (de 99%) is pro-
duced by Candida rugosa lipase when the sequence is
started from 1c whereas the other sugar derivatives are less
suited to the reaction with lipase.
also been reported.[7] Selectivity in these reactions is caused
by internal asymmetric induction and leads to dia-
stereomerically enriched products.
Introduction
Modification of natural carbohydrates is a powerful ap-
proach both for synthetic objectives and to the elucidation
of carbohydrate function in chemical biology studies.[1]
When such modifications are required to generate new
asymmetric centers, high stereochemical control, in ad-
dition to the control caused by the stereostructure of the
carbohydrate itself, is necessary. The nonreducing end of a
monosaccharide is an important site for synthetic modifica-
tions, for instance, for biological recognition processes. On
the other hand, the C-6 position of aldohexoses is attractive
for structural modifications that allow specific interactions
with external substrates. For instance, cyclodextrins modi-
fied with cyanohydrin moieties at C-6 can function as ef-
ficient hydrolase-like catalysts.[2]
A strategy often employed in C-6 modifications of sugar
derivatives consists of initial oxidation of the primary
alcohol functionality to the corresponding aldehyde fol-
lowed by the desired addition reaction. Examples of such
reactions include a Staudinger reaction initiated process,[3]
reductive amination,[4] Diels–Alder reaction,[5] and Grig-
nard reaction.[6] The addition of formaldehyde dialkylhy-
drazones to sugar aldehydes followed by oxidative cleavage
to yield free or benzyl-protected sugar cyanohydrins has
Cyanohydrins (α-hydroxynitriles) form a versatile class of
compounds for synthetic chemistry as they allow easy ac-
cess to important compounds, such as α-hydroxy acids and
amides, α-functionalized nitriles, 1,2-amino alcohols, and
N-heterocycles.[8] In our previous work, a cyanohydrin moi-
ety was introduced into the aglycon part of peracetylated
α-d- and α-l-mannosylacetaldehydes to afford novel
mannosylglycosides.[9] Stereoselectivity was then achieved
through lipase catalyzed acylation after a nonselective hy-
drocyanation step, affording the diastereomerically enriched
epimers of the sugar cyanohydrin and cyanohydrin acetate.
We now report the results of our studies on the introduction
of a cyanohydrin moiety at the C-6 position of monosac-
charides. The chemoenzymatic synthesis includes the depro-
tection of peracetylated methyl α-d-glycosides (formation of
compound 1, Step a), followed by a one-pot reaction se-
quence with oxidation (aldehyde 2 formation, Step b), hy-
drocyanation (C–C bond formation, Step c or cЈ), and acyl-
ation (Step d) at the C-6 position of a sugar pyranoside 3
under mild reaction conditions (Scheme 1). The synthesis
was optimized for methyl α-d-galactopyranoside, and the
method was then applied to the corresponding α-d-man-
nose- and α-d-glucose-based glycosides with 1H NMR
spectroscopic analysis of the reaction.
[a] Institute of Biomedicine/Department of Pharmacology, Drug
Development and Therapeutics/Laboratory of Synthetic Drug
Chemistry, University of Turku,
The absolute configurations of the cyanohydrin products
in Scheme 1 are depicted as was expected on the basis of
the models for (R)-oxynitrilase [(R)-HNL] and many lipase
enzymes. Accordingly, the (6S)-cyanohydrin was expected
20014 Turku, Finland
E-mail: liisa.kanerva@utu.fi
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201200086.
Eur. J. Org. Chem. 2012, 2729–2737
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2729

A. Hietanen, L. T. Kanerva
FULL PAPER
After peracetylation by using literature methods,[11] a set of
readily available lipase preparations was screened to assess
their ability to deprotect position C-6 by regioselective
methanolysis in diisopropyl ether (DIPE), affording com-
pounds 1a–c (Table 1). Lipase PS-D and PS-C II prepara-
tions (Burkholderia cepacia lipase immobilized on diato-
maceous earth and ceramic, respectively) were effective with
the galactoside substrate but displayed very low or no ac-
tivity toward the other sugar acetates (Table 1, entries 1 and
2). Lipases A (CAL-A adsorbed on Celite) and B (CAL-
B as a commercial Novozym 435 catalyst) from Candida
antarctica and Candida rugosa lipase (CRL, adsorbed on
Celite) gave generally low product yields for 1a–c (Table 1,
entries 3–5). Thermomyces lanuginosus lipase (Lipozyme TL
IM) allowed highly efficient deprotection irrespective of
which sugar was used (Table 1, entry 6), although the reac-
tion was not perfectly selective for C-6, particularly with
substrate 1c. Nevertheless, Lipozyme TL IM was used for
the preparative syntheses of 1a–c in good isolated yields
(64–84%).
Scheme 1. Synthetic strategy from carbohydrates: (a) regioselective
deprotection, (b) oxidation, (c) enzymatic hydrocyanation, (cЈ)
chemical hydrocyanation, (d) enzymatic acylation, and (e) enzy-
matic deacylation.
Table 1. Formation of 1a–c by lipase-catalyzed methanolysis of per-
acetylated methyl α-d-glycopyranosides.[a]
to be the major diastereomer obtained from the hydrocyan-
ation reaction catalyzed by (R)-oxynitrilase (Scheme 2, a),
whereas the formation of the (6R)-cyanohydrin ester was
suggested to be the major diastereomer with lipase cata-
lyzed O-acylation (Scheme 2, b).[10] NMR analysis was used
to examine the presumed absolute configurations and it was
found that these did not quite follow the expectations with
(R)-oxynitrilases.
Entry
Lipase
1a[%]
1b [%]
1c [%]
1
2
3
4
5
6
Lipase PS-D
Lipase PS-C II
Novozym 435
80
72
14
0
0
6
6
0
17
36
0
CAL-A on Celite 10
CRL on Celite 15
28[b]
64
Results and Discussion
Lipozyme TL IM 95 (81[c]) 92 (84[c])
69 (64[c])
[a] [Substrate]
= 0.05 m, [MeOH] = 0.25 m, lipase content
Chemoenzymatic Preparation of 1a–c (Step a)
50 mg mL–1 in DIPE. [b] Reaction time 72 h. [c] Isolated yields.
Our approach to the target cyanohydrins and cyano-
hydrin esters required the secondary alcohol functionalities
in the sugar moieties to be protected to block their reacti-
vity later in the lipase catalyzed acylation and to promote
solubility in organic solvents, whereas the primary alcohol
Oxidation of 1a (Step b)
Both biocatalytic and chemical oxidation are potential
group at C-6 was to remain unprotected so that the desired methods for the preparation of aldehydes 2a–c from the
cyanohydrin functionality could be introduced (Scheme 1). corresponding alcohols 1a–c (Step b of Scheme 1). The bio-
Scheme 2. Suggested configurations at C-6 based on (a) the benzaldehyde-mandelonitrile model for (R)-HNL and (b) the secondary
alcohol model for acylation by lipase PS catalysis.
2730
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 2729–2737

Chemoenzymatic Synthesis of Sugar Cyanohydrin Esters
catalytic oxidation of the primary alcohol group of hexose mation of 7a (Table 2, entry 2). Compared to soluble
sugars at C-6 reported in the literature largely rely on the TEMPO, the use of TEMPO immobilized on polystyrene
use of galactose oxidase in aqueous solutions, restricting gave low product yields (15%); thus, soluble TEMPO was
the studies to galactosyl moieties[12] and predisposing the used throughout the work. In addition to purification by
product to side reactions.[13] Methods that rely on the oxi- column chromatography, aqueous work-up was also delete-
dation of a primary alcohol to the aldehyde stage by chloro- rious for the produced aldehyde, causing product loss to the
peroxidase from Caldariomyces fumago in organic solvents aqueous phase and the formation of impurities. Therefore,
have been extensively studied,[14] but our efforts to use this subsequent steps from 1a in the reaction cascade were per-
approach failed with both unprotected methyl α-d-galacto- formed without isolating the aldehyde.
pyranoside and 1a as substrates.
Literature reports concerning chemical oxidation can be
be classified into three broad categories. Dimethyl sulfox-
Hydrocyanation of 2a–c (Steps c and cЈ)
ide-based oxidation of acetyl-protected sugars gave α,β-un-
A set of (R)-oxynitrilases were first studied for the trans-
saturated aldehyde 7a as the main product.[15] Catalytic
hydrocyanation between 2a and acetone cyanohydrin in
2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPO) coupled with
buffered microaqueous toluene containing all the reaction
a stoichiometric oxidant[16] or with a O2/laccase system[17]
components from the previous oxidation Step b (Step c in
for unprotected sugars carries the possibility of overoxid-
Scheme 1). The reaction was performed (1) in a one-com-
ation, affording the carboxylic acid rather than the alde-
partment reactor (all reagents and catalysts present in situ),
hyde, particularly when water or a base is present. On the
and (2) in a two-compartment reactor (acetone cyanohydrin
other hand, Dess–Martin oxidation is an effective method
decomposes to HCN in the presence of Amberlite IRA-
900 HO– ion exchange resin in one of the compartments,
thereafter diffusing freely into the enzymatic reaction mix-
ture in the second compartment).[21] In Method 1, (R)-oxy-
nitrilase in toluene containing tartrate buffer (4 vol.-%,
0.10 m, pH 5.4) was responsible for the decomposition of
acetone cyanohydrin and for the formation of the sugar cy-
anohydrin 3a. Low diastereomeric excess (27% de) and
product yield (17%) for (6R)-3a obtained by almond meal
catalysis (a rich source of (R)-oxynitrilase) were found to
be drawbacks of the reaction (Table 3, entry 1). Similar re-
sults were obtained with commercial (R)-oxynitrilase cata-
lysts from Prunus amygdalis (PaHNL) and Arabidopsis thal-
iana (AtHNL). Evidently, the rapid rate of HCN generation
(and resulting in high HCN concentration) in Method 1
caused enzyme inhibition and resulted in low yield in the
one-compartment reactor. In Method 2, the reaction yield
of (6R)-3a increased considerably with a moderate improve-
ment in the diastereomeric excess (Table 3, entry 2). Labile
3a was isolated in 35% yield (37% de) by column
chromatography, the rest of the product decomposing to
for sugar oxidation, but the method is typically used in sol-
vents such as dimethyl sulfoxide (DMSO), MeCN, or
CH2Cl2 that are less suitable for subsequent enzymatic
steps.[5,18] The method based on iodine(III) reagents with
catalytic TEMPO, also related to the Dess–Martin oxi-
dation, is particularly attractive because the isomerization
of chirally labile aldehydes and overoxidation to carboxylic
acids should be completely avoided.[18,19] For example, a di-
acetoxyiodobenzene/TEMPO oxidation of γ,δ-unsaturated
primary alcohols coupled to enzymatic hydrocyanation is
reported to provide advantages of better atom economy and
a more facile work-up compared to the use of Dess–Martin
periodinane alone.[20] Accordingly, the oxidation of 1a with
diacetoxyiodobenzene and catalytic TEMPO in dichloro-
methane resulted in 62% reaction yield for the desired alde-
hyde 2a with one major impurity among other unidentified
minor impurities (Table 2, entry 1). Whereas 2a was un-
stable upon chromatographic purification, the main impu-
rity was separated and shown to be the unsaturated alde-
hyde 7a. Toluene and DIPE are solvents commonly used
with oxynitrilases and lipases.[21] Oxidation in DIPE was
not achieved, evidently due to the low solubility of the hy-
pervalent iodine oxidant (Table 2, entry 3), whereas a yield
of 75% for 2a in toluene was accomplished without the for-
Table 3. Effect of catalyst and cyanide source on the (trans)hydro-
cyanation of 2a.
Table 2. Solvent effects on the oxidation of 1a (0.050 m) with
PhI(OAc)2 (0.055 m)/TEMPO (0.05 m).
Entry
Catalyst
(buffer vol.-%)
CN–
source
Yield
[%]
de
[%]
1
2
3
4
5
Almond meal[a] (4)
Almond meal[a] (4)
– (4)
ACH[c]
HCN[d]
ACH[c]
ACH[c]
ACH[c]
17[d]
74[d]
0
27
37
–
–
52
Almond meal[a] (0)
0[d]
85
Entry Solvent
1a [%]
2a [%]
7a [%]
IRA-900 OH–[b] (0)
1
2
3
CH2Cl2
toluene
DIPE
0
0
81
62
75
4
7
0
0
[a] 50 mgmL–1. [b] 10 mgmL–1. [c] 2 equiv. of acetone cyanohydrin
(one-compartment reactor). [d] HCN formed from acetone cyano-
hydrin (5 equiv., two-compartment reactor).
Eur. J. Org. Chem. 2012, 2729–2737
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2731

A. Hietanen, L. T. Kanerva
FULL PAPER
aldehyde 7a and other side products. Surprisingly, the pre-
vailing diastereomer had the 6R rather than the originally
expected 6S configuration (Scheme 2, a). There was no re-
action in the absence of the enzyme under otherwise iden-
tical conditions, indicating low enzymatic diastereoselectiv-
ity (Table 3, entry 3). The stereostructure of the galactose
ring evidently steers the stereochemical outcome of 3a. The
presence of some water is necessary for (R)-oxynitrilase ca-
talysis (Table 3, entry 4) and, accordingly, the use of almond
meal is preferred over other (R)-oxynitrilase catalysts be-
cause the water is mainly impregnated in the meal and thus
easy to remove for the next step in the reaction cascade.[9]
In addition to the decomposition of acetone cyanohydrin
(in the two-compartment reactor case above), the formation
of cyanohydrins was also catalyzed by bases (Scheme 1,
Figure 1. Yields of 2a (᭿,ᮀ) and 3a (᭹,᭺) with time scale; oxi-
dation/hydrocyanation (20 mgmL–1 Amberlite IRA-900 HO–) in
one-pot by add-on (filled symbols) and at once (open symbols)
Step cЈ). Thus, Amberlite IRA-900 in the HO– form was reagent additions.
used to transform 2a in the reaction mixture into dia-
stereomerically enriched (6R)-3a (52% de) in high yield in
24 hours using acetone cyanohydrin as the cyanide source
(Table 3, entry 5). Thus, asymmetric induction from the gal-
actopyranose ring gave a diastereomeric mixture containing
76 mol-% (6R)-3a and 24 mol-% (6S)-3a.
To conclude, chemical hydrocyanation under anhydrous
conditions by the add-on principle is most suitable for the
production of (6R)-3a (52% de in favor of the 6R-dia-
stereomer), whereas the almond meal method gives the
product with somewhat lower reaction yield (37% de). Pyr-
anoside derivatives 1b and 1c were also subjected to the
chemical oxidation/hydrocyanation conditions, but the reac-
tions were prone to generate side reactions, producing the
cyanohydrins 3b and 3c in 77 and 73% reaction yields,
respectively, with negligible diastereoselectivity (19 and 5%
de, respectively). These results indicate the importance of
the axial substituent at the C-4 position of the carbohydrate
scaffold as a stereocontrol element and the need to protect
2 against acetic acid elimination (formation of 7).
Increasing the resin content from 5 to 20 mg mL–1 in-
creased the yield of 3a from 43 to 92% in 24 hours without
affecting the diastereomeric excess (Table 4). On the other
hand, the formation of 7a (3% in 3 h and 20% in 24 h) was
also clear in the absence of acetone cyanohydrin. Accord-
ingly, the base catalyzes hydrocyanation considerably faster
than elimination. When acetone cyanohydrin and the basic
resin were both introduced after the oxidation step (add-on
synthesis), hydrocyanation proceeded rapidly to 83% yield
(2–3 h), and thereafter slowly increased to 92% (Figure 1,
filled symbols). To explain the slight increase in yield at
this point, we suggest that some condensation side-products
might decompose back to the aldehyde when the aldehyde
content in the reaction mixture is lowered due to hydrocy-
anation. However, when the oxidation and hydrocyanation
Lipase-Catalyzed Acylation of 3a–c (Step d)
The production of cyanohydrin esters under basic condi-
reagents were introduced simultaneously into the toluene tions from aldehydes in the presence of acetone cyano-
solution of 1a, only a trace amount of aldehyde 2a was hydrin, a lipase, and an acyl donor is among the oldest dy-
detected and the reaction quickly halted at approximately namic kinetic resolution methods.[22,23a–23c] However, when
20% yield for 3a (Figure 1, open circles). The reagents used acetone cyanohydrin, the basic resin, lipase PS-C II, and
in Steps b and cЈ are clearly not fully compatible. Appar- vinyl butanoate were introduced simultaneously into the re-
ently, the base consumes active iodine oxidant and induces action mixture obtained through Step b (Scheme 1), the for-
condensation reactions.
mation of 4a did not exceed 16%, with the main product
being 7a. The base-catalyzed formation of 7a means that
the dynamic equilibrium between 2a and 3a had turned
towards the aldehyde. As a possible explanation, the so-
called residual water introduced with the seemingly dry lip-
ase favors hydrolysis of vinyl butanoate, disfavoring acyl-
ation,[23d] and reduces the stability of 3a.
Table 4. Effect of Amberlite IRA-900 HO– content on the trans-
hydrocyanation of 2a.
The one-pot reaction from 1a–c to diastereomerically en-
riched 3a–c through Steps aǞbǞcЈ in anhydrous toluene
was repeated according to the above principles at 23 °C to
carry out the next lipase-catalyzed Step d (and e) of the
reaction cascade (Scheme 1). On the basis of the dia-
stereomeric excess values, diastereomerically enriched 3a–c
contained 76, 60 and 52 mol-% of galactosyl-, mannosyl-
and glucosylpyranoside derivatives as the major dia-
Entry Resin [mgmL–1] 2a [%]
3a [%]
de of 3a [%]
1
2
3
4
0
5
10
20
75
22
0
0
–
43
85
92
52
52
52
0
2732
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 2729–2737

Chemoenzymatic Synthesis of Sugar Cyanohydrin Esters
Table 5. Lipase screening for the acylation of 3a–c with vinyl butanoate in the reaction sequence starting from 1a–c.[a]
Entry Lipase
4a
Yield [%]
4b
Yield [%]
4c
de [%]
de [%]
Yield [%]
de [%]
1
2
3
4
5
6
Lipase PS-D
Lipase PS-C II
Novozym 435
CAL-A on Celite
CRL on Celite
Lipozyme TL IM
42
75
Ͻ1
10
0
87
85
–
83
–
65
65
76
69
77
Ͻ1
30
33
12
25
16
–
69
56
71
5
27
0
19
42
15
–
99[b]
–
4
–
[a] Reagents and conditions: Step b: 1a–c (0.05 m), PhI(OAc)2 (1.1 equiv.), TEMPO (0.1 equiv.), toluene, 23 °C, 12 h; Step cЈ: acetone
cyanohydrin (2 equiv.), Amberlite IRA-900 HO– (20 mgmL–1), one-pot reactor, 23 °C, 6 h, followed by decantation; Step d: vinyl butano-
ate (3 equiv.), lipase (50 mgmL–1), 23 °C, 72 h, absolute configurations at C-6 of 4b and 4c unconfirmed. [b] Opposite diastereomeric
preference to other lipases.
stereomers (evidently 6R also with 3b and 3c), respectively.
A set of commonly employed microbial lipases and vinyl
butanoate was studied next for the acylation of cyano-
hydrins 3a–c in toluene in the one-pot reaction mixtures
obtained after first removing the basic resin by filtration
(Table 5, entries 1–6). Lipases PS-D and PS-C II accepted
all three cyanohydrin pyranosides as substrates, although
the diastereoselectivity was good only for 4a (85–87% de,
Table 5, entries 1 and 2). Novozym 435 did not accept ga-
lactoside 3a, and the enzyme displayed very low diastereo-
selectivities with 3b and 3c (Table 5, entry 3). CAL-A
(Table 5, entry 4) and CRL (Table 5, entry 5) were markedly
active only with mannoside 3b. The diastereoselectivity of
CRL to produce 4c was excellent (99% de) although oppo-
site [evidently producing (S)-4c] to that of the other lipases
studied. In acylation, the stereoselectivity of CRL is known
to depend strongly on the structure of the racemic alcohol,
with reversed enantioselectivity to common lipases (like li-
pase PS) being observed, for instance, with cyanohydrin
substrates.[24] Lipozyme TL IM was completely inactive for
the acylation of cyanohydrins 3a–c (Table 5, entry 6). At
this point, investigations with 3b and 3c were discontinued
because of the disapointing results.
Next, optimization for the preparation of 4a was contin-
ued by studying the effects of lipase PS-C II and vinyl but-
anoate concentration on the acylation of cyanohydrin 3a.
Acylation was faster when 100 mgmL–1 of enzyme was used
Figure 2. (a) Reaction yields of 4a vs. time, and (b) diastereomeric
than with 50 mgmL–1 with a set amount of the acyl donor,
excess vs. reaction yields for the lipase PS-C II-catalyzed acylation
of 3a (92%, 52% de) in toluene at 23 °C. (᭿) lipase (50 mgmL–1)
and vinyl butanoate (3 equiv.); (ᮀ) lipase (50 mgmL–1) and vinyl
butanoate (6 equiv.); (᭹) lipase (100 mgmL–1) and vinyl butanoate
(3 equiv.); (᭺) lipase (100 mgmL–1) and vinyl butanoate (6 equiv.).
however, 80–90% yields were reached irrespective of the
amount of acyl donor (Figure 2, a). At the same time, dia-
stereomeric excess values depended only on conversion, al-
lowing the formation of (6R)-4a at 89–92% diastereomeric
excess up to a reaction yield of ca. 60% (Figure 2, b); at
this point a sharp drop in the diastereomeric excess was
The length of the acyl part of the acyl donor clearly af-
observed as the less reactive (6S)-3a diastereomer became fected reactivity, whereas the diastereomeric excess of 4a–
more active in product formation.
5a mainly depended on conversion (Table 6). Thus, al-
Eur. J. Org. Chem. 2012, 2729–2737
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2733

A. Hietanen, L. T. Kanerva
FULL PAPER
though the reaction was initially most effective with vinyl of the major diastereomer are clearly in an anti orientation
acetate (reaction yield 45% in 24 h, Table 6, entry 2), the whereas there is more gauche character in the minor dia-
reaction yield of 75% [approximately the theoretical yield stereomer. When the torsion angle energies were analyzed
for (6R)-4a] was reached only with vinyl butanoate in 72 h with MM2 molecular mechanics, the gauche conformations
(Table 6, entry 1). Acylation with vinyl laurate (to give 6a) along the C-5–C-6 bond were found to be somewhat lower
was somewhat more selective but impractically slow for ef- in energy with (6S)-3a (ca. 2 kcalmol–1 higher than anti
ficient application (Table 6, entry 3).
conformation) than with (6R)-3a (ca. 5 kcalmol–1 higher
than anti conformation), suggesting that (6S)-3a is the
minor diastereomer.
Table 6. Effect of an acyl donor (3 equiv.) on the acylation of 3a
with lipase PS-C II (50 mgmL–1).
Secondly, the diastereomeric mixture of 3a (37% de,
Table 3, entry 2) was dissolved with (R)- and (S)-mandelic
acid, both in the presence of 4-(dimethylamino)pyridine
1
(DMAP), and the H NMR spectra were measured (Fig-
ure 3).[25] The downfield shift of H-6 of the minor dia-
stereomer of 3a in the complex with (R)-mandelate-
DMAPH+ compared to the shift with the (S)-mandelate-
DMAPH+ (ΔδRS = +0.032 ppm) implies homochirality
with (R)-mandelonitrile and, therefore, the assignment (6S)-
3a; the major diastereomer of 3a had nearly equal shift with
both (R)- and (S)-mandelate-DMAPH+. However, this was
also observed with diastereomerically pure 3a and full veri-
fication of the absolute configuration by this method could
not be obtained.
Entry Vinyl ester
Product 24 h
Yield
72 h
Yield
[%]
de
[%]
de
[%]
[%]
1
2
3
butanoate
acetate
laurate
4a
5a
6a
34
45
13
90
91
96
75
66
34
85
81
96
Unlike free cyanohydrin 3a, products 4a and 5a allowed
the diastereomeric excess to be enhanced upon column
chromatography. However, the recoverable yield of the pure
diastereomer (deϾ 99%) remained low due to considerable
tailing in the separation, even when highly diastereo-
merically enriched esters (de of the order of 85%) were puri-
fied. Thus, in the preparative synthesis, the fractions of both
(6R)-4a and (6R)-5a were enriched to diastereomeric ex-
cesses of more than 99%, with isolated yields of 39–41%
(reaction yield in the order of 75%) as calculated over the
three steps from 1a. The activity of lipase PS-C II was then
used in the reverse direction by subjecting cyanohydrin
acetate (6R)-5a to methanolysis (under the conditions de-
tailed in Table 1, reaction time 24 h). Free cyanohydrin
(6R)-3a was obtained by simple filtration of the enzyme
without chromatography and epimerization in 73% yield.
Figure 3. 1H NMR signals of the minor (downfield) and major
(upfield) diastereomer of 3a (37% de) in complex with (R)-mand-
elic acid–DMAP (top) and (S)-mandelic acid–DMAP (bottom).
The models for the respective complexes of (6S)-3a are shown on
left.
Absolute Configuration
As described above, the (R)-oxynitrilase and lipase mod-
els predict different stereoisomeric preferences for the prod-
ucts 3 and 4 (Scheme 2). However, the same diastereomer
preferentially forms in the base- (Step cЈ) as well as in the
(R)-oxynitrilase-catalyzed (Step c) hydrocyanation, and this
is also the preferentially reacting diastereomer in lipase PS-
C II catalyzed acylation (Step d). Accordingly, NMR analy-
sis of the diastereomeric mixture of 3a could be used to
confirm the absolute configuration of the obtained cyano-
hydrin esters.
Overall, the NMR experiments support, although do not
unambiguously confirm, the assignment of (6R)-3a (con-
trary to the prediction of Scheme 2, a) and (6R)-4a configu-
rations to be preferentially formed in the hydrocyanation
and lipase PS-C II catalyzed acylation steps.
Conclusions
First, both stereoisomers (6R)- and (6S)-4a have a mini-
mum energy conformation in which H-5 and H-6 are in
A straightforward, multistep synthesis and kinetic dia-
an anti-relationship along the C-5–C-6 bond. The coupling stereoresolution sequence has been developed based on bio-
3
constants J5,6 measured were 5.4 Hz for the minor and and chemo-catalytic reaction steps, starting from peracetyl-
8.7 Hz for the major diastereomer of 3a. Thus, H-5 and H-6 ated methyl α-d-glycosides. Essentially, three main reaction
2734
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 2729–2737

Chemoenzymatic Synthesis of Sugar Cyanohydrin Esters
Anomeric signals were assigned based on 13C and HSQC spectra
and 1-CH3O signals correlated to those based on peak integrals.
For galactoside derivatives (Xa) the signals were confirmed by full
assignment of the prepared compounds, however, the assignments
are tentative for mannosyl (Xb) and glucosyl (Xc) compounds. Col-
umn chromatography was performed on dried silica gel (60 Å, 230–
400 mesh).
steps, oxidation [PhI(OAc)2, TEMPO], hydrocyanation (ba-
sic Amberlite IRA-900 ion exchange resin and acetone cya-
nohydrin), and acylation (lipase PS-C II and a vinyl ester)
taking place through a cascade sequence in anhydrous tolu-
ene turned out to be most applicable. It is essential to note
that the sugar aldehydes and cyanohydrins are unstable dur-
ing chromatographic separation, emphasizing the impor-
tance of proceeding without purifying the intermediate
products and obtaining stable cyanohydrin esters. The but-
anoate ester of galactosyl cyanohydrin was prepared with
high diastereoselectivity (85% de) and reaction yield (75%).
Further diastereomeric purification of the ester product by
column chromatography afforded fractions of cyanohydrin
acetate 5a and butanoate 4a as (6R)-diastereomers with dia-
stereomeric excesses of more than 99%. However, the yield
remained low (31–41%) due to imperfect diastereomeric
separation. When coupled to enzymatic deacylation of
(6R)-4a with methanol, the corresponding relatively labile
free (6R)-cyanohydrin (de Ͼ 99%) was obtained through
simple filtration of the lipase. It has been shown that the
selectivity of the method depends on the sugar configura-
tion because mannose- and glucose-derived substrates gave
inferior results under the same conditions. In these cases,
the method was hampered partly due to side reactions trig-
gered by acetate elimination at the C-4 position of the acet-
ylated sugar aldehyde. Changes in protective group strategy
might enable improvements in synthesis yield and selectiv-
ity.
Table 7. Partial assignment and chemical shifts of the 1H NMR
signals used in the analysis of galactosides (Xa).
H-1 [ppm]
H-6 [ppm]
OCH3 [ppm]
1a
2a
5.00
9.54
3.68, 3.52
5.16
3.42
3.46
3a (R/S)
4a (R/S)
5a (R/S)
6a (R/S)
7a
5.06/5.11
5.08/5.12
5.08/5.12
5.08/5.12
9.25
4.40/4.68
5.48/n.d.
5.46/n.d.
5.47/n.d.
5.15
3.48/3.45
3.49/3.43
3.49/3.43
3.49/3.43
3.52
1
Table 8. Tentative assignment and chemical shifts of the H NMR
signals used in the analysis of mannosides (Xb) and glucosides (Xc).
Mannosides (Xb)
H-1 [ppm] OCH3 [ppm]
Glucosides (Xc)
H-1 [ppm]
OCH3 [ppm]
1
2
4.73
4.81
3.41
3.45
3.50/3.45
3.49/3.44
4.97
5.07
5.09/5.03
5.09/5.02
3.41
3.45
3.50/3.45
3.49/3.44
3 (R/S) 4.83/4.79
4 (R/S) 4.82/4.78
Syntheses of Alcohols 1a–c. Typical Procedure: Methyl α-d-galacto-
pyranoside (1.01 g, 5.20 mmol) was stirred with I2 (50 mg) in Ac2O
(5 mL) for 10 min at 23 °C. The reaction was diluted with EtOAc
(10 mL) and washed with 10% aq. Na2CO3 (2ϫ10 mL). The aque-
ous phases were combined and extracted with EtOAc (2ϫ10 mL).
The organic phases were combined, dried with Na2SO4, concen-
trated, and filtered through a silica pad (EtOAc/hexane, 4:1; Rf
= 0.85) to collect the tetraacetylated product (1.79 g, 4.94 mmol,
95%).
Experimental Section
Materials and Methods: Chemical reagents were purchased from
commercial sources and used as received unless otherwise men-
tioned. Methyl-α-d-glycosides, PhI(OAc)2, TEMPO (soluble and
on polystyrene, 1.0 mmolg–1), and acetone cyanohydrin were pur-
chased from Aldrich; vinyl acetate, butanoate, and laurate were
purchased from Fluka. Almond meal (β-glucosidase from al-
monds) was acquired from Sigma, immobilized PaHNL from Chi-
ralVision, aq. AtHNL from Evocatal, lipase PS-D and PS-C II
(Burkholderia cepacia lipase) from Amano Europe, and Novozym
435 (Candida antarctica lipase B) and Lipozyme TL IM (Thermo-
myces lanuginosus lipase) from Novozymes. CAL-A (Candida ant-
arctica lipase A, from Novozymes) and CRL (Candida rugosa li-
pase, from Sigma) powders were immobilized on Celite in the pres-
ence of sucrose.[26] Chloroperoxidase from Caldariomyces fumago
(lyophilized and immobilized preparations) were kindly obtained
from Bio-Research Products, Inc. (Iowa, USA). Amberlite IRA-
900 ion exchange resin (Aldrich) was conditioned to the HO– form
before use.[23] NMR spectra were measured with a Bruker Avance
500 MHz spectrometer, and high-resolution mass spectra were re-
corded with a Bruker micro-TOF-Q quadrupole-TOF spectrometer
operating in the ESI+ mode. Optical rotations were measured with
a Perkin–Elmer 241 polarimeter, and [α]2D5 values are given in units
of 10–1 degcm2 g–1. Analytical scale reactions were performed on
1–5 mL scale at room temp. (ca. 23 °C) or at 48 °C. Samples (100–
200 μL) were analyzed by NMR (samples concentrated and redis-
solved in CDCl3 with TMS as a calibrant). Whenever sampling
resulted in more than ca. 10% loss in reaction volume, parallel
reactions were run and sampled once each. Reactions were ana-
One of the peracetylated sugars (450 mg, 1.24 mmol) and Lipo-
zyme TL IM (620 mg) were shaken in diisopropyl ether (12.15 mL)
and MeOH (251 μL, 6.20 mmol) at 48 °C. After 24 h, the enzyme
was filtered off and the product was purified by column chromatog-
raphy (EtOAc/hexane, 4:1; Rf = 0.55] to yield, for instance, 1a
(323 mg, 1.01 mmol, 81%).
The NMR spectra of 1a–c are in accordance with those reported
in ref.[27] and HRMS correspond to the sodium adducts of the mo-
lecular ion.
Almond Meal Catalyzed Synthesis of Cyanohydrin 3a: In a dried
glass vessel with two compartments connected through a class tube
between the headspace of the compartments, 1a (141 mg,
0.440 mmol) was dissolved in anhydrous toluene (8.8 mL) in one
of the compartments (compartment 1). PhI(OAc)2 (156 mg,
0.484 mmol) and TEMPO (6.9 mg, 0.044 mmol) were added, and
the reaction mixture was stirred at 23 °C. After 12 h, tartrate buffer
(0.10 m, pH 5.4, 350 μL) and almond meal (435 mg) were added.
For the production of HCN, in compartment 2, hexane (8.8 mL),
Amberlite IRA-900 HO– (88 mg) and acetone cyanohydrin
(200 μL, 2.19 mmol) were introduced. The reaction mixtures in
both compartments were stirred at 23 °C for 24 h. From the hydro-
cyanation mixture (compartment 1) the almond meal was filtered
and the crude product (74%, 37% de by NMR analysis) was puri-
fied by column chromatography (EtOAc/hexane, 6:4; Rf = 0.50) to
1
lyzed using assigned H NMR signals according to Tables 7 and 8.
Eur. J. Org. Chem. 2012, 2729–2737
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2735

A. Hietanen, L. T. Kanerva
collect 3a (54 mg, 0.156 mmol, 35%, 37% de). (6R)-3a: H NMR (OCOCH3), 169.8 (OCOCH3), 168.2 (OCOCH3), 115.3 (CN), 97.5
FULL PAPER
1
(500.13 MHz, CDCl3, 25 °C): δ = 5.51 (dd, J4,3 = 3.4 Hz, J4,5
=
(1-C), 67.7 (2-C), 67.4 (5-C), 67.0 (3-C), 66.1 (4-C), 58.3 (6-C), 56.0
(OCH3), 21.0 (OCOCH3), 20.6 (OCOCH3), 20.5 (OCOCH3), 20.0
(OCOCH3) ppm (contains traces of 7a). HRMS: calcd. for
C16H21NO10Na+ [M + Na]+ 410.1058; found 410.1074.
1.0 Hz, 1 H, 4-H), 5.35 (dd, J3,2 = 10.9 Hz, 1 H, 3-H), 5.18 (dd,
J2,1 = 3.7 Hz, 1 H, 2-H), 5.06 (d, 1 H, 1-H), 4.40 (d, J6,5 = 8.7 Hz,
1 H, 6-H), 4.17 (dd, 1 H, 5-H), 3.7 (br., 1 H, OH), 3.48 (s, 3 H,
OCH3), 2.20 (s, 3 H, COCH3), 2.10 (s, 3 H, COCH3), 2.01 (s, 3 H,
COCH3) ppm. 13C NMR (500.13 MHz, CDCl3, 25 °C): δ = 170.5
(OCOCH3), 170.3 (OCOCH3), 170.1 (OCOCH3), 118.2 (CN), 97.3
(1-C), 68.9 (5-C), 67.9 (2-C), 67.5 (4-C), 67.4 (3-C), 60.1 (6-C), 55.8
(OCH3), 20.8 (OCOCH3), 20.7 (OCOCH3), 20.8 (OCOCH3) ppm.
Enzymatic Methanolysis of (6R)-5a to give (6R)-3a: Cyanohydrin
acetate (R)-5a (106 mg, 0.274 mmol, 99% de) and lipase PS-C II
(273 mg) were shaken in diisopropyl ether (5.42 mL) at 48 °C, and
MeOH (55 μL, 1.36 mmol) was added. After 24 h the lipase was
filtered to yield (R)-3a (69 mg, 0.200 mmol, 73%, de Ͼ 99%).
[α]2D5 = +117 (c = 1.0, CHCl3). NMR spectra and HRMS corre-
spond to those given for (6R)-3a.
1
(6S)-3a: H NMR (500.13 MHz, CDCl3, 25 °C): δ = 5.57 (dd, J4,3
= 3.5 Hz, J4,5 = 1.1 Hz, 1 H, 4-H), 5.37 (dd, J3,2 = 10.9 Hz, 1 H,
3-H), 5.23 (dd, J2,1 = 3.6 Hz, 1 H, 2-H), 5.11 (d, 1 H, 1-H), 4.68
(d, J6,5 = 5.4 Hz, 1 H, 6-H), 4.17 (dd, 1 H, 5-H), 3.7 (br., 1 H,
OH), 3.45 (s, 3 H, OCH3), 2.23 (s, 3 H, COCH3), 2.10 (s, 3 H,
COCH3), 2.03 (s, 3 H, COCH3) ppm. 13C NMR (125.77 MHz,
CDCl3, 25 °C): δ = 117.4 (CN), 97.5 (1-C), 67.8 (5-C), 67.8 (2-C),
67.8 (4-C), 67.0 (3-C), 60.4 (6-C), 56.1 (OCH3), 21.1 (OCOCH3),
20.9 (OCOCH3) ppm (C=O signals too weak for detection). (6R/
6S)-3a: HRMS: calcd. for C14H19NO9Na+ [M + Na]+ 368.0952;
found 368.0919.
Supporting Information (see footnote on the first page of this arti-
1
cle): H and 13C NMR spectra of the prepared compounds.
Acknowledgments
The authors thank the Academy of Finland for financial support
(research grant number 121983). This work was sponsored by the
European Union (EU) in the frame of the COST Action CM0701.
One-Pot Synthesis of Cyanohydrin Butanoate (6R)-4a: Alcohol 1a
(438 mg, 1.37 mmol) was dissolved in anhydrous toluene (27.4 mL),
PhI(OAc)2 (485 mg, 1.51 mmol) and TEMPO (21.4 mg,
0.137 mmol) were added and 2a was formed after 12 h as described
above. Amberlite IRA-900 HO– resin (540 mg) and acetone cya-
nohydrin (248 μL, 2.72 mmol) were added into the reaction mix-
ture. After 6 h the solution contained (6R)-3a (86%, 55% de). The
basic resin was removed by decanting the solution part into a sec-
ond vessel loaded with lipase PS-C II (1.35 g) followed by the ad-
dition of vinyl butanoate (495 μL, 3.90 mmol). The mixture was
shaken at 23 °C. After 72 h, the lipase was filtered off from the
mixture containing (6R)-4a (75%, 85% de), and the filtrate was
washed with EtOAc (3ϫ30 mL). The combined filtrates were con-
centrated and purified by column chromatography (EtOAc/hexane,
6:4; Rf = 0.62) to collect (6R)-4a (233 mg, 0.56 mmol, 41%, de Ͼ
99%). [α]2D5 = +95 (c = 1.0, CHCl3). 1H NMR (500.13 MHz,
CDCl3, 25 °C): δ = 5.48 (d, J6,5 = 9.5 Hz, 1 H, 6-H), 5.42 (dd, J4,5
= 1.0, J4,3 = 3.5 Hz, 1 H, 4-H), 5.35 (dd, J3,2 = 11.0 Hz, 1 H, 3-
H), 5.14 (dd, J2,1 = 3.5 Hz, 1 H, 2-H), 5.08 (d, 1 H, 1-H), 4.32 (dd,
1 H, 5-H), 3.49 (s, 3 H, OCH3), 2.29 (m, 2 H, CH2CH2CH3), 2.11
(s, 3 H, COCH3), 2.10 (s, 3 H, COCH3), 1.98 (s, 3 H, COCH3),
1.65 (m, 2 H, CH2CH2H3), 0.95 (t, JCH2–CH3 = 7.5 Hz, 3 H,
CH2CH2CH3) ppm. 13C NMR (125.77 MHz, CDCl3, 25 °C): δ =
170.9 (OCOPr), 170.3 (OCOCH3), 170.0 (OCOCH3), 169.8 (OC-
OCH3), 115.4 (CN), 97.5 (1-C), 67.7 (2-C), 67.4 (5-C), 67.0 (3-C),
66.1 (4-C), 58.1 (6-C), 56.0 (OCH3), 35.1 (CH2CH2CH3), 20.8 (OC-
OCH3), 20.6 (OCOCH3), 20.5 (OCOCH3), 17.9 (CH2CH2CH3),
13.4 (CH2CH2CH3) ppm. HRMS: calcd. for C18H25NO10Na+ [M
+ Na]+ 438.1371; found 438.1402.
[1] For example, see: Carbohydrates in Chemistry and Biology,
1st ed., vol. 1–4 (Eds.: B. Ernst, G. W. Hart, P. Sinaÿ), Wiley-
VCH, Weinheim, Germany, 2000.
[2] F. Ortega-Caballero, C. Rousseau, B. Christensen, T. E. Pe-
tersen, M. Bols, J. Am. Chem. Soc. 2005, 127, 3238–3239.
[3] X. Li, Q. Yin, L. Jiao, Z. Qin, J. Feng, H. Chen, J. Zhang, M.
Meng, Carbohydr. Res. 2011, 346, 401–409.
[4] a) J. Neumann, J. Thiem, Eur. J. Org. Chem. 2010, 900–908; b)
F. Peri, J. Jiménez-Barbero, V. García-Aparicio, I. Tvarosˇka, F.
Nicotra, Chem. Eur. J. 2004, 10, 1433–1444.
[5] P. A. Burland, D. Coisson, H. M. I. Osborn, J. Org. Chem.
2010, 75, 7210–7218.
[6] H. Stepowska, A. Zamojski, Tetrahedron 1999, 55, 5519–5538.
˛
[7] a) R. Fernández, E. Martín-Zamora, C. Pareja, J. M. Lassa-
letta, J. Org. Chem. 2001, 66, 5201–5207; b) J. M. Lassaletta,
R. Fernández, E. Martín-Zamora, C. Pareja, Tetrahedron Lett.
1996, 37, 5787–5790.
[8] J. Holt, U. Hanefeld, Curr. Org. Synth. 2009, 6, 15–37.
[9] A. Hietanen, F. S. Ekholm, R. Leino, L. T. Kanerva, Eur. J.
Org. Chem. 2010, 6974–6980.
[10] R. J. Kazlauskas, A. N. E. Weissfloch, A. T. Rappaport, L. A.
Cuccia, J. Org. Chem. 1991, 56, 2656–2665.
[11] K. P. R. Kartha, R. A. Field, Tetrahedron 1997, 53, 11753–
11766.
[12] a) R. Schoevaart, T. Kieboom, Top. Catal. 2004, 27, 3–9; b) V.
Bonnet, R. Duval, C. Rabiller, J. Mol. Catal. B 2003, 24–25,
9–16.
[13] a) K. Parikka, M. Tenkanen, Carbohydr. Res. 2009, 344, 14–
20; b) A. Maradufu, A. S. Perlin, Carbohydr. Res. 1974, 32,
127–136.
[14] See, for example: a) A. Zaks, D. R. Dodds, J. Am. Chem. Soc.
1995, 117, 10419–10424; b) E. Kiljunen, L. T. Kanerva, Tetra-
hedron: Asymmetry 1999, 10, 3529–3535; c) E. Kiljunen, L. T.
Kanerva, J. Mol. Catal. B 2000, 9, 163–17; d) F. van de Velde,
F. van Rantwijk, R. A. Sheldon, Trends Biotechnol. 2001, 19,
73–80.
[15] D. M. Mackie, A. S. Perlin, Carbohydr. Res. 1972, 24, 67–85.
[16] a) M. Angelin, M. Hermansson, H. Dong, O. Ramström, Eur.
J. Org. Chem. 2006, 4323–4326; b) T. Breton, G. Bashiardes, J.-
M. Léger, K. B. Kokoh, Eur. J. Org. Chem. 2007, 1567–1570.
[17] a) M. Marzorati, B. Danieli, D. Haltrich, S. Riva, Green Chem.
2005, 7, 310–315; b) L. Baratto, A. Candido, M. Marzorati, F.
Sagui, S. Riva, B. Danieli, J. Mol. Catal. B 2006, 39, 3–8.
[18] T. Wirth, Angew. Chem. 2005, 117, 3722; Angew. Chem. Int.
Ed. 2005, 44, 3656–3665.
One-Pot Synthesis of Cyanohydrin Acetate (6R)-5a: Oxidation of
1a and its hydrocyanation were performed as described above. For
acylation, vinyl acetate (6 equiv.) and lipase PS-C II (100 mgmL–1)
were applied. After 24 h, a reaction mixture containing (6R)-5a
(71%, 83% de) was obtained. Purification by column chromatog-
raphy (EtOAc/hexane, 6:4; Rf = 0.71) gave (6R)-5a (39%, de Ͼ
99%). [α]2D5 = +128 (c = 1.0, CHCl3). 1H NMR (500.13 MHz,
CDCl3, 25 °C): δ = 5.46 (d, J6,5 = 9.5 Hz, 1 H, 6-H), 5.44 (dd, J4,5
= 1.3, J4,3 = 3.4 Hz, 1 H, 4-H), 5.35 (dd, J3,2 = 10.8 Hz, 1 H, 3-
H), 5.15 (dd, J2,1 = 3.6 Hz, 1 H, 2-H), 5.08 (d, 1 H, 1-H), 4.32 (dd,
1 H, 5-H), 3.49 (s, 3 H, OCH3), 2.12 (s, 3 H, COCH3), 2.11 (s, 3
H, COCH3), 2.10 (s, 3 H, COCH3), 1.98 (s, 3 H, COCH3) ppm. 13
NMR (125.77 MHz, CDCl3, 25 °C): δ = 170.3 (OCOCH3), 170.1
C
2736
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 2729–2737

Chemoenzymatic Synthesis of Sugar Cyanohydrin Esters
[19] A. De Mico, R. Margarita, L. Parlanti, A. Vescovi, G. Piancat-
elli, J. Org. Chem. 1997, 62, 6974–6977.
[20] D. J. Vugts, L. Veum, K. al-Mafraji, R. Lemmens, R. F.
Schmitz, F. J. J. de Kanter, M. B. Groen, U. Hanefeld, R. V. A.
Orru, Eur. J. Org. Chem. 2006, 1672–1677.
Kanerva, K. Rahiala, O. Sundholm, Biocatalysis 1994, 10, 169–
180; d) L. Veum, L. T. Kanerva, P. J. Halling, T. Maschmeyer,
U. Hanefeld, Adv. Synth. Catal. 2005, 347, 1015–1021.
[24] See, for instance: K. Lundell, T. Raijola, L. T. Kanerva, En-
zyme Microb. Technol. 1998, 22, 86–93.
[21] a) E. Kiljunen, L. T. Kanerva, Tetrahedron: Asymmetry 1997,
8, 1225–1234; b) E. Kiljunen, L. T. Kanerva, Tetrahedron:
Asymmetry 1996, 7, 1105–1116.
[22] L. T. Kanerva, A. Liljeblad, “Transesterification – Biological”
in Encyclopedia of Catalysis (Ed.: I. T. Horváth), John Wiley &
Sons, New York, 2010; DOI: 10.1002/0471227617.
[23] a) M. Inagaki, J. Hiratake, T. Nishioka, J. Oda, J. Am. Chem.
Soc. 1991, 113, 9360–9361; b) M. Inagaki, J. Hiratake, T. Ni-
shioka, J. Oda, J. Org. Chem. 1992, 57, 5643–5649; c) L. T.
[25] a) L. S. Moon, R. S. Jolly, Y. Kasetti, P. V. Bharatam, Chem.
Commun. 2009, 1067–1069; b) L. S. Moon, M. Pal, Y. Kasetti,
P. V. Bharatam, R. S. Jolly, J. Org. Chem. 2010, 75, 5487–5498.
[26] O. Sundholm, L. T. Kanerva, J. Chem. Soc. Perkin Trans. 1
1993, 2407–2410.
[27] A. R. Moen, T. Anthonsen, Biocatal. Biotransform. 2009, 27,
226–236.
Received: January 25, 2012
Published Online: March 19, 2012
Eur. J. Org. Chem. 2012, 2729–2737
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2737