Transketolase and fructose-1,6-bis-phosphate aldolase, complementary tools for access to new ulosonic acid analogues

Article abstract of DOI:10.1016/j.tet.2003.12.069

A new approach to the synthesis of ulosonic acids KDO and DAH is described. The key step is the C₅-C₆ bond formation catalysed by fructose-1,6-bisphosphate aldolase (for KDO) or transketolase (for DAH) using substituted acrylonitrile α-hydroxyaldehyde. All asymmetric carbon configurations are determined in an enzymatic step by the means of deshydrogenase or lipase. This strategy, using a non-metabolism pathway, allows access to novel precursors of KDO, DAH and analogues.

Full text of DOI:10.1016/j.tet.2003.12.069

Tetrahedron 60 (2004) 2417–2425
Transketolase and fructose-1,6-bis-phosphate aldolase,
complementary tools for access to new ulosonic acid analogues
*
Dominique Crestia, Colette Demuynck and Jean Bolte
` ` `
Laboratoire de Synthese et Etude de Systemes a Interet Biologique, UMR 6504 du CNRS, Universite Blaise Pascal,
´ ˆ
`
63177 Aubiere Cedex, France
´
Received 14 July 2003; revised 24 November 2003; accepted 23 December 2003
Abstract—A new approach to the synthesis of ulosonic acids KDO and DAH is described. The key step is the C5–C6 bond formation
catalysed by fructose-1,6-bisphosphate aldolase (for KDO) or transketolase (for DAH) using substituted acrylonitrile a-hydroxyaldehyde.
All asymmetric carbon configurations are determined in an enzymatic step by the means of deshydrogenase or lipase. This strategy, using a
non-metabolism pathway, allows access to novel precursors of KDO, DAH and analogues.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
priate aldolase or synthetase.5 In all cases, the C3–C4 bond
is created with control of configuration in C4 (Scheme 1).
3-Deoxy-2-ulosonic acids are known as a variety of
important natural products including sialic acids (N-acetyl-
neuraminic acid and its derivatives), 3-deoxy-D-arabino-2-
heptulosonic acid 7-phosphate (DAH-7-P), 3-deoxy-D-
manno-2-octulosonic acid (KDO), and 3-deoxy-D-glycero-
D-galacto-2-nonulosonic acid (KDN). In the course of our
studies, we were especially interested in close analogues of
KDO and DAH. KDO is an integral component of the
lipopolysaccharides of Gram-negative bacteria. New synth-
eses of KDO may be useful in developing analogues capable
of disrupting the biosynthesis of bacterial cell-wall com-
ponents, and thereby lead to new antibacterial agents.1
DAH-7-P is an important intermediate in the shikimic acid
pathway along which the aromatic amino acids are
biosynthesised in bacteria, fungi and plants. Disruption of
aromatic amino acid biosynthesis (in plants) is purported to
be an extremely effective vehicle for herbicide action.1,2
This specific occurrence of DAH-7-P makes its biosynthetic
pathway a possible target. Some analogues have been tested
as herbicide agents.3 For these reasons, syntheses of KDO,
DAH and analogues have remained the subject of
investigation over the last few years. Total syntheses of
these rare carbohydrates have been based on readily
accessible sugar precursors.4 Most enzymatic approaches
involve specific aldol condensation of pyruvate or phospho-
enolpyruvate (PEP) with aldoses catalysed by the appro-
Although these methods can afford KDO and DAH in
moderate to good yields, they do not allow synthesis of all
analogues due to the specificity of the enzymes for pyruvate,
PEP and close analogues of D-arabinose.
We looked for a more versatile approach using enzymes not
involved in the biosynthetic pathway of both targeted
ulosonic acids. Our strategy was based on the formation of
the C5–C6 bonds using fructose-1,6-bis-phosphate aldolase
from rabbit muscle (RAMA) or yeast transketolase (TK).
The utility of these enzymes in organic syntheses is well
known6 for the synthesis of monosaccharide analogues.
Our retrosynthetic strategy is shown in Scheme 2. For KDO,
the key step would be the condensation of dihydroxyacetone
phosphate (DHAP) onto aldehyde A catalysed by aldolase
(RAMA). In the same way, the transfer of a hydroxyacetyl
group from hydroxypyruvic acid (HPA) onto aldehyde A,
catalysed by transketolase (TK) would afford a precursor of
3-deoxy-D-arabino-2-heptulosonic acid (DAH). In these
reactions, both enzymes catalyse the C5–C6 bond formation
and determine the stereochemistry of these centres in the
configuration present in KDO (C5, C6) and DAH (C5),
according to Scheme 2. The configuration at C6 (DAH) or
C7 (KDO) would be obtained by chemical or enzymatic
stereospecific reduction of the keto group. The acryloyl
pattern in A is a good precursor for the pyruvyl moiety of
3-deoxy-2-ulosonic acids, and numerous syntheses of
3-deoxy-D-manno-2-octulosonic acid (KDO) or sialic acids
are based on the reaction of ethyl bromomethacrylate with
aldehyde to introduce this function.7 The use of a
Keywords: C–C bond; Aldolase; Transketolase; KDO; DAH; Polyol
dehydrogenase.
*
Corresponding author. Tel.: þ33-4-73-40-71-28; fax: þ33-4-73-40-77-
17; e-mail address: jbolte@chimie.univ-bpclermont.fr
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2003.12.069

2418
D. Crestia et al. / Tetrahedron 60 (2004) 2417–2425
Scheme 1.
methylidene group as a precursor of the ketone also offers
the advantage of simplifying the identification of the various
intermediate products by limiting the number of cyclic
hemiacetal forms.
2. Results and discussion
2.1. Preparation of a-hydroxyaldehydes substrates
In a previous paper,9 we described the synthesis of synthons
1 and 2 in an enantiomerically pure form by enzymatic
resolution with Candida rugosa lipase, or by microbial
reduction of the corresponding ketones using Aspergillus
niger and Lactobacillus kefir.
This strategy could also lead to analogues of ulosonic
acids. For example, using the enantiomer of A in aldolase
catalysed synthesis would provide the C4 epimer of KDO.
Also, the 4-deoxy analogues of A are substrates of
aldolase, as was demonstrated in a preliminary report8 and
would lead to 4-deoxy KDO. Moreover, during the
reduction of C6 or C7 keto group, compounds of the L
series can be obtained.
(þ)-1 and (2)-1 were hydrolysed in the presence of Dowex
Hþ resin (Scheme 3). No racemisation occurred during the
hydrolysis as indicated by the specific rotation of the
resulting aldehydes (þ)-3 and (2)-3. In contrast, usual
acidic conditions for the acetal hydrolysis of 2 led to
racemic (^)-4. Moreover, we observed the formation of
a-hydroxyketone which implies an enediol intermediate. It
In this paper, we report the activity of the RAMA and TK
towards two precursors of aldehyde A, and the syntheses of
various KDO and DAH analogues.
Scheme 2.

D. Crestia et al. / Tetrahedron 60 (2004) 2417–2425
2419
Scheme 3.
can be supposed that the presence of the ester group
catalyses this isomerisation. To avoid this racemisation, we
tried a deprotection method under non-acidic conditions: the
deprotection of aldehyde (^)-2 was carried out by reaction
with LiBF4 in acetonitrile with 34% yield.10 This yield was
too low for the method to be considered useful for the
deprotection of the optically pure (þ)-4 and (2)-4. Racemic
4 and/or optically pure obtained aldehydes 3 were submitted
to the action of aldolase and transketolase.
2.2.1. Syntheses from aldehyde 3. We have performed the
enzymatic syntheses with each enantiomer of (^)-3
(Scheme 3). The reaction was carried out on 5 mmol of
DHAP and an excess of aldehyde in the presence of RAMA
(Scheme 4). The progress of the reaction was followed by
enzymatic assay of DHAP in the mixture. After 45 h, no
more DHAP was present in the solution. The phosphate
group was hydrolysed at pH 4 in the presence of phytase
acid from Aspergillus ficuum. After only one purification on
silica gel, the expected octuloses 5a and 5b were obtained
with 75 and 72% yield, respectively, from (R)-(þ)-3 and
(S)-(2)-3. They were easily characterized by 13C NMR
spectroscopy as well as mass spectroscopy. Compounds
5a and 5b exist in both a and b hemiacetalic forms. On
the basis of quantitative 13C NMR experiments, we have
attributed their proportions as a90%/b10%. Analysis of
400 MHz 1H NMR spectrum of 5b established the
stereochemistry of the aldol adduct. As expected, the
threo stereochemistry was obtained during reactions of
RAMA with sugar phosphates as indicated by the small
coupling constant between the C(3) and C(4) protons
(2.9 Hz). This implies a trans-diequatorial arrangement
of these protons. A range of other examples supports this
conclusion.12
2.2. Syntheses with RAMA
Aldehydes (^)-3, (þ)-(R)-3, (2)-(S)-3 and (^)-4 were
assayed as substrates for aldolase. The reaction was
monitored by enzymatic titration of residual DHAP
(DHAP was prepared and assayed according to Gefflaut
et al.11) and the characteristic constants of the reaction, the
¨
apparent Michaelis constant Km and the maximum rate Vmax
were calculated. The Vmax values are given in percent of the
activity of RAMA with the natural substrate, D-glyceralde-
hyde-3-P. The results are reported in Table 1.
Table 1. Kinetic constants for the reactions of hydroxyaldehydes 3 and 4
catalysed by aldolase (RAMA) and transketolase (TK)
Enzyme
Aldolase (RAMA)
Transketolase
We also investigated the enzymatic reactions in
preparative scale starting from racemic aldehyde (^)-
3. The two diastereomers 5a and 5b were obtained in
equivalent quantity with a overall yield of 60%, but the
purification has been difficult because two successive
separations on silica gel have been necessary to purify
them.
Substrate
V
max (%)
Km (mM)
Vmax (%)
Km (mM)
(2)-1!(2)-3
(þ)-1!(þ)-3
(^)-1!(^)-3
(^)-2!(^)-4
22
34
27
21
66
48
57
53
1
31
45
0
273
9
72
—
All tested aldehydes are substrates for RAMA. It appears
that racemic aldehydes (^)-3 and its enantiomers have no
difference in activity towards RAMA. This result confirms
that RAMA is not enantioselective toward these unphos-
phorylated a-hydroxyaldehydes.
2.2.2. Synthesis from aldehyde 4. Using the same
procedure, we have carried out the enzymatic reaction on
a preparative scale starting from the racemic aldehyde (^)-4
(Scheme 4). The two diastereomers 6a and 6b were obtained
in equivalent quantity with an overall yield of 11%. Due to
Scheme 4.

2420
D. Crestia et al. / Tetrahedron 60 (2004) 2417–2425
Scheme 5.
this low yield and the difficult separation of 6a and 6b, this
method was not pursued further.
preparative scale starting from the optically pure aldehyde
(R)-(þ)-3 with an excess of HPA. HPA was completely
consumed in 14 h. The heptulose 7 was obtained with a
yield of 73% (Scheme 5). This was better than with the
racemic acceptor substrate (^)-3 which is consistent with
kinetic studies.
2.3. Synthesis with TK
Compounds 3 and 4 were tested as substrates of TK. The
results are reported in Table 1. Unlike most a-hydroxyalde-
hydes, theester(^)-4wasnotasubstrateforTK. However, the
nitrile (R)-(þ)-3 displays a remarkable reactivity with a low
Km value of 9 mM. As expected, the (S)-(2)-3 is not a
substrate. The low reactivity was probably due to the presence
of trace amount of the R enantiomer. These results allowed us
to attribute the absolute configuration of (2)-3 and (þ)-3 in
our previous (^)-3 enzymatic resolution study.9 The Km value
of racemic (^)-3 is higher than the supposed value of about
18 mM. The enantiomer (S)-(2)-3 could inhibit the enzyme
and thus explains such a difference.
2.4. Reduction of the carbonyl group
The synthesis of the keto acids first requires the reduction of
the carbonyl group of both octuloses 5a and 5b and the
heptulose 7. In spite of our efforts, all the tested chemical
methods have resulted in the reduction of the carbon–
carbon double bond.
To solve this problem, reduction of the carbonyl using
isolated enzyme or microorganisms was investigated.
Since TK exhibits a high enantioselectivity towards racemic
a-hydroxyaldehydes, in spite of the higher Km value, the
synthesis could be carried out from racemic (^)-3 or from
the R enantiomer.
2.4.1. Reduction of carbonyl group with purified enzyme.
For each RAMA and TK adducts, respectively, 5a, 5b and 7,
enzymatic reduction was performed at pH 7. This utilised
available iditol dehydrogenase (IDH) from Candida utilis
coupled with the formate dehydrogenase (FDH), and
sodium formate to regenerate the NADH in situ (Scheme
6). Progress of the reaction was monitored by thin-layer
chromatography (silica gel, 10% methanol in dichloro-
methane). Due to the diastereostelectivity of the reaction
catalysed by IDH,13 the newly created asymmetric centres
in C6 or C7 are of the S configuration. Three compounds 8a,
8b and 9 were obtained in yields from 66 to 69%. They were
easily characterised by 13C NMR spectroscopy since all
signals were assigned: J mod experiments confirmed that
only one diastereomer has been formed in each reduction.
There are no reported enzymes which form the opposite R
enantiomer in C6 or C7. Therefore, whole cell microorgan-
isms were tested for this activity (Scheme 6).
2.3.1. Synthesis from aldehyde (6)-3. Racemic a-hydro-
xyaldehyde (^)-3 and one half equivalent of b-hydro-
xypyruvic acid (HPA) were reacted in the presence of the
TK from S. cerevisiae and their required cofactors (thiamine
pyrophosphate TPP and magnesium) (Scheme 5). The
experiment was performed on a 5 mmol scale of the
acceptor substrate in TRIS buffer at pH 7.5. Progress of
the reaction was monitored by thin-layer chromatography
(silica gel, 10% methanol in dichloromethane) and by
enzymatic assay of HPA in the mixture. After 14 h, no more
HPA was present in the solution. The heptulose 7 was
obtained after silica gel chromatography with a yield of
38%. The heptulose 7 was easily characterised by 13C NMR
spectroscopy since all signals were assigned: J mod
experiment confirmed that only one diastereomer had been
formed. This allowed us to assign the C6 carbon of the
carbonyl group and the C2 carbon of the hydroxymethyl
group, respectively, near 213 and 68 ppm. CI mass
spectroscopy results were also consistent with the proposed
structure with a signal at 186 (MþHþ) for ketose 7.
2.4.2. Reduction of carbonyl group with microorgan-
isms. Compounds 5a, 5b and 7 were submitted to the action
of baker’s yeast, using commercial freeze-dried cells under
non-fermenting conditions. In each case, only one com-
pound according to J mod NMR spectrum has been obtained
with a modest yield (25–30%). We assigned the structures
8a, 8b and 9 to these reduction products on the basis of their
NMR spectra. This conclusion was confirmed by the optical
values. In this reduction, enzyme acting in baker’s yeast
possesses the same stereostelectivity as IDH leading to
saccharides in the S series. For large-scale synthesis, the less
expensive baker’s yeast method would be preferred.
However, the biotransformations performed in the presence
of washed resting cells of Aspergillus niger appeared to be
less stereoselective because we obtained a mixture of two
The residual aldehyde (S)-(2)-3 was isolated in 43% yield.
Nevertheless, the separation of the heptulose 7 and
unreacted aldehyde (S)-(2)-3 has been difficult. Due to
this, we investigated an enzymatic synthesis catalysed by
TK from aldehyde (R)-(þ)-3.
2.3.2. Synthesis from aldehyde (1)-R-3. Using the same
procedure, we have carried out the enzymatic reaction on

D. Crestia et al. / Tetrahedron 60 (2004) 2417–2425
2421
Scheme 6. Key: (i) IDH (EC 1.1.1.14), FDH (EC 1.1.1.14), NADH, HCO2Na; (ii) NaH, BrBn, IN(Bu)4; (iii) 1. NaOH, 110 8C, 9 days, 2. MeI, DMF.
diastereoisomers, according to NMR 13C experiments.
Moreover, the yields in these microbial reductions are
low. Nevertheless, the use of Aspergillus niger for the
bioconversions would allow the synthesis of alcohols with
the R absolute configuration at C6 or C7.
3. Conclusion
We have demonstrated that complex monosaccharides such
as ulosonic acids can be synthesised by a chemo-enzymatic
strategy different from their biosynthesis. In our study, all
the asymmetric carbon configurations were determined by
an enzymatic step: a lipase resolution or a microbial
reduction for C4, an aldolase catalysed C–C bond formation
for C5–C6 in the case of KDO and a TK catalysed C–C
bond formation for C5–C6 in the case of DAH. Finally, C7
(KDO or epimer) and C6 (DAH or epimer) were obtained by
enzymatic or microbial reduction. The key step in the
synthesis was catalysed by RAMA and TK which appear as
complementary tools in organic chemistry.
After the reduction step, the deprotection of the keto acid
group was required. This necessitates the hydrolysis of the
nitrile group and the oxidation of the carbon–carbon double
bond.
2.5. Hydrolysis of nitrile group
The hydrolysis step must take place before the oxidation step
in order to avoid the formation of an unstable a-ketonitrile
group. Initially, we wanted to hydrolyse enzymatically the
compounds 8a, 8b and 9. Indeed, mild reaction conditions
(pH 7 and room temperature) do not necessitate hydroxyl
groups protection. But, preliminary analytic assays have
showed that the compounds 8a, 8b and 9 were not substrates
of nitrile hydratase or nitrilase from Rhodococcus equi A4.
Because of this we performed the hydrolysis step chemi-
cally. First, the hydroxyl groups were protected as benzyl
ethers by reaction with benzyl bromide. Compounds 10a,
10b and 11 were obtained in yields from 61 to 69%
(Scheme 6). Secondly, we carried out the hydrolysis of the
nitrile group at 110 8C under alkaline conditions.14 The
compounds 10b and 11 were stable and the hydrolysis of
nitrile function is complete after 9 days. The corresponding
carboxylates were esterified with methyl iodide. Methyl
esters 12 and 13 have been isolated with yields of 78 and
73%, respectively (Scheme 6). These were easily charac-
terized by 13C NMR spectroscopy since all signals were
assigned: J mod experiments confirmed the disappearance
of the nitrile function (about 119 ppm) and the presence of
methyl ester near 177 ppm.
Thus, this method is a novel and versatile approach to the
chemoenzymatic synthesis of ulosonic acids. Indeed, it
allowed the preparation of various precursors and analogues
of DAH and KDO which are inaccessible by other
chemoenzymatic or chemical approaches.
4. Experimental
4.1. General methods
Rabbit muscle aldolase (E.C. 4.1.2.13), acid phytase from
Aspergillus ficuum (E.C. 3.1.3.8), L-sorbitol dehydrogenase
from sheep liver (E.C. 1.1.1.14), and formate dehydrogen-
ase from yeast (E.C. 1.2.1.2) were purchased from Sigma.
1H (400.134 MHz) and 13C (100.61 MHz) spectra were
recorded on a Bruker AC 400 spectrometer. Mass spectra
were obtained on a Helwett–Packard 5989 A spectrometer.
Optical rotations were determined on a JASCO polarimeter.
IR spectra were recorded on a Perkin–Elmer 881 spectro-
photometer. UV analyses were performed on a Hitachi
(U-2010) spectrophotometer. Solvents were distilled over
an appropriate desiccant and stored under argon. Trans-
ketolase was produced and purified in our laboratory as
precedently described.16 All reagents were obtained from
Aldrich, TLC plates of silica gel 60F254 from Merck.
Completion of the synthesis of KDO and DAH is simple
since these a-methylene esters are known to lead to ulosonic
acids by oxidation of the carbon–carbon double bond
according to methods widely described in the literature.15

2422
D. Crestia et al. / Tetrahedron 60 (2004) 2417–2425
Merck silica gel for column chromatography 60/230–400
and 60/40–63 mesh were used. Aldol condensation was
monitored enzymatically by DHAP consumption.17 The
phosphate-catalysed hydrolysis was monitored by TLC
(silica gel from Merck). DHAP was prepared and assayed
according to Gefflaut et al.11 The microorganisms were all
laboratory-grown except freeze-dried baker’s yeast, which
was a commercial product (VAHINE, Monteux, France).
Preculture and culture conditions for the fungus Aspergillus
niger ATCC 9142 have already been described elsewhere.18
5b (major, 88%) d: 134.2 (C9); 121.3 (C8); 120.0 (C7);
104.3 (C2); 81.0/78.2/77.6 (C3, C4, C5); 65.7 (C1); 35.7
(C6). 5b (minor, 12%) d: 134.4 (C9); 121.45 (C8); 120.2
(C7); 107.4 (C2); 81.8/78.6/78.3(C3, C4, C5); 66.7 (C1);
36.9 (C6). IR (neat) 3351 (O–H); 2360 (CuN); 1641
(CvC) cm21. MS (CI/CH4)¼m/z 256 ((MþC3H5)þ); 244
((MþC2H5)þ); 216 ((MþH)þ); 198 ((MH2H20)þ).
0
0
´
4.2.3. (1)-6-Deoxy-6-(1 -ethyl-oxycarbonyl-2 -0vinyl)-
´
(a,b)-D-fructofuranose 6a and (2)-6-deoxy-6-(1 -ethyl-
oxycarbonyl-20-vinyl)-(a,b)-D-sorbofuranose 6b. To
solution of 4-hydroxy-2-methylene-5-oxo-pentanoic acid
ethyl ester 4 (26 mL) prepared as described previously
(dimethyl acetal; 5.34 mmol) was mixed 17.4 mL of a
solution containing 3.87 mmol of DHAP, and the resulting
solution was adjusted to pH 7.5 with 1 N NaOH. 250 U of
RAMA were added. The total volume was ca. 43.7 mL. At
24 and 48 h, 50 U of RAMA were added. After 56 h DHAP
analysis indicated 100% conversion, the pH was adjusted to
3.9 with 1 N HCl. Acid phytase (250 U) was added and the
mixture was heated at 30 8C for 24 h. Water was removed
under reduced pressure. The two diastereoisomers 6a and 6b
were purified and separated by silica gel chromatography
(CH2Cl2/MeOH, 98:2 to 85:15) to yield 111.5 mg (11%)
(6a/6b¼1:1) as colorless oil.
4.2. Synthesis with RAMA
4.2.1. (1)-6-Deoxy-6-(10-cyanovinyl)-(a,b)-D-fructofura-
nose 5a. (S)-(2)-4-Hydroxy-2-methylene-5-oxo-pentane-
nitrile 3 was formed in situ as follows: the dimethyl acetal
(S)-(2)-5,5-dimethoxy-4-hydroxy-2-methylene-pentane-
nitrile 1 (7.63 mmol) was dissolved in water (33 mL), and
Dowex 50W-X8 (Hþ form, 200–400 mesh) was added until
pH,2. The mixture was heated at 45 8C for 12 h, and then
the resin was filtered off and washed with water. A solution
of DHAP (29.07 mL, 5.56 mmol) was added, and the
mixture was adjusted to pH 7.5 with 1 N NaOH. RAMA
(250 U) was added, the mixture (total volume¼50.9 mL)
was stirred gently at rt. After 48 h DHAP analysis indicated
100% conversion, the pH was adjusted to 3.9 with 1 N HCl.
Acid phytase (250 U) was added and the mixture was heated
at 30 8C for 24 h. Water was removed under reduced pressure,
and (þ)-6-deoxy-6-(10-cyanovinyl)-(a,b)-D-fructofuranose 5a
was purified by silica gel chromatography (CH2Cl2/MeOH,
90:10) to yield 897.4 mg as colorless oil (4.17 mmol: 75% from
DHAP). [a]2D5¼þ12.7 (c 2.22, MeOH). Rf¼0.27 (CH2Cl2/
MeOH, 90:10). 1H NMR (400.134 MHz, CD3OD) d: 6.16 (s,
0
0
´
4.2.3.1. (1)-6-Deoxy-6-(1 -ethyl-oxycarbonyl-2 -
vinyl)-(a,b)-D-fructofuranose 6a. [a]2D5¼þ8 (c 2.25,
MeOH). Rf¼0.15 (CH2Cl2/MeOH, 85:15). 1H NMR
(400.134 MHz, CD3OD) d: 6.13 (s, 1H, H9); 6.01 (s, 1H,
0
H9 ); 4.30 (q, 2H, J¼7.5 Hz, CH2 ester); 3.85–3.92 (m, 1H,
H4); 3.86 (d, 1H, J¼7.5 Hz; H3); 3.61–3.66 (m, 1H, H5);
3.37 (s, 2H, H1); 2.36–2. 19 (m, 2H, H6); 1.37 (t, 3H,
J¼7.5 Hz, CH3 ester). 13C NMR (100.61 MHz, CD3OD) d:
6a (major, 78%) 167.9 (C8); 144.9 (C7); 125.7 (C9); 104.5
(C2); 79.2/78.7/77.1 (C3, C4, C5); 66.3 (C1); 60.4 (CH2
ester); 36.3 (C6); 19.5 (CH3 ester). 6a (minor, 22%) 167.9
(C8); 144.7 (C7); 125.3 (C9); 108.1 (C2); 84.4/81.4/78.25
(C3, C4, C5); 65.4 (C1); 60.4 (CH2 ester); 33.7 (C6); 19.5
(CH3 ester). IR (neat) 3354 (O–H); 1735 (CvO); 1637
(CvC); 1041 (C–O) cm21. MS (CI/CH4)¼m/z 263
((MþH)þ); 245 ((MH2H20)þ).
0
1H, H9); 6.14 (s, 1H, H9 ); 4.25–4.10 (m, 2H, H3 and H4); 4.05–
4.00 (m, 1H, H5); 3.8 (s, 2H, H1); 2.82–2.70 (m, 2H, H6). 13
C
NMR (100.61 MHz, CD3OD) d:5a (major, 90%) d: 134.5 (C9);
120.7 (C8); 119.9 (C7); 103.4 (C2); 81.7/80.2/79.9 (C3, C4,
C5); 64.6 (C1); 39.0 (C6). 5a (minor, 10%) d: 134.6 (C9); 120.8
(C8); 120.1 (C7); 106.3 (C2); 84.2/80.8/80.05 (C3, C4, C5);
68.0(C1);39.9(C6). IR(neat)3351(O–H);2360(CuN);1641
(CvC) cm21. MS (CI/CH4)¼256 ((MþC3H5)þ); 244
((MþC2H5)þ); 216 ((MþH)þ); 198 ((MH2H20)þ).
4.2.2. (2)-6-Deoxy-6-(10-cyanovinyl)-(a,b)-L-sorbofura-
nose 5b. To solution of (R)-(þ)-4-hydroxy-2-methylene-5-
oxo-pentanenitrile 3 (32 mL) prepared as described previously
(dimethyl acetal; 6.56 mmol) was mixed 22.4 mL of a solution
containing 4.72 mmol of DHAP, and the resulting solution was
adjusted to pH 7.5 with 1 N NaOH. 250 U of RAMA were
added. The total volume was ca. 57.7 mL. After 46 h DHAP
analysis indicated 100% conversion, the pH was adjusted to 3.9
with 1 N HCl. Acid phytase (250 U) was added and the mixture
was heated at 30 8C for 24 h. Water was removed under
reduced pressure, and (2)-6-deoxy-6-(10-cyanovinyl)-(a,b)-
L-sorbofuranose 5b was purified by silica gel chromato-
graphy (CH2Cl2/MeOH, 90:10) to yield 730.6 mg as
colorless oil (72% from DHAP). [a]2D5¼222.6 (c 1.72,
MeOH). Rf¼0.39 (CH2Cl2/MeOH, 90:10). 1H NMR
(400.134 MHz, CD3OD) d: 5.99 (s, 1H, H9); 5.98 (s, 1H,
0
0
´
4.2.3.2. (2)-6-Deoxy-6-(1 -ethyl-oxycarbonyl-2 -
vinyl)-(a,b)-D-sorbofuranose 6b. [a]2D5¼25.6 (c 1.5,
MeOH). Rf¼0.25 (CH2Cl2/MeOH, 85:15). 1H NMR
(400.134 MHz, CD3OD) d: 6.10 (s, 1H, H9); 6.05 (s, 1H,
0
H9 ); 4.25 (q, 2H, J¼6.5 Hz, CH2 ester); 4.04–4.14 (m, 2H,
H4 et H5); 3.89 (d, 1H, J¼4.5 Hz, H3); 3.42 (s, 2H, H1);
2.09–2.26 (m, 2H, H6); 1.25 (t, 3H, J¼6.5 Hz, CH3 ester).
13C NMR (100.61 MHz, CD3OD) d: 6b (major, 83%) 167.9
(C8); 144.7 (C7); 125.5 (C9); 103.8 (C2); 80.5/79.4/78.7
(C3, C4, C5); 66.4 (C1); 61.4 (CH2 ester); 34.7 (C6); 21.5
(CH3 ester). 6b (minor, 17%) 167.9 (C8); 144.9 (C7); 125.0
(C9); 105.7 (C2); 79.6/78.9/77.7 (C3, C4, C5); 67.1 (C1);
61.4 (CH2 ester); 34.7 (C6); 21.5 (CH3 ester). IR (neat) 3354
(O–H); 1735 (CvO); 1637 (CvC); 1041 (C–O) cm21. MS
(CI/CH4)¼m/z 263 ((MþH)þ); 245 ((MH2H20)þ).
4.3. Synthesis with TK
0
H9 ); 4.54 (ddd, 1H, J¼8.8, 4.9, 4.4 Hz, H5); 4.27 (dd, 1H,
4.3.1. (2)-(4R,5S)-4,5,7-Trihydroxy-2-methylene-6-oxo-
heptanenitrile 7. 40 mL of enzyme extract containing
200 U of TK, 290 mg of Tris buffer (50 mM, pH 7.5), 0.6 g
of (R)-(þ)-4-hydroxy-2-methylene-5-oxo-pentanenitrile 3
J¼4.4, 2.9 Hz, H4); 4.27 (d, 1H, J¼2.9 Hz, H3); 3.57 (s, 2H,
H1); 2.75 (dd, 1H, J¼14.8, 4.9 Hz, H6); 2.68 (dd, 1H,
J¼14.8, 8.8 Hz, H6 ). 13C NMR (100.61 MHz, CD3OD) d:
0

D. Crestia et al. / Tetrahedron 60 (2004) 2417–2425
2423
(4.79 mmol, 100 mM), 0.55 g of commercial hydroxy-
pyruvate (5 mmol, 100 mM), 19.5 mg of MgCl2
(9.58.1022 mmol, 2 mM), and 66 mg (1.44.1022 mmol,
3 mM) thiamine pyrophosphate (TPP) were placed in a
round-bottomed flask. The reaction mixture was deoxygen-
ated with argon and left at rt in the dark. After complete
disappearance of hydroxypyruvate (14 h), MeOH (200 mL)
was added and the precipitate was removed by centrifu-
gation. The supernatant was concentrated under reduced
pressure to a thick syrup. Purification of this syrup by
chromatography on silica gel afforded (2)-(4R,5S)-4,5,7-
trihydroxy-2-methylene-6-oxo-heptanenitrile 7 as a color-
less oil (648 mg, 73%). [a]2D5¼þ8.5 (c 2.17, MeOH).
Rf¼0.48 (CH2Cl2/MeOH, 8:2). 1H NMR (400.134 MHz,
(400.134 MHz, CD3OD) d: 6.0 (s, 1H, H8); 5.93 (s, 1H,
0
H8 ); 3.96 (m, 1H, J¼8.9, 4.4 Hz, H6); 3.77 (dd, 1H, J¼9.2,
4.8 Hz, H5); 3.68 (dd, 1H, J¼14.9, 7.0 Hz, H7); 3.63 (dd,
0
1H, J¼14.9, 5.6 Hz, H7 ); 2.58 (dd, 1H, J¼8, 4.2 Hz, H3);
2.52 (dd, 1H, J¼9, 5.5 Hz, H3 ). 13C NMR (100.61 MHz,
0
CD3OD) d: 134.1 (C8); 122.1 (C2); 120.1 (C1); 74.1 (C5);
74.0 (C6); 71.7 (C4); 64.5 (C7); 40.0 (C3). IR (neat) 3364
(O–H); 2227 (CN) cm21
((MþH)þ).
.
MS (GC/CI)¼m/z 188
4.4.2. General procedure for the reduction of RAMA
and TK adducts with freeze-dried baker’s yeast or
Aspergillus niger. After culture at 27 8C for the times as
indicated by Besse and co-workers, the fungus was filtered
on sintered glass or centrifuged, and then washed four times
with NaCl solution (8 g L21). Mycelium or freeze-dried
baker’s yeast (5 g) was placed in a 500 mL conical flask
with 50 mL of distilled water and 50 mL of substrate. After
incubation at 27 8C on a rotating table set at 200 rpm, the
mixture was filtered on sintered glass or centrifuged for
10 min at 8000 rpm. The liquor was then continuously
extracted with ethyl acetate for 24 h. The ethyl acetate phase
was dried on MgSO4 and the solvent evapored off under
vaccum. The products extracted after bioconversion were
separated on a silica gel column. The eluent was CH2Cl2/
MeOH, 95:5 to 80:20. Bioconversions were stopped when
the ketose was disappeared completely as indicated by TLC
analysis (CH2Cl2/MeOH, 80:20).
0
CDCl3) d: 6.25 (s, 1H, H8); 6.15 (s, 1H, H8 ); 4.76 (d, 1H,
0
J¼12 Hz, H7); 4.64 (d, 1H, J¼12 Hz, H7 ); 4.50 (d, 1H,
J¼3 Hz, H5); 4.39 (m, 1H, H4); 2.73 (d, 2H, J¼6 Hz, H3).
13C NMR (100.61 MHz, CDCl3) d: 213.4 (C6); 134.6 (C8);
121.3 (C2); 119.8 (C1); 78.9 (C5); 71.6 (C4); 68.1 (C7);
39.55 (C3). IR (neat) 3364 (O–H); 2227 (CN); 1733 (CvO)
cm21. MS (CI/CH4)¼m/z 186 ((MþH)þ). HRMS calcd for
208.0586, found 208.0587 (MþH)þ.
4.4. Enzymatic reduction of RAMA and TK adducts
4.4.1. General procedure for the reduction of RAMA
and TK adducts with sorbitol dehydrogenase. To a
solution of RAMA or TK adduct 5a, 5b or 7 (0.1 mM) in
100 mM phosphate buffer (pH 7.0) were added sodium
formate (3 equiv.), NADH sodium salt (0.05 equiv.),
L-sorbitol dehydrogenase (E.C. 1.1.1.14, 15 U) and formate
dehydrogenase (E.C. 1.2.1.2, 15 U). The mixture reaction
was stirred at room temperature until the ketose was
disappeared completely as indicated by TLC analysis
(CH2Cl2/MeOH, 80:20). The mixture was concentrated
under reduced pressure and chromatographed over silica gel
(CH2Cl2/MeOH, 95:5 to 85:15) to give purified product as
8a, 8b or 9.
The results were reported on Table 2.
Table 2. Enzymatic and microbial reduction of the keto group of the KDO
and DAH precursors 5 and 7
IDH
Baker’s yeast
Aspergillus niger
[a]2D5 8C Yield (%) [a]D25 8C Yield (%) [a]D25 8C Yield (%)
5a 211.7 69
(8a, (S)-C7)
5b 211.2 67
(8b, (S)-C7)
þ4.6 66
(9, (S)-C6)
210.9 31
210.7 23
þ5.6
24.9
22
17
6
4.4.1.1. (1)-(4R,5R,6S,7S)-4,5,6,7,8-Pentahydroxy-2-
methylene-octanenitrile 8a. [a]2D5¼þ11.7 (c 0.7, MeOH).
1
Rf¼0.25 (CH2Cl2/MeOH, 80:20). H NMR (400.134 MHz,
7
þ5
24
7
0
CD3OD) d: 6.20 (s, 1H, H9 ); 6.11 (s, 1H, H9); 4.17 (m, 1H,
H6); 4.02 (m, 1H, H5); 3.95 (m, 1H, H7); 3.85 (m, 1H, H4);
3.75 (m, 1H, H8); 2.95 (dd, 1H, H3, J¼15, 8 Hz); 2.85 (dd,
1H, H3 , J¼15, 6 Hz). 13C NMR (100.61 MHz, CD3OD) d:
0
4.4.3. General procedure for the benzylation of L-SDH
products. To a stirred solution of compound 8a, 8b, or 9
(60 mM) in dry THF was added slowly NaH (1 equiv. per
hydroxyl group) at 220 8C under inert atmosphere. After
the addition, the mixture reaction was homogeneous,
IN(Bu)4 (0.5 equiv. per hydroxyl group), and benzyl
bromide (1 equiv. per hydroxyl group) were added at
room temperature. The mixture was stirred for 1 h, then
evaporated. Column chromatography (cyclohexane/ethyl
acetate, 8:2) gave purified penta or tetrabenzylated com-
pound as 10a, 10b or 11.
134.0 (C9); 122.0 (C2); 120.1 (C1); 75.1/74.8/73.5/72.6
(C4, C5, C6, C7); 64.7 (C8); 40.1(C3). IR (neat) 3354 (O–
H); 2232 (CN); 1647 (CvC) cm21. MS (CI/CH4)¼m/z 218
((MþH)þ).
4.4.1.2. (2)-(4S,5R,6S,7S)-4,5,6,7,8-Pentahydroxy-2-
methylene-octanenitrile 8b. [a]2D5¼210.9 (c 1.6,
MeOH). Rf¼0.17 (CH2Cl2/MeOH, 80:20). 1H NMR
0
(400.134 MHz, CD3OD) d: 6.20 (s, 1H, H9 ); 6.11 (s, 1H,
H9); 4.17 (m, 1H, H6); 4.02 (m, 1H, H5); 3.95 (m, 1H, H7);
3.85 (m, 1H, H4); 3.75 (m, 1H, H8); 2.95 (dd, 1H, H3, J¼15,
8 Hz); 2.85 (dd, 1H, H3 , J¼15, 6 Hz). 13C NMR
0
4.4.3.1. (1)-(4R,5R,6S,7S)-4,5,6,7,8-Pentakis-benzyl-
oxy-2-methylene-octanenitrile 10a. Yield¼61% (colorless
oil). [a]2D5¼þ16.9 (c 1.565, CHCl3). Rf¼0.48 (cyclohexane/
ethyl acetate, 8:2). 1H NMR (400.134 MHz, CDCl3) d: 7.45
(100.61 MHz, CD3OD) d: 134.1 (C9); 122.1 (C2); 120.0
(C1); 74.8/73.55/72.9/71.2 (C4, C5, C6, C7); 64.6 (C8);
40.0 (C3). IR (neat) 3354 (O–H); 2232 (CN); 1647 (CvC)
cm21. MS (CI/CH4)¼m/z 218 ((MþH)þ).
0
(m, 25H, H ar.); 5.85 (s, 1H, H9); 5.60 (s, 1H, H9 ); 4.85–
´
4.4.1.3. (1)-(4R,5S,6S)-4,5,6,7-Tetrahydroxy-2-
4.35 (m, 10H, 5£OCH2); 4.06 (m, 1H, H6); 3.9 (m, 1H, H5);
methylene-heptanenitrile 9. [a]2D5¼þ4.6 (c 1.22,
3.84 (m, 1H, H7); 3.71 (m, 3H, H4 and H8); 3.60 (dd, 1H,
0
MeOH). Rf¼0.48 (CH2Cl2/MeOH, 80:20). 1H NMR
J¼15, 4 Hz, H8 ); 2.675 (dd, 1H, J¼16, 4 Hz, H3); 2.60 (d,

2424
D. Crestia et al. / Tetrahedron 60 (2004) 2417–2425
1H, J¼16 Hz, H3 ). 13C NMR (100.61 MHz, CDCl3) d:
CDCl3) d: 176.5 (C1); 135.8–135.3 (Cipso); 134.3 (C8);
124.6 (C2); 75.7 (C5); 74.6 (C6); 73.9 (C4); 71.6/71.1/70.6/
70.4 (4£OCH2); 66.9 (C7); 33.5 (C3); 15.8 (CH3 ester). IR
(neat) 2927 (C–H); 1732 (CvO); 1667 (CvC); 1090 (C–
O) cm21. MS (CI/isobutane)¼m/z 581 ((MþH)þ).
0
138.6/138.4/138.3/138.2/138.1 (Cipso); 132.6 (C9); 128.5–
127.6 (Cortho, Cmeta and Cpara); 120.7 (C2); 119.1 (C1);
´
79.92/79.25/79.1/78.4 (C4, C5, C6, C7); 74.9/74.4/73.4/
72.8/72.1 (5£OCH2); 69.6 (C8); 35.9 (C3). IR (neat) 3367
(O–H); 2226 (CN); 1667 and 1621 (CvC); 1047 (C–O)
cm21. MS (CI/CH4)¼m/z 688 ((MþH)þ).
4.4.3.2. (2)-(4S,5R,6S,7S)-4,5,6,7,8-Pentakis-benzyl-
oxy-2-methylene-octanenitrile 10b. Yield¼58% (colorless
oil). [a]2D5¼215.5 (c 2.06, CHCl3). Rf¼0.31 (cyclohexane/
ethyl acetate, 8:2). 1H NMR (400.134 MHz, CDCl3) d: 7.45
References and notes
1. Unger, F. M. Adv. Carbohydr. Chem. Biochem. 1981, 38,
323–388.
0
(m, 25H, H ar.); 6.01 (s, 1H, H9); 5.94 (s, 1H, H9 ); 4.90–
4.42 (m, 10H, 5£OCH2); 4.03 (m, 1H, H6); 3.92 (m, 1H,
2. Reimer, L. M.; Conley, D. L.; Pompliano, D. L.; Frost, J. W.
J. Am. Chem. Soc. 1986, 108, 8010. Draths, K. M.; Frost, J. W.
J. Am. Chem. Soc. 1990, 112, 1657.
H5); 3.87 (m, 1H, H7); 3.74 (m, 3H, H4 and H8); 3.63 (dd,
0
1H, J¼15, 4 Hz, H8 ); 2.68 (dd, 1H, J¼16, 4 Hz, H3); 2.62
(d, 1H, J¼16 Hz, H3 ). 13C NMR (100.61 MHz, CDCl3) d:
0
´
3. Le Marechal, P.; Froussios, C.; Level, M.; Azerad, R.
138. 6/138.4/138.3/138.2/138.1 (Cipso); 132.3 (C9); 128.5–
127.6 (Cortho, Cmeta and Cpara); 120.6 (C2); 119.0 (C1); 79.7/
Carbohydr. Res. 1981, 94, 1.
´
4. (a) Ghalambor, M. A.; Levine, E. M.; Heath, E. C. J. Biol.
Chem. 1966, 241, 3207–3215. (b) Frick, W.; Kru¨lle, T.;
Schmidt, R. R. Liebigs Ann. Chem. 1991, 435. (c) Burke, S. D.
Org. Lett. 1999, 1, 71–74. (d) Adlersberg, M.; Sprinson, D. B.
Carbohydr. Res. 1984, 127, 9. (e) Hermann, K. M.; Poling,
M. D. J. Biol. Chem. 1975, 250, 6817. (f) Ramage, R.;
MacLeod, A. M.; Rose, G. W. Tetrahedron 1991, 47, 5625. (g)
Lopez-Herrera, F. P.; Sarabia-Garcia, F. Tetrahedron Lett.
1994, 36, 6705. (h) Barton, D. H. R.; Liu, W. Tetrahedron Lett.
1997, 38, 367. (i) Barton, D. H. R.; Liu, W. Tetrahedron Lett.
1997, 53, 12067.
79.1/78.8/78.2 (C4, C5, C6, C7); 74.9/74.4/73.4/72.8/72.1
(5£OCH2); 69.3 (C8); 33.9 (C3). IR (neat) 3358 (O–H);
2227 (CN); 1667 and 1620 (CvC); 1053 (C–O) cm21. MS
(CI/CH4)¼m/z 688 ((MþH)þ).
4.4.3.3. Heptanenitrile 4,4,3,3(1)-(4R,5S,6S)-4,5,6,7-
´
tetrakis-benzyloxy-2-methylene-11. Yield¼44% (colorless
oil). [a]2D5¼þ7.1 (c 3.95, CHCl3). Rf¼0.44 (cyclohexane/
ethyl acetate, 8:2). 1H NMR (400.134 MHz, CDCl3) d:
7.42–7.27 (m, 20H, H ar.); 5.83 (s, 1H, H8); 5.52 (s, 1H,
0
H8 ); 4.8–4.48 (m, 8H, 4£OCH2); 3.92 (m, 1H, H6); 3.88
(m, 1H, H5); 3.68 (m, 2H, H4 and H7); 3.60 (dd, 1H, J¼10,
5. (a) Sugai, T.; Shen, G.-J.; Ichikawa, Y.; Wong, C.-H. J. Am.
0
5.3 Hz, H7 ); 2.56 (dd, 1H, J¼14.5, 4.4 Hz, H3); 2.41 (dd,
´
Chem. Soc. 1993, 115, 413. (b) Auge, C.; Delest, V.
Tetrahedron: Asymmetry 1995, 6, 863.
1H, J¼14.5, 8.2 Hz, H3 ). 13C NMR (100.61 MHz, CDCl3)
0
d: 135–134.4 (Cipso); 129.7 (C8); 125.3–124.5 (Cortho
Cmeta and Cpara); 117.1 (C2); 115. 6 (C1); 75.6 (C5); 74.2
,
6. (a) Machajewski, T. D.; Wong, C.-H. Angew. Chem., Int. Ed.
2000, 39, 1352. (b) Kobori, Y.; Myles, D. C.; Whitesides,
G. M. J. Org. Chem. 1992, 57, 5899–5907. (c) Wymer, N.;
Toone, E. J. Curr. Opin. Chem. Biol. 2000, 4, 110. (d) Wong,
C.-H.; Whitesides, G. M. J. Org. Chem. 1983, 48, 3199. (e)
Effenberger, F.; Null, V. Liebigs Ann. Chem. 1992,
1211–1212. (f) Ziegler, T.; Straub, A.; Effenberger, F.
Angew. Chem., Int. Ed. Engl. 1988, 27, 716–717. (g) Bolte,
J.; Demuynck, C.; Samaki, H. Tetrahedron Lett. 1987, 28,
5525–5528.
´
(C6); 73.7 (C4); 71.2/70.2/70.1/69.6 (4£OCH2); 66.3 (C7);
33.0 (C3). IR (neat) 2236 (CN); 1659 (CvC); 1027 (C–O)
cm21. MS (CI/CH4)¼m/z 688 ((MþH)þ). HRMS calcd for
548.2801, found 548.2804 (MþH)þ.
4.4.4. General procedure for the hydrolysis of nitrile
group. To a solution of penta- or tetrabenzylated compound
as 10b or 11 (3 mM) in ethylene glycol, water, THF, and
MeOH (1:1:1:2) was added NaOH (4 equiv.). The mixture
was heaten at 110 8C for 9 days. Then at rt, carboxylate
group was esterified by addition of DMF (1 mM), and
methyl iodide (10 equiv.) The mixture reaction was stirred
for 24 h, then diluted with water, extracted with CH2Cl2,
dried (MgSO4) and evaporated. Column chromatography
(cyclohexane/ethyl acetate, 8:2) gave purified methyl ester
as 12 or 13.
7. (a) Enoch, K.; Gordon, D. M.; Whitesides, W. G. J. Org.
Chem. 1993, 58, 5500–5507. (b) Chan, T.-H.; Lee, M.-C.
J. Org. Chem. 1995, 60, 4228–4232. (c) Prenner, R. H.;
Schmid, W. Monatsh. Chem. 1996, 127, 1045–1050. (d)
Martin, L. D.; Stille, J. K. J. Org. Chem. 1982, 47,
3630–3633.
´
8. Guerard, C.; Demuynck, C.; Bolte, J. Tetrahedron Lett. 1999,
40, 4181–4182.
4.4.4.1. (2)-(4S,5R,6S,7S)-4,5,6,7,8-Pentakis-benzyl-
oxy-2-methylene-octanoic acid methyl ester 12.
Yield¼78% (colorless oil). [a]2D5¼24.1 (c 1.0, CHCl3).
Rf¼0.33 (cyclohexane/ethyl acetate, 8:2). 13C NMR
(100.61 MHz, CDCl3) d: 176.5 (C1); 138.8–138.3 (Cipso);
134.1 (C9); 125.2 (C2); 79.5/78.6/78.1/78 (C4, C5, C6, C7);
74.3/73.3/73.1/72.6/72.4 (5£OCH2); 69.9 (C8); 33.7 (C3);
14.8 (CH3 ester). IR (neat) 2923 (C–H); 1731 (CvO); 1667
(CvC); 1090 (C–O) cm21. MS (CI/isobutane)¼m/z 701
((MþH)þ).
´
9. Crestia, D.; Guerard, C.; Veschambre, H.; Hecquet, L.;
Demuynck, C.; Bolte, J. Tetrahedron: Asymmetry 2001, 12,
869.
10. Lipshutz, B. H.; Harvey, D. F. Synth. Commun. 1982, 12(4),
267.
11. Gefflaut, T.; Lemaire, M.; Valentin, M.-L.; Bolte, J. J. Org.
Chem. 1997, 62, 5920.
12. Bernardski, M. D.; Waldmann, H. J.; Whitesides, G. M.
Tetrahedron Lett. 1986, 27, 5807.
´
4.4.4.2. (1)-(4R,5S,6S)-4,5,6,7-Tetrakis-benzyloxy-2-
13. (a) Kobori, Y.; Myles, D. C.; Whitesides, G. M. J. Org. Chem.
1992, 57, 5899–5907. (b) Lindstad, R. I.; Koll, P.; Mckinley-
Mckee, J. S. Biochem. J. 1998, 330, 479–487.
14. Stevens, R. V.; Beaulieu, N.; Chan, W. H.; Daniesky, T. T.;
methylene-heptanoic acid methyl ester 13. Yield¼73%
(colorless oil). [a]D25¼þ6.2 (c 2.34, CHCl3). Rf¼0.48
(cyclohexane/ethyl acetate, 8:2). 13C NMR (100.61 MHz,

D. Crestia et al. / Tetrahedron 60 (2004) 2417–2425
2425
Waldner, A.; Williard, P. G.; Zutter, U. J. Am. Chem. Soc.
1986, 108, 1039.
16. Hecquet, L.; Demuynck, C.; Bolte, J. J. Mol. Catal. B: Enzym.
2001, 6, 797–802.
15. (a) Chan, T.-H.; Li, C.-J. J. Chem. Soc., Chem. Commun.
1992, 10, 747. (b) Bloor, S. J. Tetrahedron Lett. 1993, 34,
5617. (c) Hon, Y.-S.; Lin, W.-C. Tetrahedron Lett. 1995,
36, 7693. (d) Steger, M.; Young, D. W. Tetrahedron 1999,
55, 7935.
17. Bergmeyer, H. U.; 3rd ed. Methods of enzymatic analysis,
Verlag Chemie: Deerfield, FL, 1984; Vol. 2, p 146.
18. Besse, P.; Baziard-Mouysset, G.; Boubekeur, K.; Palvadeau,
P.; Veschambre, H.; Payard, M.; Mousset, G. Tetrahedron:
Asymmetry 1999, 10, 4745–4754.