In situ proton NMR study of acetyl and formyl group migration in mono-O-acyl D-glucose

Article abstract of DOI:10.1002/mrc.2395

Acetyl and formyl group migration, mutarotation, and hydrolysis of mono-O-acylated glucose are studied by in situ 1D and 2D ¹H NMR spectroscopy. α-D-Glucosyl-1-acetate and α-D-glucosyl-1 -formate serve as sole starting materials. They are gener

Full text of DOI:10.1002/mrc.2395

Research Article
Received: 23 August 2008
Revised: 11 October 2008
Accepted: 1 December 2008
Published online in Wiley Interscience: 26 January 2009
(www.interscience.com) DOI 10.1002/mrc.2395
In situ proton NMR study of acetyl and formyl
group migration in mono-O-acyl D-glucose†
Lothar Brecker,a∗ Marek Mahut,a Alexandra Schwarzb and Bernd Nidetzkyb
Acetyl and formyl group migration, mutarotation, and hydrolysis of mono-O-acylated glucose are studied by in situ 1D and 2D
1H NMR spectroscopy. α-D-Glucosyl-1-acetate and α-D-glucosyl-1-formate serve as sole starting materials. They are generated
in situ by configuration retaining glucosyltransfer from α-D-glucosyl-1-phosphate to formate and acetate, which is catalyzed
by the Glu-237 → Gln mutant of Leuconostoc mesenteroides sucrose phosphorylase. Temporary accumulated regio-isomeric
mono-O-acyl D-glucoses are identified, characterized, and quantified directly from the reaction mixture. Time courses of the
transformations give insight into pH dependence of acyl group migration and mutarotation as well as into the stability of
c
various regioisomers. Copyright ꢀ 2009 John Wiley & Sons, Ltd.
Keywords: NMR; 1H; in situ 1D and 2D NMR; mono-O-acyl D-glucose; acyl group migration; Leuconostoc mesenteroides sucrose
phosphorylase
Introduction
Results and Discussion
Migration of small acyl groups in partially O-acylated carbo-
hydrates or polyhydroxy compounds is well known since the
1920s[1,2] and has occasionally been used in synthetic strategies.[3]
However, it makes the O-acetyl group an unusual protect-
ing group when selective protection of a secondary hydroxyl
group in a saccharide is required.[4] Further on it complicates
structural investigations of partially O-acetylated oligosaccha-
rides and polysaccharides, which naturally occur in plants and
bacteria.[5–7] Migrating acyl groups also impede studies of lipases
[EC 3.1.1.3] and acetylesterases [EC 3.1.1.6] acting on partially acy-
lated saccharides.[8,9] Irreversible acyl group migration (AcM) from
secondary to primary hydroxyl groups is well documented in some
cases,[10,11] butreversibleAcMinpolyhydroxycompoundshasonly
been occasionally studied with respect to kinetic, thermodynamic,
and positional aspects.[12–14] This discrepancy is likely caused by
difficulties to analyze intermediates, which are accumulated in
small amounts for a short time.
During the last decades there is a growing interest to determine
the details of AcM reaction mechanism and its kinetic.[12–19]
NMR spectroscopy has frequently been applied as an excellent
analytical tool for this purpose.[13,14,17–19] Acylated saccharides
used as starting materials in some of these investigations are
isomeric mixtures, which make AcM difficult to analyze.[13] For
some other studies, laboriously synthesized selective O-acylated
glycopyranosides serve as model substrates. These tailor-made
compounds are not subjected to mutarotation and AcM is only
possible over a constricted number of hydroxyl functions.[12,14]
We now use the recently described sucrose phosphorylase
catalyzed glucosyl transfer for selective generation of α-D-
glucosyl-1-acetate (α-1a) and α-D-glucosyl-1-formate (α-2a).[20]
These compounds serve as excellent starting materials to study
several details of acetyl and formyl group migration as well as of
accompanied mutarotation in mono-O-acyl glucose. 1D and 2D
in situ1HNMRspectroscopyisusedasthemostversatiletechnique
to identify and quantify up to 14 intermediates and products in situ
and in parallel during the reasonable short reaction times.
Wild-type sucrose phosphorylase from Leuconostoc mesenteroides
[LmSPase, EC 2.4.1.7] catalyses reversible glucosyl transfer from
sucrose (3) to phosphate with retention of the α-configuration
(Scheme 1(A)). The equilibrium is by far on the side of the
products fructose (4) and α-D-glucosyl-1-phosphate (5).[21] The
corresponding LmSPase Glu-237 → Gln mutant (E237Q) also catal-
yses α-retaining glucosyl transfer from α-D-glucosyl-1-phosphate
(5) to small acceptor molecules like acetate and formate
(Scheme 1(B)).[20] The formed α-D-glucosyl-1-acetate (α-1a) and
α-D-glucosyl-1-formate (α-2a) are subjected to spontaneous AcM,
mutarotation, and partial hydrolysis in aqueous solution, which
has made the use of in situ 1H NMR necessary to identify and
investigate the E237Q catalyzed transformations.[20]
We apply 1D and 2D in situ 1H NMR to gain insight into the
spontaneous consecutive reactions without influences during
work up procedures. Successively recorded 1H NMR spectra allow
identification of known compounds and enable efficient analysis
of the reaction time courses.[22–24] 2D COSY and TOCSY as well
as 1D selTOCSY spectra are recorded directly from the reaction
mixture for additional structural analysis of not yet described
intermediates.[25,26]
An in situ recorded 1H NMR spectrum is given in Fig. 1(A)
and shows indicative signals of all sufficiently accumulated
∗
Correspondenceto:Lothar Brecker,UniversityofVienna,DepartmentofOrganic
¨
Chemistry, Wahringer Straße 38, A-1090 Wien, Austria.
E-mail: lothar.brecker@univie.ac.at
¨
†
Dedicated to the memory of Dr. Marion F. Kogl. Her insights, enthusiasm, and
unique humanity have been a scientific and personal enrichment to many, and
her presence is greatly missed.
¨
a
University of Vienna, Department of Organic Chemistry, Wahringer Straße 38,
A-1090 Wien, Austria
b
Graz University of Technology, Institute of Biotechnology and Biochemical
Engineering, Petersgasse 12, A-8010 Graz, Austria
c
Magn. Reson. Chem. 2009, 47, 328–332
www.soci.org
Copyright ꢀ 2009 John Wiley & Sons, Ltd.

NMR of acyl group migration in mono-O-acyl D-glucose
Scheme 1. (A) Wild-type LmSPase catalyzed formation of sucrose (3) by glucosyl transfer from α-D-glucosyl-1-phosphate (5) to fructose (4). (B) E237Q
catalyzed syntheses of sucrose (3), α-D-glucosyl-1-acetate (α-1a), and α-D-glucosyl-1-formate (α-2a) by glucosyl transfer from α-D-glucosyl-1-phosphate
(5) to fructose (4), acetate, and formate, respectively.
Scheme 2. Reaction scheme of acetyl group migration and mutarotation between D-glucosyl-1-acetate (1a) and the O-acetyl D-glucoses (1b-e).
compounds during α-D-glucosyl-1-formate (α-2a) generation,
Table 1. Proton [δH (ppm)] chemical shifts of O-formyl D-glucoses
formyl group migration, and hydrolysis. The corresponding time
course (Fig. 1(B)) indicates the E237Q catalyzed glycosyl transfer
to formate to be reasonably slow, which is caused by moderate
catalytic efficiency (kcat/Km = 3.7 × 10−4 s−1 mM−1).[20] Several
steps of the formyl group migration, however, are fast and lead
to only ∼400 µM maximum concentrations of 2a. Intermittently
generated 2-O-formyl D-glucose (2b) and 3-O-formyl D-glucose
(2c) are accumulated in detectable amounts only in a time frame
of 6–8 h. Concentrations of both anomers of 4-O-formyl D-glucose
(2d) are below the detection limit, because fast and irreversible
AcM causes direct formation of the two 6-O-formyl-D-glucose
anomers (2e).[10] A simultaneous and rapid formyl ester hydrolysis
of all intermediates and products leads to fast formation of free
glucoseandformate.Thequitelowconcentrationsofaccumulated
O-formyl glucoses (2) allow only to determine a small number of
1H NMR data (Table 1), which are in good accordance with sparse
resultspublishedearlier.[20,27] Furtheranalysisoftheintermediates,
however, is cumbered by fast hydrolysis of the formyl esters.
Using acetate as acceptor E237Q catalyses formation of α-
D-glucosyl-1-acetate (α-1a) with moderate catalytic efficiency
(kcat/Km = 3.9 × 10−4 s−1 mM−1).[20] The consecutive AcM is
strongly pH dependant and small pH changes lead to pronounced
Compound
H-1
H-6a
H-6b
HCOO-
α-D-glucosyl-1-formate (α-2a)
2-O-formyl D-glucose (2b)
5.36
ND
ND
ND
ND
ND
8.21
8.27
8.31a
8.30a
8.15
8.15
3-O-formyl α-D-glucose (α-2c)
3-O-formyl β-D-glucose (β-2c)
6-O-formyl α-D-glucose (α-2e)
6-O-formyl β-D-glucose (β-2e)
5.25
ND
ND
ND
ND
ND
5.18
4.63
4.42
4.47
4.38
4.32
ND, not determined due to signal overlap.
a Exchangeable.
differences in concentrations of the intermittently generated
mono-O-acetyl D-glucoses (1).[28] A transformation at pH 7.8
causes fast AcM only leading to accumulation of the terminally
formed 6-O-acetyl D-glucose (1e). At pH 6.8 the biotransformation
and AcM are very slow resulting in exclusive formation of small
amounts 2-O-acetyl D-glucose (1b) after 10 h. At pH 7.2–7.3 the
enzyme catalyzed transformation, AcM, and mutarotation are
well balanced and allow in situ NMR monitoring of time courses
between12 hand90 h. During thelongerreactiontime, additional
c
Magn. Reson. Chem. 2009, 47, 328–332
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/mrc

L. Brecker et al.
(A)
(A)
(B)
(C)
(B)
(D)
(E)
Figure 2. In situ1HNMRstructuralanalysisof3-O-acetylα-D-glucose(α-1c).
In the 1H NMR (A) H-1 shows a doublet (5.22 ppm; 3JH−H = 4.1 Hz), which
indicates the α-pyranose form. DQF-COSY cutting (B) shows consecutive
couplings to H-2 (3.67 ppm), H-3 (5.12 ppm), and H-4 (3.55 ppm).
Reasonable low field shift of H-3 indicates the hydroxyl group in this
position to be acetylated. The glucose configuration is confirmed by large
coupling constants of H-3 and H-4, which are also detectable in 1H NMR.
Theyindicatenoenzymecatalyzedmodificationofthesaccharideskeleton.
H-1 and H-3 traces in 2D TOCSY with 100 ms mixing time (C) indicate the
shift of H-5(3.87 ppm). 1DselTOCSY with 25 ms mixingtime and irradiation
at H-3 (D) confirms the shifts of H-2, H-4, and H-5. Corresponding 1D
selTOCSY with 100 ms mixing time (E) allows additional determination of
H-6a and H-6b (3.80 ppm and 3.74 ppm). Spectra correspond to the time
course shown in Fig. 4(B) and are recorded between 30 h and 40 h after
the start of the experiment.
Figure 1. (A) 1H NMR spectrum taken during E237Q catalyzed glucosyl
transfer from α-D-glucosyl-1-phosphate (5) to formate and consecutive
formyl group migration, mutarotation, and hydrolysis. Representative
signals of the accumulated intermediates and terminally formed glucose
areindicated(HCOO-:protoninformylgroup).(B) In situ1HNMRmonitored
time course of the reaction cascade at pH 7.2. Shown are concentrations of
the initially formed α-D-glucosyl-1-formate (α-2a) as well as the summed
concentrations of the respective anomers of O-formyl D-glucoses (2b),
(2c), and (2e), and of the terminal formed glucose (Fm: formyl group). The
compounds are assigned by the indicative signals shown in (A) and by the
order of their appearance in the reaction cascade. Starting concentration
of α-D-glucosyl-1-phosphate (5) and of formate are 10 mM and 250 mM,
respectivelyandglucoseconcentrationafter630 minis3.1 mM.4-O-Formyl
D-glucose (2d) does not accumulate in detectable amounts.
acetate groups of all compounds are assigned to the mono-O-
acetyl D-glucoses (1) by parallel time courses of corresponding
H-1 and CH3 group signal integrals (Fig. 3). All detectable 1H NMR
shifts of the accumulated intermediates are listed in Table 2. They
are in good accordance with a few earlier reported shifts of these
compounds.[20,29–33]
in situ 1D selTOCSY as well as 2D COSY, TOCSY, and NOESY spectra
can be recorded, when intermediates are accumulated in higher
concentrations. These measurements enable for the first time a
detailedstructureelucidationofseveralmono-O-acetylD-glucoses
(1) in parallel. A comprehensive analysis is exemplarily shown in
Fig. 2 for identification of 3-O-acetyl α-D-glucose (α-1c).
Apart from the initially generated α-D-glucosyl-1-acetate (α-
1a), both anomers of 2-O-acetyl D-glucose (1b) and 3-O-acetyl
D-glucose (1c) are accumulated in amounts suitable for detailed
characterization. β-D-Glucosyl-1-acetate (β-1a), however, is only
formed in small concentrations by AcM from position two to one
in β-1b. Little signals of the two further acetyl groups indicate
the two anomers of 4-O-acetyl D-glucose (1d) to be accumulated
only in small amounts. Their low concentration is caused by
the fast and irreversible AcM from position four to six in gluco-
configured aldohexoses.[10] Hence, 6-O-Acetyl D-glucose (1e) is
finally generated in larger amounts (Scheme 2). Signals of the
1H NMR signals of the acetyl groups and of some H-1 are
used to determine time courses of the reactions monitored at
pH 7.3 and 7.2 over 13 and 90 h, respectively. This small pH
variation leads to very similar transformation rates of the E237Q
catalyzed reaction, but the AcM rates show reasonable differences
(Fig. 4(A) and (B)). For example, maximum concentration of 2-O-
acetyl D-glucose (1b) is reached after ∼6 and ∼20 h, respectively.
However, several details of the consecutive AcMs are very similar
in both time courses. At the beginning, 2-O-acetyl α-D-glucose
(α-1b) is formed in excess above the corresponding β-anomer
β-1b before mutarotation inverted their proportion towards the
thermodynamically favored ratio after a couple of minutes. The
correspondingequilibrationof3-O-acetylD-glucose(1c)isreached
considerably later, which is very likely caused by faster AcM from
position two to three in the α-anomer compared to the β-anomer
as well as by slower mutarotion of 1c compared to those of
c
www.interscience.wiley.com/journal/mrc
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2009, 47, 328–332

NMR of acyl group migration in mono-O-acyl D-glucose
identification of accumulated intermediates and gives insight into
time courses of acyl group migration as well as into accompanied
mutarotation. Such investigations complement AcM studies using
partly protected model substrates, which allow monitoring of only
a restricted number of migration steps.[12,14] The resulting data are
valuable for future identification of partially O-acetylated natural
saccharides as well as for analysis of biocatalyzed regiospecific
acylation/deacylation of carbohydrates.
Experimental
Chemicals
All chemicals are purchased from Sigma-Aldrich Chemical Co., St
Louis, USA in the highest available purity and used without further
purification.
Figure 3. 1H NMR signals of the acetate groups assigned to the anomers of
D-glucosyl-1-acetate (1a) and of the O-acetyl D-glucoses (1b-e). Spectrum
matches the last in situ 1H NMR measurement of the time course shown in
Fig. 4(B).
Site-directed mutagenesis, protein production, and purifica-
tion
1b (Fig. 4(A)). The anomeric ratios of intermediate 1d and of
terminal product 1e do not change during the reaction, as they
are formed after the equilibrium conditions of the mutarotation
arereached.Progressofthetotalproductformationshowsthatthe
E237Q catalyzed transformation decelerated, caused by product
inhibition and enzyme denaturation (Fig. 4(C)). Hydrolysis of the
acetates is quite small as by far no free glucose is monitored. The
pH hence drops only very slightly during the reaction (−0.03 pH)
in spite of high buffer concentration. However, even this small
decrease in pH leads to the slowing of AcM. Such changes during
transformations impede a sensible determination of the large
number of pH dependant rate constants (Scheme 2), which make
the reported system more complex compared to recently studied
AcM in partly protected α-D-galactose.[14]
Site-directed mutagenesis, protein production, and purification of
E237Q have been described previously.[20]
NMR sample preparation
Forin situNMRmeasurementsthereactionsareperformeddirectly
in a NMR sample tube. Samples contain 10 mM α-D-glucosyl-
1-phosphate (5), 9 µM E237Q, and 250 mM sodium formate or
sodium acetate, respectively, in D2O (0.70 mL, 99.9% D). The pH
value is adjusted by addition of 1.0 M NaOD or DCl solutions in
D2O and measured at the start and at the end of the reaction.
Small amounts of MES buffer (1 mM), which originate from protein
purification, serve for in situ estimation of pH changes during the
reaction.
In situ 1H NMR measurements
Conclusions
LmSPase E237Q mutant catalyzed reactions provide the specific
synthesis of α-D-glucosyl-1-acetate (α-1a) and α-D-glucosyl-1-
formate (α-2a) which are excellent starting materials for a
first-time study of free formyl and acetyl group migration in
unprotected mono-O-acyl D-glucose (1). 1D and 2D in situ 1H
NMR monitoring is an optimal analytical technique for parallel
Reactions taking 13–14 h are performed in the magnet and up
to 64 1H NMR spectra are directly taken from the sample at
regular intervals.[22] The reactions over 90 h are performed in the
magnet and temporarily kept in a temperature-controlled water
bath (303 0.2 K). 1D and 2D H NMR measurements of these
reactions are measured in non-uniform intervals.[25,26]
1
Table 2. Proton [δH (ppm)] chemical shifts of O-acetyl D-glucoses
Compound
H-1
H-2
H-3
H-4
H-5
H-6a
3.75
H-6b
3.43
CH3CO-
α-D-glucosyl-1-acetate (α-1a)
β-D-glucosyl-1-acetate (β-1a)
2-O-acetyl α-D-glucose (α-1b)
2-O-acetyl β-D-glucose (β-1b)
3-O-acetyl α-D-glucose (α-1c)
3-O-acetyl β-D-glucose (β-1c)
4-O-acetyl α-D-glucose (α-1d)
4-O-acetyl β-D-glucose (β-1d)
6-O-acetyl α-D-glucose (α-1e)
6-O-acetyl β-D-glucose (β-1e)
6.05
5.49
5.32
4.76
5.22
4.69
3.74
3.80
3.75
3.75
2.141
2.170
2.118
2.113
2.124
2.130
2.104a
2.107a
2.070
2.074
ND
4.66
4.61
3.67
3.37
3.85
3.62
5.12
4.93
3.46
3.45
3.55
3.55
ND
3.81
3.80
3.87
3.51
3.91
3.84
3.80
3.88
3.82
3.67
3.74
3.69
ND
5.18
4.62
3.69
3.23
3.50
3.46
3.42
3.64
3.99
4.37
4.32
4.37
4.25
4.21
ND, not determined due to low concentration.
a Exchangeable.
c
Magn. Reson. Chem. 2009, 47, 328–332
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/mrc

L. Brecker et al.
(B)
(A)
(C)
Figure 4. In situ 1H NMR time course of α-D-glucosyl-1-acetate (α-1a) formation, AcM, and mutarotation at pH 7.3 (A) and 7.2 (B) over 13 h and 90 h,
respectively. [Symbols: close circle (β-1a); open circle (α-1a); close square (β-1b); open square (α-1b); close triangle (β-1c); open triangle (α-1c); star
(β-1d); cross (α-1d); close diamond (β-1e); open diamond (α-1e).] Concentration of summed products in the reaction performed at pH 7.2 over 90 h
is shown in (C). Starting concentrations of α-D-glucosyl-1-phosphate (5) and of acetate are 10 mM and 250 mM, respectively. Signals of H-1 and of the
acetate groups are used to obtain the progress curves of the intermediates and products.
NMR spectroscopic measurements
[6] A. Teleman, M. Nordstrom, M. Tenkanen, A. Jacobs, O. Dahlman,
Carbohydr. Res. 2003, 338, 525.
All spectra are recorded on a DRX-600 AVANCE spectrometer
(Bruker, Rheinstetten, Germany), equipped with a triple resonance
xyz-gradient inverse probe and processed using the Topspin 1.3
software.1Hirradiationandmeasurementfrequencyis600.13 MHz
and the sample temperature is adjusted to 303.15 0.01 K. The
1D 1H and selTOCSY spectra are recorded with a 30◦ 1H-pulse,
acquisition of 32 768 data points, digital FID resolution of 0.18 Hz,
a relaxation delay of 1.0 s, and accumulation of 16 to 512 scans.[26]
[7] H. A. Schols, A. G. J. Voragen, in HandbookofFoodEnzymology, (Eds:
J. R. Whitaker, A. G. J. Voragen, D. W. S. Wong), Marcel Dekker: New
York and Basel, 2003, pp 829.
[8] H. G. Park, J. H. Do, H. N. Chang, Biotechnol. Bioprocess Eng. 2003, 8,
1.
[9] B. La Ferla, Monatsh. Chem. 2002, 133, 351.
[10] K.-F. Hsiao, H.-J. Lin, D.-L. Leu, S.-H. Wu, K.-T. Wang, Bioorg. Med.
Chem. Lett. 1994, 4, 1629.
[11] K. Kefurt, Z. Kefurtova, J. Jary, I. Horakova, M. Marek, Carbohydr. Res.
1992, 223, 137.
1
The overwhelming HDO signal in H NMR spectra is suppressed
[12] M. Mastihubova, P. Biely, Carbohydr. Res. 2004, 339, 1353.
[13] V. Molinier, K. Wisniewski, A. Bouchu, J. Fitremann, Y. Queneau, J.
Carbohydr. Chem. 2003, 22, 657.
by presaturation. For selective excitation in selTOCSY Gaussian
pulses with 50 ms duration are used and the mixing times varied
between 25 and 100 ms. After zero filling to 65 536 data points
the free induction decays are Fourier transformed to spectra of
6000 Hz range. 2D TOCSY and NOESY spectra are recorded with
1024 data points and 8 scans in the t2-dimension, as well as with
256 experiments in t1-dimension leading to measurement times
between 75 and 100 min and to digital FID resolutions of 4.7 Hz
(t2-dimension) and 18.7 Hz (t1-dimension). Zero filling to 512 data
points in t1-dimension, appropriate sinusoidal multiplication, and
Fourier transformation lead to spectra with 6000 Hz range in both
dimensions. Mixing time in 2D NOESY spectra is 800 ms. All spectra
are referenced to external acetone (δ 1H: 2.225 ppm).
[14] M. U. Roslund, O. Aitio, J. Warnaa, H. Maaheimo, D. Y. Murzin,
R. Leino, J. Am. Chem. Soc. 2008, 130, 8769.
[15] M. A. Rangelov, G. N. Vayssilov, D. D. Petkov, Int. J. Quantum Chem.
2006, 106, 1346.
[16] E. A. Pritchina, N. P. Gritsan, G. T. Burdzinski, M. S. Platz, J. Phys.
Chem. A 2007, 111, 10483.
[17] D. Horton, W. E. Mast, K. D. Philips, J. Org. Chem. 1967, 32, 1471.
[18] M. R. Vignon, J. A. Vottero, Tetrahedron Lett. 1976, 28, 2445.
[19] J. M. Brown, P. A. Chaloner, A. Colens, J. Chem. Soc., Perkin Trans. 2
1979, 71.
[20] A. Schwarz, L. Brecker, B. Nidetzky, Biochem. J. 2007, 403, 441.
[21] J. J. Mieyal, M. Simon, R. H. Abeles, J. Biol. Chem. 1972, 247, 532.
[22] L. Brecker, D. W. Ribbons, Trends Biotechnol. 2000, 18, 197.
[23] J.-P. Grivet, A.-M. Delort, J.-C. Portais, Biochimie 2003, 85, 823.
[24] P. Sedmera, P. Halada, C. Peterbauer, J. Volc, TetrahedronLett. 2004,
45, 8677.
Acknowledgment
[25] P. Sedmera, P. Halada, E. Kubatova, D. Haltrich, V. Prikrylova, J. Volc,
J. Mol. Catal. B: Enzym. 2006, 41, 32.
We thank Susanne Felsinger (University of Vienna) for recording
several NMR spectra.
[26] C. Tyl, S. Felsinger, L. Brecker, J. Mol. Catal. B: Enzymol. 2004, 28, 55.
[27] J. M. Indelicato, B. A. Stewart, G. L. Engel, J. Pharm. Sci. 1980, 69,
1183.
[28] pH values were calculated according to: pH
= pD + 0.4.
(a) P. R. Mussini, T. Mussini, S. Rondinini, Pure Appl. Chem. 1997,
69, 1007; (b) D. Luo, Huaxue Shijie 1985, 26, 24.
References
[29] S. Hanessian, V. Mascitti, P. Lu, H. Ishida, Synthesis 2002, 14, 1959.
[30] M. Bols, H. C. Hansen, Acta Chem. Scand. 1993, 47, 818.
[31] M. Abbadi, C. Morin, Bioorg. Med. Chem. Lett. 1999, 9, 1779.
[32] K. M. Taba, R. Ko¨ster, W. V. Dahlhoff, Synthesis 1983, 1036.
[33] K. Nomura, K. Sugimoto, H. Nishiura, K. Ohdan, T. Nishimura,
H. Hayashi, T. Kuriki, Biosci. Biotechnol. Biochem. 2008, 72, 82.
[1] E. Fischer, Ber. Dtsch. Chem. Ges. 1920, 53, 1621.
[2] B. Helferich, W. Klein, Liebigs Ann. Chem. 1927, 455, 173.
[3] R. Schauer, Meth. Enzymol. 1987, 138, 611.
[4] T. W. Greene, P. G. M. Wuts, Protective Groups in Organic Synthesis
(3rd Edn), John Wiley & Sons, Inc.: New York, 1999, pp 149.
[5] M. A. Kabel, P. de Waard, H. A. Schols, A. G. J. Voragen, Carbohydr.
Res. 2003, 338, 69.
c
www.interscience.wiley.com/journal/mrc
Copyright ꢀ 2009 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2009, 47, 328–332