Synthesis and conformational analysis of carbasugar bioisosteres of α-l-iduronic acid and its methyl glycoside

Article abstract of DOI:10.1016/j.carres.2010.03.002

The synthesis of two novel carbasugar analogues of α-l-iduronic acid is described in which the ring-oxygen is replaced by a methylene group. In analogy with the conformational equilibrium described for α-l-IdopA, the conformation of the carbasugars was investigated by ¹H and ¹³C NMR spectroscopy. Hadamard transform NMR experiments were utilised for rapid acquisition of ¹H,¹³C-HSQC spectra and efficient measurements of heteronuclear long-range coupling constants. Analysis of ¹H NMR chemical shifts and J_H,H coupling constants extracted by a total-lineshape fitting procedure in conjunction with J_H,C coupling constants obtained by three different 2D NMR experiments, viz., ¹H,¹³C-HSQC-HECADE, J-HMBC and IPAP-HSQC-TOCSY-HT, as well as effective proton-proton distances from 1D ¹H,¹H T-ROE and NOE experiments showed that the conformational equilibrium ⁴C₁?²S_5a?¹C₄ is shifted towards ⁴C₁ as the predominant or exclusive conformation. These carbasugar bioisosteres of α-l-iduronic acid do not as monomers show the inherent flexibility that is anticipated to be necessary for biological activity.

Full text of DOI:10.1016/j.carres.2010.03.002

Carbohydrate Research 345 (2010) 984–993
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Synthesis and conformational analysis of carbasugar bioisosteres
of a-L-iduronic acid and its methyl glycoside
*
Elin Säwén, Mattias U. Roslund, Ian Cumpstey, Göran Widmalm
Department of Organic Chemistry, Arrhenius Laboratory, Stockholm University, S-106 91 Stockholm, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis of two novel carbasugar analogues of
gen is replaced by a methylene group. In analogy with the conformational equilibrium described for
a-L-iduronic acid is described in which the ring-oxy-
Received 18 December 2009
Received in revised form 15 February 2010
Accepted 1 March 2010
a-L-
IdopA, the conformation of the carbasugars was investigated by 1H and 13C NMR spectroscopy. Hadamard
transform NMR experiments were utilised for rapid acquisition of 1H,13C-HSQC spectra and efficient mea-
surements of heteronuclear long-range coupling constants. Analysis of 1H NMR chemical shifts and JH,H
coupling constants extracted by a total-lineshape fitting procedure in conjunction with JH,C coupling con-
stants obtained by three different 2D NMR experiments, viz., 1H,13C-HSQC-HECADE, J-HMBC and IPAP-
HSQC-TOCSY-HT, as well as effective proton–proton distances from 1D 1H,1H T-ROE and NOE experi-
ments showed that the conformational equilibrium 4C1 ꢀ 2S5a ꢀ 1C4 is shifted towards 4C1 as the predom-
Available online 6 March 2010
Presented at the SMASH 2008 NMR
Conference, Santa Fe, NM, USA, September
7–10, 2008
inant or exclusive conformation. These carbasugar bioisosteres of
show the inherent flexibility that is anticipated to be necessary for biological activity.
Ó 2010 Elsevier Ltd. All rights reserved.
a-L-iduronic acid do not as monomers
Keywords:
Heparin
Conformational equilibrium
NMR
Molecular modelling
1. Introduction
OH
OH
Hpro-S
OH
Monosaccharides are usually assumed to have a rigid ring-con-
formation. This is a good approximation for hexapyranoses with up
to one non-anomeric axial hydroxy group, which includes the
majority of monosaccharide residues found in oligosaccharides of
biological or biochemical importance. Iduronic acid, which is found
HOOC
HOOC
1C4
O
HOOC
6
5
1
OH
OH
3
HO
OH
a-(1?4)-linked in nature in the glycosaminoglycans heparin, hep-
OH
OH
OH
aran sulfate and dermatan sulfate, is unusual among biologically
relevant hexoses in that in its pyranose form it does not exclusively
occupy the 4C1 conformation, but rather is a flexible entity with
more than one low-energy conformation accessible.1 The relative
population of the three low-energy conformations 4C1, 2SO and
1C4 (Fig. 1) is dependent on the surrounding residues and sulfa-
tion.2,3 It has been proposed that the flexibility of iduronic acid is
key to the strong binding of certain of these glycans to their recep-
tors.1 One important interaction is the binding of a heparin penta-
2SO
HO
HO
1
O
Hpro-R
HOOC
HO
OMe
OH
5a
4
O
HO
HO
4C1
2
OH
OH
COOH
OH
antithrombin.4
Different
conformationally
2
saccharide
to
constrained mimics of iduronic acid have been synthesised and
incorporated into heparin sequences whose affinities for anti-
thrombin indicate that the 2SO skew conformation is important
for this binding event.5–7 The conformational preferences of idu-
ronic acid and heparin-derived oligosaccharides have been studied
by NMR spectroscopy, molecular modelling, molecular dynamics
Figure 1. Schematic of possible conformations for
and the carbasugar analogues thereof synthesised herein (right column).
a-L-iduronic acid (left column)
simulations, ab initio quantum mechanical computations and den-
sity functional theory methodology.8–16
Carbasugars (formerly pseudosugars) are carbohydrate-like
molecules in which the ring-oxygen has been formally replaced
by carbon.17–19 They thus differ from natural glycosides in that
pseudodisaccharides based on carbasugars will be hydrolytically
* Corresponding author.
E-mail address: gw@organ.su.se (G. Widmalm).
0008-6215/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2010.03.002

E. Säwén et al. / Carbohydrate Research 345 (2010) 984–993
985
stable, an important factor when considering enzyme inhibitor de-
OBn
OBn
sign. Some ido-configured carbasugars have been synthesised be-
fore: both anomers of carbaidose have been described, as have
some C1 ether pseudodisaccharides.20–25 Only two carbaiduronic
acid derivatives have been synthesised, both as protected deriva-
OH
OH
BnO
OBn
tives only, and both with the non-natural b-D
-configuration.26,27
BnO
OBn
OBn
3
Recently, two novel carbasugars were isolated from Streptomyces
OBn
5
lincolnesis.28
We were interested as to whether carbaiduronic acid deriva-
tives would have a conformational profile that would give them
the potential to be useful mimics of iduronic acid, and we therefore
undertook the synthesis of carbasugar 1 and its 1-O-methyl ether
derivative 2 (Fig. 1) and investigated their ring-conformation(s),
the results of which we report in this paper.
(i)
OBn
OBn
OH
OMe
(v)
BnO
OBn
BnO
OBn
2. Results and discussion
OBn
OBn
8
4
2.1. Synthesis
(ii)
(vi)
Our synthesis started from the known cyclohexene derivative 3
that is prepared from L
-sorbose in nine steps (Scheme 1).29 Hydro-
OH
OH
genation of the C@C double bond proceeded stereoselectively with
addition of hydrogen from the same face as the free OH group to
give the protected carbaidose 4 as the major product, separable
by chromatography from the minor gluco compound 5 (ido:gluco;
4:5; 4:1).30 Selective cleavage of the primary benzyl ether from
O6 in 4 was achieved by acetolysis with TFA and Ac2O,31 followed
by deacetylation with NaOMe in MeOH to give the known diol 6.30
Initial attempts at selective oxidation to iduronic acid 7 failed how-
ever: treatment of diol 6 with TEMPO and diacetoxyiodobenzene
under two-phase conditions with CH2Cl2 and water gave clean for-
mation of the C6 aldehyde,32 which was not transformed into the
acid 7 even after several hours. Using a mixture of acetone and
water as reaction solvent in the oxidation did not change the out-
come. Nevertheless, the crude aldehyde could be oxidised to the
carboxylic acid 7 with sodium chlorite.33 The secondary alcohol
at C1 remained untouched during these two steps. Hydrogenolysis
of the benzyl ethers gave the deprotected carbasugar 1.
OH
OMe
BnO
OBn
BnO
OBn
OBn
OBn
9
6
(vii)
(iii)
OH
OH
OH
OMe
O
O
BnO
OBn
BnO
OBn
OBn
7
OBn
10
In order to access the carbasugar analogue 2 of the methyl gly-
coside, OH1 of carbaidose 4 was methylated with methyl iodide to
give the methyl ether 8. The OH6 protection was selectively re-
moved as described above to give the primary alcohol 9. In this
case, treatment with TEMPO and diacetoxybenzene in acetone/
water gave the carboxylic acid 10 directly, although a rather long
reaction time (48 h) was needed for complete conversion. Deben-
zylation as described above gave the methyl glycoside analogue 2.
(iv)
(viii)
OH
OH
OH
OMe
O
O
HO
OH
HO
OH
OH
OH
1
2
2.2. Conformational analysis
Scheme 1. Reagents and conditions: (i) H2, Pd (C), Et3N, EtOAc; 4, 78%; 5, 19%; (ii)
TFA, Ac2O; then NaOMe, MeOH, 84%; (iii) TEMPO, PhI(OAc)2, CH2Cl2, water; then
DMSO, tBuOH, dimethoxybenzene, NaClO2, NaH2PO4, 67%; (iv) Pd(OAc)2, H2, EtOH,
AcOH, 96%; (v) NaH, MeI, THF, 92%; (vi) TFA, Ac2O; then NaOMe, MeOH, 65%; (vii)
TEMPO, PhI(OAc)2, acetone, water, 64%; (viii) Pd(OAc)2, H2, MeOH, AcOH, 89%.
The conformational analysis of carbasugar 1 and its 1-O-methyl
ether derivative 2 (Fig. 1) is based on NMR spectroscopy and
molecular modelling. 1H and 13C NMR chemical shift assignments
of the novel compounds used, besides standard 1D 1H and 13C
experiments, 2D 1H,1H-TOCSY, 1H,13C-H2BC,34 1H,13C-HMBC, and
Hadamard transform (HT) 1H,13C-HSQC experiments.35 The
1H,13C-HSQC-HT experiment was used to rapidly and efficiently
obtain the one-bond 1H,13C-correlations. Since the 13C resonances
from both 1 and 2 were resolved in the 1D 13C NMR spectra and
the compounds have, respectively, six and seven carbons that carry
protons, a Hadamard-8 matrix was optimal for irradiation. In par-
ticular, as the ‘all-plus’ column of the Hadamard matrix is normally
not used since it is not effective in reducing instrumental arte-
facts.36 The 1H,13C-HSQC-HT spectra of 1 are presented in Figure 2
and the 1H and 13C NMR chemical shift assignments of the two
compounds are compiled in Table 1. Analysis of the homo- and
heteronuclear coupling constants (vide infra) showed that the con-
formational equilibria in the two compounds are very similar, and
therefore the detailed conformational analysis is focussed on com-
pound 1.
The conformational analysis is based on the following: 3JH,H val-
3
ues, JH,C values, effective 1H,1H-distances derived from cross-
relaxation rates in 1D T-ROESY and NOESY experiments and 1H
chemical shifts. The JH,H coupling constants were extracted by a to-
tal-lineshape analysis37 carried out with the PERCH NMR spin sim-
ulation software, which resulted in an excellent agreement
between the experimental and the simulated 1H NMR spectra

986
E. Säwén et al. / Carbohydrate Research 345 (2010) 984–993
one may in the HSQC-HECADE or the J-HMBC experiments alter a
30
40
50
60
70
80
scaling factor for the F1-peak separation to resolve the overlap.
The IPAP-HSQC-TOCSY-HT experiment requires significantly less
experimental time than the other two experiments, which have
C5a
C5
n
conventional t1-evolution times. The JH,C coupling constants,
which were quantified and were consistent between the experi-
mental spectra within 0.5 Hz of the average, are compiled in Ta-
ble 2. Two-bond coupling constants are negative whereas three-
bond coupling constants are positive. In the same way as for the
3JH,H coupling constants, the 3JH,C coupling constants were analysed
for the three ring-conformations using the Karplus-type relation-
ship 3JH,C = 4.26 ꢀ 1.00cosh + 3.56cos2h as proposed by Wasylishen
and Schaefer.47 The results for the three conformations are plotted
in Figure 4b, and again the 4C1 conformation shows the best agree-
ment between experimental and calculated values from the en-
ergy-minimised structures.
C1
C4
C3
C2
3.5 3.0
2.5
2.0 1.5
1H /ppm
Effective proton–proton distances may be obtained from 1D
1H,1H T-ROESY and 1H,1H NOESY experiments48,49 in which proton
resonances are selectively targeted and build-up curves are created
as a function of the mixing times used (Fig. 6). The cross-relaxation
Figure 2. 1H,13C-HSQC-Hadamard transform spectra of 1 (1H-decoupled: black
cross-peaks; 13C-coupled: red cross-peaks).
rates (
r
) are subsequently obtained from the initial slopes and the
1=6, com-
3
(Fig. 3a and b). The JH,H values thus obtained (Table 1) could be
effective distances (rij) are derived from rij ¼ rref
ð
rref rijÞ
=
used in the conformational analysis. Based on previous conforma-
tional analyses of a-L-IdopA and its derivatives, it is reasonable to
piled in Table 3. The agreement between the effective proton–pro-
ton distances from T-ROE and NOE is excellent, and a comparison
between experimentally derived distances and those obtained
from the molecular models confirms that the 4C1 conformation is
the predominant one (Table 3 and Fig. 4c). In this analysis, the
effective distances between H5 and H5apro-S and between H5apro-R
and H2 differ substantially between the three conformations,
whereas the distance between H5apro-S and H2 was approximately
3.83 Å in all three conformations and was not used to investigate
conformational preferences.
consider the following three conformations in 1: 4C1, S5a and
1C4. For a conformational equilibrium between the three ring-con-
formations, the NMR observables will be averaged by rapid ex-
change. An indication of the predominant conformation may be
obtained by predicting the 1H NMR chemical shifts and proton–
proton coupling constants for each conformation. This was carried
out using the PERCH NMR simulation software for which the 4C1,
2S5a and 1C4 energy-minimised conformations of 1 were given as
input. Comparison of the predicted 1H NMR spectra with the
experimental 1H NMR spectrum (Fig. 3c–e vs a) shows that the
4C1 conformation agrees the best. This result indicates that this
conformation should be present for 1. A comparison between
2
Based on 1H chemical shifts and homo- and heteronuclear cou-
pling constants together with effective proton–proton distances,
the equilibrium in 1 (and 2) is shifted far towards the 4C1 confor-
mation (Fig. 7), which is the predominant or exclusive conforma-
3
tion with only small if any populations of S5a and 1C4, since
2
experimental and predicted JH,H values, using the Karplus-type
relationships proposed by Haasnot et al.38 and implemented in
the Janocchio software,39 confirms this conclusion (Fig. 4a). All
the JH,H values of compound 2 (Table 1) are very similar to those
of 1. Consequently, whatever conformation or conformational
equilibrium is present in 1 is also present in 2.
increased populations of these conformations would not lead to
significantly better agreement with experimental data (cf. Fig. 4).
The exchange of the ring-oxygen in
a-L-IdopA for a methylene
group results in the absence of an anomeric effect that would sta-
bilise the conformation with an axial C1ꢀO1 bond, i.e., the 1C4 con-
former. Thus, the driving force to populate the 1C4 conformation in
1H,13C-Heteronuclear coupling constants give, like JH,H, infor-
n
4C1 ꢀ 2S5a ꢀ 1C4 equilibrium is not present to any significant ex-
mation on conformational preferences, in particular, of torsion an-
gles via their Karplus-type relationships. Several approaches40 to
extract these J-couplings have been reported, e.g., from 1D NMR
spectra.41,42 In the present study we employ three different 2D
NMR experiments to measure nJH,C, viz., (i) 1H,13C-HSQC-HECADE,43
(ii) J-HMBC44 and (iii) IPAP-HSQC-TOCSY-HT.35,45,46 NMR spectra
resulting from the three experiments are presented in Fig. 5a–c,
a
tent in these carbasugars. The importance of the anomeric effect in
general and the exo-anomeric effect in particular has also been
noted for carba- and C-fucopyranosides50,51 where, in its absence,
conformations that would not be populated to any significant ex-
tent then are present. The fact that internal
fo- -IdopA in oligosaccharide sequences related to heparin only
populate the 1C4 and 2SO conformations whereas non-sulfated
-IdopA as a terminal group populates also the 4C1 conformation3
a-L-IdopA and 2-O-sul-
a
-L
n
respectively. For all spectra, JH,C values are obtained from a
a-
peak-to-peak separation. The signs of the coupling constants may
be obtained from the relative peak separation in the HSQC-HECADE
and IPAP-HSQC-TOCSY-HT experiments. If spectral overlap occurs,
L
indicates that the carbaiduronic acid bioisosteres may still be suit-
able as hydrolytically stable residues in synthetic oligosaccharides.
Table 1
1H and 13C NMR chemical shifts and 1H,1H coupling constants of compounds 1 and 2 at 27 °C in D2O
1
2
3
4
5
5a
6
Me
3JH1,H5a(pro-R)/H5a(pro-S)
3JH2,H1/H3
3JH3,H4
3JH4,H5
3JH5,H5a(pro-R)/H5a(pro-S)
2JH5a(pro-R)/H5a(pro-S)
1
2
1H
d ppm
nJH,H/Hz
d ppm
3.56
11.94, 4.67
70.5
3.20
9.34, 9.30
78.0
3.60
9.78
75.2
3.53
5.57
73.6
2.81
4.95, 2.97
44.1
1.49 (pro-R)
ꢀ13.54
32.0
2.29 (pro-S)
2.50 (pro-S)
13C
1H
181.2
181.0
d/ppm
nJH,H/Hz
d ppm
3.267
11.43, 4.36
80.3
3.274
9.57, 9.33
76.7
3.61
9.71
75.4
3.53
5.54
73.5
2.84
4.72, 3.10
43.8
1.34 (pro-R)
ꢀ13.44
28.4
3.44
57.5
13C

E. Säwén et al. / Carbohydrate Research 345 (2010) 984–993
987
e
d
c
b
a
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1H /ppm
Figure 3. 1H NMR analysis at 700 MHz of 1: (a) experimental spectrum (black), (b) simulated spectrum by total-lineshape analysis using the PERCH NMR software to obtain
1H,1H coupling constants (blue), (c) predicted by PERCH NMR software for the 4C1 conformation (green), (d) for the 2S5a conformation (purple) and (e) for the 1C4 conformation
(red).
3
3
chemical shifts, JH,H, and JH,C obtained by three different 2D
NMR experiments as well as proton–proton distances from two
different types of nuclear Overhauser experiments. The carbasugar
bioisosteres do not exhibit the conformational flexibility that has
3. Conclusions
The conformational preferences of the two novel carbaiduronic
acid derivatives 1 and 2 of
a-L
-IdopA were studied by 1H NMR

988
E. Säwén et al. / Carbohydrate Research 345 (2010) 984–993
43.6
a
b
c
a
43.8
44.0
44.2
44.4
44.6
10
5
A
2.30
1H /ppm
2.28
5
10
Experimental 3JHH /Hz
b
70.0
10
5
B
C
70.5
71.0
3.2
3.0
2.8
1H /ppm
5
10
Experimental 3JCH /Hz
c
D
E
4
3
2
1900
1800
1700
1H /Hz
2
3
4
Figure 5. Two- and three-bond heteronuclear correlations in 1: (a) 1H,13C-HSQC-
HECADE spectrum (mixing time 40 ms) showing the H5apro-S to C5 correlation (A);
j = 20.6 showing the H2 to C1 (B) and the H5 to C1 (C)
correlations (c) IPAP-HSQC-TOCSY-HT (mixing time 50 ms) showing the H2 to C5a
(D) and the H5 to C5a (E) correlations.
Experimental rHH /Å
(b) J-HMBC spectrum with
Figure 4. Experimental versus calculated data of 1 for the three conformations 4C1
(green diamonds), S5a (purple squares) and 1C4 (red triangles): (a) JH,H; (b) JC,H; (c)
2
1H,1H distance.
carbasugar, as has been seen for iduronic acid itself, would be
the most obvious extension of this work. It is hoped that oligosac-
charides having these chemically modified and hydrolytically sta-
ble residues may be of importance as therapeutic agents of
medical significance related to cancer processes.52
been reported for a-L-iduronic acid in oligo- or polysaccharides and
which is of great importance for its interaction with some protein
receptors, such as antithrombin. The present study shows that the
conformational equilibrium in the six-membered ring may be
investigated in detail using different NMR techniques. To be able
to use these carbasugars in (pseudo)oligosaccharides that mimic
the biological activity of heparin, the influence by a 2-O-sulfo
group or by sugar residues at O1 and O4 of the carbasugar on its
conformation should be investigated. Substitution with a 2-O-sulfo
group and incorporation of the carbasugar into a heparin-like oli-
gosaccharide that may affect the conformational properties of the
4. Experimental section
4.1. General methods
Proton nuclear magnetic resonance (1H) spectra (4ꢀ10) were
recorded on Bruker Avance II 400 (400 MHz) and Varian Mercury

E. Säwén et al. / Carbohydrate Research 345 (2010) 984–993
989
Table 2
Two- and three-bond heteronuclear coupling constants extracted from 1H,13C-HSQC-HECADE, J-HMBC and IPAP-HSQC-TOCSY-HT experiments
Carbon
Proton
Compound 1
J-HMBC
Compound 2
J-HMBC
HECADE
IPAP-TOCSY
HECADE
IPAP-TOCSY
J/Hz
|J|/Hz
J/Hz
J/Hz
|J|/Hz
J/Hz
C1
H2
H3
ꢀ5.1
1.0
4.7
1.1
ꢀ5.1
1.3
5.4
1.8
1.5
2.3
H5
10.0
ꢀ6.0
ꢀ5.7
10.2
6.2
5.7
10.6
ꢀ6.0
ꢀ5.0
>0
10.1
6.3
5.5
H5apro-R
H5apro-S
OMe
ꢀ6.2
ꢀ5.4
4.6
C2
H1
H3
H4
4.1
4.6
1.2
5.2
4.2
ꢀ5.1
ꢀ4.5
ꢀ4.1
1.1
H5apro-R
H5apro-S
3.4
10.4
3.0
10.7
2.8
10.6
2.9
10.7
C3
C4
H1
H2
H4
H5
1.3
4.8
6.2
8.6
ꢀ5.0
7.8
9.2
8.4
H2
H3
H5
H5apro-R
H5apro-S
1.2
ꢀ5.8
ꢀ6.7
1.5
1.3
5.8
6.4
1.5
11.0
1.4
ꢀ5.4
7.2
6.5
1.8
ꢀ6.6
2.6
10.5
ꢀ6.1
2.1
10.8
10.7
C5
H1
H3
H4
H5apro-R
H5apro-S
1.2
1.1
2.7
3.4
4.3
1.6
ꢀ2.8
ꢀ3.4
ꢀ4.2
0.7
ꢀ2.6
ꢀ3.5
ꢀ4.6
1.1
2.1
ꢀ3.6
ꢀ3.7
ꢀ5.1
ꢀ2.4
ꢀ3.6
ꢀ4.4
3.6
4.2
C5a
C6
H1
H2
H4
H5
ꢀ2.0
1.8
1.1
1.2
5.0
ꢀ2.2
1.2
1.2
>0
ꢀ4.5
ꢀ5.1
ꢀ5.5
4.9
ꢀ4.6
H4
H5
H5apro-R
H5apro-S
9.3
6.4
6.6
9.3
2.2
400 (400 MHz) spectrometers; multiplicities are quoted as singlet
(s), broad singlet (br s), doublet (d), doublet of doublets (dd), etc.
or multiplet (m). Carbon nuclear magnetic resonance (13C) spectra
(4ꢀ10) were recorded on Bruker Avance II 400 (100 MHz) and Var-
ian Mercury 400 (100 MHz) spectrometers. Spectra were assigned
using COSY, HSQC and DEPT experiments. Chemical shifts and cou-
pling constants were recorded in units of ppm and Hz. Residual sol-
vent signals were used as an internal reference for CDCl3:
dH = 7.26 ppm and dC = 77.16 ppm for compounds 4ꢀ10. In order
to extract accurate coupling constant values, computational spec-
tral analyses of all the 1H NMR spectra were performed with the
PERCH NMR software. For molecules with several benzyl groups,
the substitution positions have not been determined but they are
distinguished by different number of primes. Low- and high-reso-
lution (HRMS) electrospray (ES) mass spectra were recorded using
a Bruker Microtof instrument. Infra-red spectra were recorded on a
Perkin–Elmer Spectrum One FT-IR spectrometer using the thin film
method on NaCl plates. Optical rotations were measured on a Per-
kin–Elmer 241 polarimeter with a path length of 1 dm; concentra-
tions are given in g/100 mL. Thin layer chromatography (TLC) was
carried out on Merck Kieselgel sheets, pre-coated with 60F254 sil-
ica. The plates were visualised with UV light or developed using
10% sulfuric acid or an ammonium molybdate (10% w/v) and cer-
ium (IV) sulfate (2% w/v) solution in 10% sulfuric acid. Flash col-
4.1.1. 5a-Carba-2,3,4,6-tetra-O-benzyl-a-L-idopyranose (4)
Unsaturated alcohol 3 (157 mg, 0.29 mmol) was dissolved in
EtOAc (4 mL), and palladium (10% on carbon, 15 mg) and triethyl-
amine (41 lL, 0.29 mmol) were added. The mixture was degassed
and stirred under an atmosphere of hydrogen. After 2 h, the mix-
ture was filtered through Celite and concentrated in vacuo. The
residue was purified by flash column chromatography (CH2Cl2/
ether, 30:1) to give protected carbaidose 4 (123 mg, 78%) as a col-
ourless oil. ½a 2D2
ꢁ
ꢀ34.6 (c 1.0 in CHCl3);
m
max(film)/cmꢀ1 3376 (br,
OH st); dH (400 MHz, CDCl3) 1.42 (1H, ddd, J5a(pro-S),5a(pro-R)
ꢀ13.6 Hz, H5apro-S), 2.24 (1H, ddd, H5apro-R), 2.46 (1H, br s, OH1),
0
2.52 (1H, ddddd, J5,5a(pro-S) 4.8 Hz, J5,5a(pro-R) 4.3 Hz, J5,6 4.7 Hz, J5,6
8.3 Hz, H5), 3.35 (1H, dd, J2,3 8.2 Hz, H2), 3.53 (1H, dd, H60),
0
3.720 (1H, dd, J4,5 5.0 Hz, H4), 3.721 (1H, dd, J6,6 ꢀ9.4 Hz, H6),
3.73 (1H, dd, J3,4 8.6 Hz, H3), 3.89 (1H, ddd, J1,2 8.4 Hz, J1,5a(pro-S)
10.7 Hz, J1,5a(pro-R) 4.6 Hz, H1), 4.54, 4.57 (2H, 2 ꢂ d, J ꢀ12.2 Hz,
PhCH2), 4.62, 4.66 (2H, 2 ꢂ d, J ꢀ11.5 Hz, PhCH0 ), 4.69, 4.95 (2H,
2
2 ꢂ d, J ꢀ11.4 Hz, PhCH020), 4.73, 4.89 (2H, 2 ꢂ d, J ꢀ10.9 Hz,
PhCH000), 7.27ꢀ7.40 (20H, m, Ph); dC (100 MHz, CDCl3) 29.7 (C5a),
2
34.9 (C5), 68.5 (C6), 68.7 (C1), 72.4, 73.3, 75.0, 75.2 (PhCH2),
81.2, 81.6 (C3, C4), 85.3 (C2), 127.6, 127.7, 127.7, 127.7, 127.9,
127.9, 127.9, 128.0, 128.1, 128.4, 128.5, 128.5, 128.7 (Ph–CH),
138.5, 138.5, 138.6, 138.7 (Ph–C); HRMS (ES+) Calcd for
C35H38O5Na (MNa+) 561.2611; Found 561.2617. Along with a mix-
ture of gluco 5 and ido 4 isomers (6:1, 30 mg, 19%).
umn chromatography was carried out on silica gel (35–70 lm,
Grace). Dichloromethane was distilled from calcium hydride. THF
was distilled from sodium benzophenone ketyl radical. Reactions
performed under an atmosphere of hydrogen or nitrogen were
maintained by an inflated balloon.
4.1.2. 5a-Carba-2,3,4-tri-O-benzyl-a-l-idopyranose (6)
Alcohol 4 (37 mg, 0.069 mmol) was dissolved in a mixture of
Ac2O (1 mL) and TFA (0.25 mL) and stirred at rt. After 55 min,

990
E. Säwén et al. / Carbohydrate Research 345 (2010) 984–993
with NaHCO3 (saturated aqueous, 30 mL). The aqueous phase
H3H4 H2
H5
H5apro-S
H5apro-R
was re-extracted with CH2Cl2 (15 mL). The combined organic ex-
tracts were dried (Na2SO4), filtered and concentrated in vacuo.
The residue was dried under high vacuum then dissolved in MeOH
(1.6 mL). Na (8 mg, 0.35 mmol) was added to MeOH (1.6 mL) and
the resulting solution was added to the sugar solution and the mix-
ture was stirred at rt. After 2 h 40 min, TLC (pentane/EtOAc, 1:1)
showed the formation of a major product (Rf 0.1). The reaction
was quenched with ammonium chloride (saturated aqueous,
30 mL) and extracted with ether (30 mL + 15 mL). The combined
organic extracts were dried (Na2SO4), filtered and concentrated
in vacuo. The residue was purified by flash column chromatogra-
phy (EtOAc/pentane, 3:2) to give the diol 6 (26 mg, 84%) as a col-
3.5
3.0
2.5
1H /ppm
2.0
1.5
ourless oil. ½a 2D1
ꢁ
ꢀ22.1 (c 1.0 in CHCl3) (lit.30 ꢀ19.0 (c 1.07 in
CHCl3));
m
max(film)/cmꢀ1 3384 (br, OH st); dH (400 MHz, CDCl3)
1.45 (1H, ddd, J5a(pro-S),5a(pro-R) ꢀ13.7 Hz, H5apro-S), 1.99 (1H, ddd,
14
10
6
H5apro-R), 2.42 (1H, ddddd, J5,5a(pro-S) 5.0 Hz, J5,5a(pro-R) 4.3 Hz, J5,6
0
4.9 Hz, J5,6 8.2 Hz, H5), 2.63 (1H, br s, OH1), 2.63 (1H, br s, OH6),
0
3.31 (1H, dd, J2,3 8.3 Hz, H2), 3.59 (1H, dd, J6,6 ꢀ11.3 Hz, H6),
3.72 (1H, ddd, J1,2 8.4 Hz, J1,5a(pro-S) 11.0, J1,5a(pro-R) 4.6 Hz, H1),
3.76 (1H, dd, J4,5 5.1 Hz, H4), 3.81 (1H, dd, J3,4 8.7 Hz, H3), 3.95
(1H, dd, H60), 4.67, 4.72 (2H, 2 ꢂ d, J ꢀ11.3 Hz, PhCH2), 4.68, 4.94
(2H, 2 ꢂ d, J ꢀ11.4 Hz, PhCH02), 4.80, 4.88 (2H, 2 ꢂ d, J ꢀ10.9 Hz,
PhCH00), 7.13ꢀ7.36 (15H, m, Ph); dC (100 MHz, CDCl3) 30.3 (C5a),
2
2
0
36.9 (C5), 63.9 (C6), 68.8 (C1), 73.5, 75.2, 75.4 (PhCH2), 81.4 (C3),
82.8 (C4), 85.4 (C2), 127.9, 128.0, 128.1, 128.1, 128.1, 128.6,
128.7, 128.8 (Ph–CH), 137.9, 138.4, 138.6 (Ph–C); HRMS (ES+) Calcd
for C28H32O5Na (MNa+) 471.2142; Found 471.2130.
100
200
300
400
Mixing time /ms
Figure 6. (top) 1D 1H,1H-NOESY spectrum of 1 with selective excitation of the
resonance from H5 and a mixing time of 400 ms; (bottom) 1H,1H cross-relaxation
build-up curves obtained from the 1D 1H,1H NOESY spectra of 1; H5 to H5apro-S
(blue filled squares) and to H4 (cerise filled diamond).
4.1.3. 5a-Carba-2,3,4-tri-O-benzyl-a-L-idopyranuronic acid (7)
Diol 6 (26 mg, 0.058 mmol) was dissolved in CH2Cl2 (1.5 mL),
and water (0.6 mL) was added. TEMPO (4 mg, 0.026 mmol) and
(diacetoxyiodo)benzene (46 mg, 0.15 mmol) were added, and the
mixture was stirred at rt. After 1 h 30 min, TLC (EtOAc) showed
the formation of a major product (Rf 0.7) and the complete con-
sumption of starting material (Rf 0.5). After a further 5 h, no change
TLC (pentane/EtOAc, 3:1) showed conversion to a major compo-
nent (Rf 0.2) and little remaining starting material (Rf 0.1). The
reaction mixture was diluted with CH2Cl2 (25 mL) and washed
Table 3
1H,1H Distances and cross-relaxation rates for 1 derived from NMR experiments and molecular mechanics models
Atom pair
NMR experiment
Molecular mechanics modelsa
T-ROEb
NOEb
4C1
2S5a
1C4
reffc/Å
r
ꢂ 103/sꢀ1
reffc/Å
r
ꢂ 103/sꢀ1
Distance/Å
H5
H5
H5
H5apro-R
H5apro-R
H5apro-S
H4
2.38
2.35
2.47
2.50
1.76d,e
2.42
5.16
2.38
2.35
2.46
2.47
1.76d,e
2.43
4.27
4.66
3.49
3.44
26.1
3.81
2.43
2.46
2.53
2.68
1.76
2.48
2.45
2.54
3.06
3.70
1.76
2.32
2.45
2.47
3.07
4.28
1.76
2.43
H5apro-R
H5apro-S
H2
H5apro-S
H1
5.41d
4.07d
2.82
30.8
4.57
a
b
c
Energy-minimised in the VEGA ZZ software and geometry optimised using the PERCH NMR software.
Selective excitation of the first resonance in the atom pair.
1=6
rij ¼ rref
ðrref
=rij
Þ
.
d
e
Mean values from excitations of both resonances in the atom pair.
Reference distance.
4C1
2S5a
1C4
2
Figure 7. The conformational equilibrium is shifted far to the left with the 4C1 conformation in 1 predominating over the S5a and the 1C4 conformations.

E. Säwén et al. / Carbohydrate Research 345 (2010) 984–993
991
0
was visible by TLC and the reaction mixture was diluted with
CH2Cl2 (20 mL) and washed with Na2S2O3 (10% aqueous, 20 mL).
The organic phase was dried (Na2SO4), filtered and concentrated
in vacuo.
3.70 (1H, dd, J6,6 ꢀ9.4 Hz, H6), 4.50, 4.59 (2H, 2 ꢂ d, J ꢀ12.2 Hz,
PhCH2), 4.61, 4.65 (2H, 2 ꢂ d, J ꢀ11.5 Hz, PhCH0 ), 4.76, 4.84 (2H,
2
2 ꢂ d, J ꢀ10.8 Hz, PhCH020), 4.80, 4.82 (2H, 2 ꢂ d, J ꢀ11.0 Hz,
PhCH000), 7.26–7.37 (20H, m, Ph); dC (100 MHz, CDCl3) 27.9 (C5a),
2
The crude aldehyde was dissolved in a mixture of n-butanol
(2.5 mL) and DMSO (2.5 mL). NaClO2 (250 mg) was dissolved in
water (5 mL) and the pH was adjusted to 4.5 by addition of sodium
phosphate (monobasic). 1,3-Dimethoxybenzene (10 lL) was added
to the aldehyde solution, then aqueous chlorite solution (0.44 mL)
35.4 (C5), 58.2 (OMe), 67.8 (C6), 72.5, 73.3, 75.7, 75.8 (PhCH2),
79.5 (C1), 81.7, 82.5 (C3, C4), 85.9 (C2), 127.6, 127.6, 127.7,
127.7, 127.8, 127.8, 128.1, 128.1, 128.4, 128.4, 128.5, 128.6 (Ph–
CH), 138.6, 138.7, 139.1, 139.1 (Ph–C); HRMS (ES+) Calcd for
C36H40O5Na (MNa+) 575.2768; Found 575.2761.
was added, and the mixture was stirred at rt. After 20 min, further
1,3-dimethoxybenzene (70
lL) and chlorite solution (0.44 mL)
4.1.6. Methyl 5a-carba-2,3,4-tri-O-benzyl-a-L-idopyranoside (9)
were added and the mixture was stirred further. After 1 h, ether
(25 mL) was added, and the mixture was washed with water
(25 mL) and brine (25 mL). The organic phase was dried (Na2SO4),
filtered and concentrated in vacuo. TLC (EtOAc) showed the pres-
ence of a major component (Rf 0.1) and a faster component (Rf
0.9). The residue was purified by flash column chromatography
Fully protected 8 (66 mg, 0.12 mmol) was dissolved in a mix-
ture of Ac2O (2 mL) and TFA (0.5 mL) and stirred at rt. After
55 min, TLC (pentane/EtOAc, 3:1) showed conversion to a major
component (Rf 0.5) and no remaining starting material (Rf 0.55).
The reaction mixture was diluted with CH2Cl2 (25 mL) and washed
with NaHCO3 (saturated aqueous, 30 mL). The aqueous phase was
re-extracted with CH2Cl2 (15 mL). The combined organic extracts
were dried (Na2SO4), filtered and concentrated in vacuo. The resi-
due was dried under high vacuum then dissolved in MeOH
(1.6 mL). Na (8 mg, 0.35 mmol) was added to MeOH (1.6 mL) and
the resulting solution was added to the sugar solution and the mix-
ture was stirred at rt. After 1 h, TLC (pentane/EtOAc, 1:1) showed
the formation of a major product (Rf 0.5). The reaction was
quenched by the addition of Dowex IR-120 (H+) resin, filtered
and concentrated in vacuo. The residue was purified by flash col-
umn chromatography (EtOAc/pentane, 4:5) to give the alcohol 9
(EtOAc) to give the acid 7 (18 mg, 67%) as a colourless oil. ½a D21
ꢁ
ꢀ31.5 (c 1.0 in CHCl3);
m
max(film)/cmꢀ1 1728 (s, C@O st); dH
(400 MHz, CDCl3) 1.48 (1H, ddd, J5a(pro-S),5a(pro-R) ꢀ13.9 Hz,
H5apro-S), 2.34 (1H, ddd, H5apro-R), 3.06 (1H, ddd, J5,5a(pro-S) 5.1 Hz,
J5,5a(pro-R) 3.9 Hz, H5), 3.29 (1H, dd, J2,3 8.4 Hz, H2), 3.75 (1H, dd,
J4,5 5.6 Hz, H4), 3.92 (1H, dd, J3,4 8.9 Hz, H3), 3.97 (1H, ddd, J1,2
8.6 Hz, J1,5a(pro-S) 10.8 Hz, J1,5a(pro-R) 4.6 Hz, H1), 4.68, 4.94 (2H,
2 ꢂ d, J ꢀ11.4 Hz, PhCH2), 4.75 (2H, s, PhCH0 ), 4.79, 4.87 (2H,
2
2 ꢂ d, J ꢀ10.9 Hz, PhCH00), 7.28ꢀ7.35 (15H, m, Ph); dC (100 MHz,
2
CDCl3) 29.0 (C5a), 40.9 (C5), 68.7 (C1), 73.9, 75.2, 75.6 (PhCH2),
80.1 (C4), 81.4 (C3), 84.9 (C2), 127.9, 128.0, 128.0, 128.2, 128.3,
128.4, 128.6, 128.7, 128.8 (Ph–CH), 137.2, 138.3, 138.6 (Ph–C),
175.4 (C@O); HRMS (ES+) Calcd for C28H30O6Na (MNa+) 485.1935;
Found 485.1923.
(36 mg, 65%) as a colourless oil. ½a D21
ꢁ
+4.8 (c 1.0 in CHCl3); mmax
(film)/cmꢀ1 3440 (br, OH st); dH (400 MHz, CDCl3) 1.34 (1H, ddd,
J5a(pro-S),5a(pro-R) ꢀ13.8 Hz, H5apro-S), 2.07 (1H, ddd, H5apro-R), 2.43
0
(1H, ddddd, J5,5a(pro-S) 5.0 Hz, J5,5a(pro-R) 3.4 Hz, J5,6 5.1 Hz, J5,6
8.3 Hz, H5), 2.72 (1H, br s, OH6), 3.35 (1H, ddd, J1,2 8.8 Hz, J1,5a(pro-S)
11.6 Hz, J1,5a(pro-R) 4.8 Hz, H1), 3.40 (1H, dd, J2,3 8.8 Hz, H2), 3.45
4.1.4. 5a-Carba-a-L-idopyranuronic acid (1)
0
Protected acid 7 (15 mg, 0.032 mmol) was dissolved in a mix-
ture of ethanol (2.25 mL) and acetic acid (0.25 mL). Palladium ace-
tate (8 mg) was added and it dissolved to give a yellow solution.
The mixture was degassed and stirred under an atmosphere of
hydrogen. After 14 h, a black suspension had appeared and the
solution was colourless. TLC (CHCl3/MeOH/AcOH/H2O, 60:30:3:5)
showed the formation of a major polar product (Rf 0.2). The mix-
ture was filtered through Celite and concentrated in vacuo. The
residue was purified on a Waters Sep-Pak C-18 cartridge eluting
with water, to give the deprotected carbasugar 1 (6 mg, 96%) as a
(3H, s, OMe), 3.58 (1H, dd, J6,6 ꢀ11.2 Hz, H6), 3.73 (1H, dd, J4,5
5.2 Hz, H4), 3.77 (1H, dd, J3,4 9.3 Hz, H3), 3.98 (1H, dd, H60), 4.69,
4.74 (2H, 2 ꢂ d, J ꢀ11.3 Hz, PhCH2), 4.81, 4.86 (2H, 2 ꢂ d, J
ꢀ10.8 Hz, PhCH02), 4.85, 4.86 (2H, 2 ꢂ d, J ꢀ10.9 Hz, PhCH020),
7.28ꢀ7.36 (15H, m, Ph); dC (100 MHz, CDCl3) 28.9 (C5a), 37.3
(C5), 58.2 (OMe), 63.8 (C6), 73.8, 75.7, 75.8 (PhCH2), 79.7 (C1),
82.1 (C3), 83.2 (C4), 85.9 (C2), 127.7, 127.7, 128.0, 128.1, 128.1,
128.1, 128.5, 128.5, 128.7 (Ph–CH), 138.1, 138.9, 139.0 (Ph–C);
HRMS (ES+) Calcd for C29H34O5Na (MNa+) 485.2298; Found
485.2303.
colourless oil. ½a D22
ꢁ
ꢀ38 (c, 0.3 in MeOH); 1H NMR and 13C NMR:
See Table 1. HRMS (ES+) Calcd for C7H12O6Na (MNa+) 215.0526;
Found 215.0530.
4.1.7. Methyl 5a-carba-2,3,4-tri-O-benzyl-a-L-
idopyranosiduronic acid (10)
Alcohol 9 (35 mg, 0.076 mmol) was dissolved in acetone (2 mL),
and water (1 mL) was added. TEMPO (5 mg, 0.030 mmol) and
diacetoxyiodobenzene (65 mg, 0.19 mmol) were added, and the
mixture was stirred at rt. After 3 h, TLC (EtOAc/pentane, 2:1)
showed the formation of a major product (Rf 0.9) and the complete
consumption of starting material (Rf 0.6). After a further 5 h, signif-
icant amounts of a new product (Rf 0.2) had appeared, but much of
the initially formed product remained. After a further 18 h, further
diacetoxyiodobenzene (37 mg, 0.11 mmol) was added. After a fur-
ther 24 h, TLC (EtOAc/pentane, 2:1) showed the polar compound
(Rf 0.2) to be the major reaction component, and very little of the
initially formed product (Rf 0.9) remaining. The mixture was di-
luted with CH2Cl2 (30 mL) and washed with Na2S2O3. HCl (1 M,
20 mL) was added to the aqueous phase, which was then extracted
with CH2Cl2 (20 mL). The combined organic extracts were dried
(Na2SO4), filtered and concentrated in vacuo. The residue was puri-
fied by flash column chromatography (EtOAc/pentane, 2:1) to give
4.1.5. Methyl 5a-carba-2,3,4,6-tetra-O-benzyl-a-L-idopyranoside
(8)
Alcohol 4 (70 mg, 0.13 mmol) was dissolved in THF (2 mL), and
methyl iodide (16 L, 0.26 mmol) and sodium hydride (60% in oil,
16 mg, 0.39 mmol) were added. The mixture was stirred under N2
and further methyl iodide (16 L, 0.26 mmol) was added after 2 h.
l
l
After a further 3 h 30 min, TLC (pentane/EtOAc, 3:1) showed the
formation of a major product (Rf 0.8), and almost no remaining
starting material (Rf 0.3). The reaction was quenched by the addi-
tion of methanol, then silica was added and the solvent was re-
moved in vacuo. The residue was purified directly by flash
column chromatography (pentane/EtOAc, 6:1–4:1) to give the
methyl ether 8 (66 mg, 92%) as a colourless oil. ½a D21
ꢀ27.2 (c 1.0
ꢁ
in CHCl3); dH (400 MHz, CDCl3) 1.24 (1H, ddd, J5a(pro-S),5a(pro-R)
ꢀ13.6 Hz, H5apro-S), 2.34 (1H, ddd, H5apro-R), 2.49 (1H, ddddd,
0
J5,5a(pro-S) 4.5 Hz, J5,5a(pro-R) 3.2 Hz, J5,6 4.2 Hz, J5,6 9.4 Hz, H5), 3.38
(1H, dd, J2,3 9.1 Hz, H2), 3.448 (1H, ddd, J1,2 9.4 Hz, J1,5a(pro-S)
11.3 Hz, J1,5a(pro-R) 5.0 Hz, H1), 3.449 (3H, s, OMe), 3.49 (1H, dd,
H60), 3.61 (1H, dd, J3,4 9.5 Hz, H3), 3.63 (1H, dd, J4,5 5.2 Hz, H4),
the acid 10 (23 mg, 64%) as a colourless oil. ½a D21
ꢀ9.1 (c 1.0 in
ꢁ
CHCl3);
m
max(film)/cmꢀ1 1731 (s, C@O st); dH (400 MHz, CDCl3)
1.31 (1H, dd, J5a(pro-S),5a(pro-R) ꢀ13.8 Hz, H5apro-S), 2.48 (1H, dd,

992
E. Säwén et al. / Carbohydrate Research 345 (2010) 984–993
H5apro-R), 3.00 (1H, ddd, J5,5a(pro-S) 4.8 Hz, J5,5a(pro-R) 3.3 Hz, H5), 3.37
(1H, dd, J2,3 8.9 Hz, H2), 3.48 (3H, s, OMe), 3.59 (1H, ddd, J1,2 8.9 Hz,
J1,5a(pro-S) 11.3 Hz, J1,5a(pro-R) 4.8 Hz, H1), 3.71 (1H, dd, J4,5 5.8 Hz,
lected 13C frequencies were irradiated by 45 ms 180° Gaussian
shaped pulses (20 Hz bandwidth) and encoded as on or off with
a Hadamard-8 matrix using 32 scans. Adiabatic 180° pulses were
used for inversion and refocusing on 13C. For 1H decoupling, the
same pulse length (45 ms) as for the irradiation was used with adi-
abatic 180° pulses of a 6.5 kHz sweep width.
H4), 3.78 (1H, dd, J3,4 9.6 Hz, H3), 4.75, 4.80 (2H, 2 ꢂ d,
J
ꢀ11.5 Hz, PhCH2), 4.79, 4.87 (2H, 2 ꢂ d, J ꢀ10.9 Hz, PhCH0 ), 4.81,
2
4.84 (2H, 2 ꢂ d, J ꢀ10.7 Hz, PhCH00), 7.26ꢀ7.35 (15H, m, Ph); dC
2
(100 MHz, CDCl3) 27.4 (C5a), 41.1 (C5), 58.4 (OMe), 74.1, 75.5,
75.9 (PhCH2), 79.4 (C1), 80.1 (C4), 81.8 (C3), 85.1 (C2), 127.6,
127.8, 128.0, 128.1, 128.2, 128.3, 128.4, 128.5, 128.5, 128.8 (Ph–
CH), 137.1, 138.6, 139.0 (Ph–C), 175.0 (C@O); HRMS (ES+) Calcd
for C29H32O6Na (MNa+) 499.2091; Found 499.2116.
The 1H,13C-HSQC-HECADE experiment43 was performed over a
spectral width of 14 ppm for 1H and 65 ppm for 13C. A 40 ms DIP-
SI-2 spin-lock with (cB1/2p
= 9.6 kHz) was applied and a tꢄ1=t1 scal-
ing factor of unity. A 2D matrix of 4096 ꢂ 256 points, 16 scans per
t1 increment and the echo/anti-echo method were used. Prior to
Fourier transformation zero-filling to 8192 ꢂ 1024 points was per-
formed and 90° shifted squared sine-bell functions were applied in
both dimensions.
4.1.8. Methyl 5a-carba-a-L-idopyranosiduronic acid (2)
Protected acid 10 (13 mg, 0.027 mmol) was dissolved in a mix-
ture of methanol (2.25 mL) and acetic acid (0.25 mL). Palladium
acetate (7 mg) was added and it dissolved to give a yellow solution.
The mixture was degassed and stirred under an atmosphere of
hydrogen. After 22 h, a black suspension had appeared and the
solution was colourless. TLC (CHCl3/MeOH/AcOH/H2O, 60:30:3:5)
showed the formation of a major polar product (Rf 0.3). The mix-
ture was filtered through Celite and concentrated in vacuo. The
residue was purified on a Waters Sep-Pak C-18 cartridge, eluting
with water to give the deprotected carbasugar 2 (5 mg, 89%) as a
In the J-HMBC experiment44 a three-fold low-pass J-filter
1
(J = 120 Hz, 150 Hz and 190 Hz) was used to suppress the JC,H
.
Scale factors between 15 and 24, calculated from
j
¼ ð0:6=nJCmHinÞ=tm1 ax, were used to scale the coupling constants in
the indirect dimension. Spectral widths of 10 ppm for 1H and
100 ppm for 13C were used. The experiments were performed with
3072 ꢂ 512 points and 64 scans per t1 increment with the echo/
anti-echo method. Forward linear prediction to 1024 points in F1
and subsequent zero-filling to 2048 ꢂ 8192 points was applied
prior to Fourier transformation. Coupling constants were extracted
from 1D-projections of the resonances of interest.
colourless oil. ½a D22
ꢁ
ꢀ52 (c 0.2 in MeOH); m/z (ES+) 435 (2 M+Na+,
40), 229 (M+Na+, 100%); 1H NMR and 13C NMR: See Table 1. HRMS
(ES+) Calcd for C16H28O12Na (2MNa+) 435.1473; Found 435.1467.
The IPAP-HSQC-TOCSY-HT experiment35,45 was carried out with
zero-quantum suppression as devised by Thrippleton and Keeler,46
4.2. NMR spectroscopy
with 512 scans and a 50 ms long spin-lock (cB1/2p = 5 kHz); other-
wise the same parameters were used as in the HSQC-HT experi-
ment. Zero-filling was carried out to 2048 ꢂ 256 points. In the 1H
dimension an exponential line-broadening factor of 1 Hz was ap-
plied. Following baseline correction in F2, coupling constants were
extracted from the 1D IPAP selections of the corresponding 13C
frequencies.
NMR experiments on 1 (44 mM) and 2 (24 mM) in D2O (pD ꢃ 6)
were performed at 27 °C on three different spectrometers, a Varian
INOVA 600 MHz spectrometer equipped with a 5 mm PFG triple
resonance probe, a Bruker AVANCE 500 MHz spectrometer and a
Bruker AVANCE III 700 MHz spectrometer; the latter two equipped
with 5 mm TCI Z-Gradient CryoProbes. 1H chemical shifts were ref-
erenced to internal TSP (dH 0.00) and 13C chemical shifts were ref-
erenced to external dioxane in D2O (dC 67.40). 1H chemical shift
assignments were performed using 1H,1H-TOCSY experiments re-
corded over 5 ppm with 1466 ꢂ 128 points in F2 ꢂ F1, respectively,
and 8 scans per t1 increment using the States-TPPI method. An
MLEV-17 spin-lock of 12.7 kHz and two different mixing times,
10 ms and 50 ms, were used. Zero-filling was performed to
4096 ꢂ 512 points. Prior to Fourier transformation a 90° shifted
sine-bell function was applied in the direct dimension and a 90°
shifted squared sine-bell function was used in the indirect dimen-
sion. For 2 a 1H,13C-H2BC experiment34 was recorded over 10 ppm
in the direct dimension and 165 ppm in the indirect dimension
with 2048 ꢂ 256 points and 64 scans per t1 increment with the
echo/anti-echo method. A constant time delay of 22 ms was used
for JH,H evolution. Zero-filling to 2048 ꢂ 1024 points and a 90°
shifted squared sine-bell function was applied prior to Fourier
transformation. The H2BC spectrum was processed in magnitude
mode. For 2 a 1H,13C-HMBC experiment was also recorded over
5 ppm in the direct dimension and 96 ppm in the indirect dimen-
sion with 2048 ꢂ 256 points and 64 scans per t1 increment with
the States-TPPI method. A 50 ms long evolution time for long-
range coupling constants was used. Zero-filling to 2048 ꢂ 1024
points and a Gaussian line-broadening factor with the maximum
at a fraction of 0.28 of the fid were applied prior to Fourier trans-
formation. 1H chemical shifts and JH,H were refined from 1D 1H
spectra with the PERCH NMR software37
1H,1H cross-relaxation rates were measured using 1D 1H,1H-
DPFGSE NOESY48 with zero-quantum suppression and 1D 1H,1H-
DPFGSE T-ROESY49 experiments at 600 MHz. Selective excitations
of H5, H5apro-R and H5apro-S were achieved using 60 Hz broad i-
SNOB-2 shaped pulses of 28 ms duration. Six different mixing
times between 50 ms and 300 ms were used. The spectral width
was 5 ppm sampled with 24,314 points. A relaxation delay of
8.5 s was used. The T-ROESY spin-lock strength was 3.0 kHz. Pulsed
field gradients of 10 ms were used. Zero-filling to 65,536 points
and an exponential line-broadening of 1 Hz were used prior to Fou-
rier transformation. All spectra were baseline corrected and inte-
grated with the same limits in all mixing times.
Acknowledgements
This work was supported by grants from the Swedish Research
Council (VR), The Knut and Alice Wallenberg Foundation, Magn.
Bergvalls Stiftelse and Magnus Ehrnrooths Stiftelse (to M.U.R.).
¯
ˇ
Drs. E. Kupce and T. Nishida are thanked for stimulating discus-
sions and the pulse sequence for the IPAP-HSQC-TOCSY-HT NMR
experiment.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.carres.2010.03.002.
13C chemical shifts were assigned using coupled and decoupled
1H,13C-HSQC-HT experiments.35 Prior to recording the Hadamard
transform (HT) experiment, a 13C spectrum with a spectral width
of 96 ppm was performed and from peak-picking, six or seven
13C frequencies were selected for 1 and 2, respectively. The se-
References
1. Ferro, D. R.; Provasoli, A.; Ragazzi, M.; Torri, G.; Casu, B.; Gatti, G.; Jacquinet, J.-
C.; Sinaÿ, P.; Petitou, M.; Choay, J. J. Am. Chem. Soc. 1986, 108, 6773.
2. Sanderson, P. N.; Huckerby, T. N.; Nieduszynski, I. A. Biochem. J. 1987, 243, 175.

E. Säwén et al. / Carbohydrate Research 345 (2010) 984–993
993
3. Ferro, D. R.; Provasoli, A.; Ragazzi, M.; Casu, B.; Torri, G.; Bossennec, V.; Perly,
B.; Sinaÿ, P.; Petitou, M.; Choay, J. Carbohydr. Res. 1990, 195, 157.
4. Jin, L.; Abrahams, J. P.; Skinner, R.; Petitou, M.; Pike, R. N.; Carrell, R. W. Proc.
Natl. Acad. Sci. U.S.A. 1997, 94, 14683.
5. Das, S. K.; Mallet, J.-M.; Esnault, J.; Driguez, P.-A.; Duchaussoy, P.; Sizun, P.;
Herault, J.-P.; Herbert, J.-M.; Petitou, M.; Sinaÿ, P. Chem. Eur. J. 2001, 7, 4821.
6. Kovensky, J.; Mallet, J.-M.; Esnault, J.; Driguez, P.-A.; Sizun, P.; Hérault, J.-P.;
Herbert, J.-M.; Petitou, M.; Sinaÿ, P. Eur. J. Org. Chem. 2002, 3595.
7. Herrera, A. J.; Beneitez, M. T.; Amorim, L.; Cañada, F. J.; Jiménez-Barbero, J.;
Sinaÿ, P.; Blériot, Y. Carbohydr. Res. 2007, 342, 1876.
27. Boyd, D. R.; Sharma, N. D.; Llamas, N. M.; Malone, J. F.; O’Dowd, C. R.; Allen, C. C.
R. Org. Biomol. Chem. 2005, 3, 1953.
28. Sedemera, P.; Halada, P.; Pospíšil, S. Magn. Reson. Chem. 2009, 47, 519.
29. Cumpstey, I.; Gehrke, S.; Erfan, S.; Cribiu, R. Carbohydr. Res. 2008, 343, 1676.
30. Paulsen, H.; Von Deyn, W. Liebigs Ann. Chem. 1987, 125.
31. Prosperi, D.; Ronchi, S.; Lay, L.; Rencurosi, A.; Russo, G. Eur. J. Org. Chem. 2004,
395.
32. van den Bos, L. J.; Codée, J. D. C.; van der Toorn, J. C.; Boltje, T. J.; van Boom, J. H.;
Overkleeft, H. S.; van der Marel, G. A. Org. Lett. 2004, 6, 2165.
33. Benowitz, A. B.; Fidanze, S.; Small, P. L. C.; Kishi, Y. J. Am. Chem. Soc. 2001, 123,
5128.
8. Cros, C.; Petitou, M.; Sizun, P.; Pérez, S.; Imberty, A. Bioorg. Med. Chem. 1997, 5,
1301.
34. Nyberg, N. T.; Duus, J. Ø.; Sørensen, O. W. J. Am. Chem. Soc. 2005, 127, 6154.
ˇ
9. Mikhailov, D.; Linhardt, R. J.; Mayo, K. H. Biochem. J. 1997, 328, 51.
10. Verli, H.; Guimarães, J. A. Carbohydr. Res. 2004, 339, 281.
11. Becker, C. F.; Guimarães, J. A.; Verli, H. Carbohydr. Res. 2005, 340, 1499.
12. Canales, A.; Angulo, J.; Ojeda, R.; Bruix, M.; Fayos, R.; Lozano, R.; Giménez-
Gallego, G.; Martín-Lomas, M.; Nieto, P. M.; Jiménez-Barbero, J. J. Am. Chem. Soc.
2005, 127, 5778.
35. Kupce, E.; Nishida, T.; Freeman, R. Prog. Nucl. Magn. Reson. Spectrosc. 2003, 42,
95.
¯
ˇ
36. Kupce, E.; Freeman, R. J. Magn. Reson. 2003, 163, 56.
37. Laatikainen, R.; Niemitz, M.; Weber, U.; Sundelin, J.; Hassinen, T.; Väpsäläinen,
J. J. Magn. Reson., Ser. A 1996, 120, 1.
38. Haasnoot, C. A. G.; de Leeuw, F. A. A. M.; Altona, C. Tetrahedron 1980, 36, 2783.
39. Evans, D. A.; Bodkin, M. J.; Baker, S. R.; Sharman, G. J. Magn. Reson. Chem. 2007,
45, 595.
13. Remko, M.; von der Lieth, C.-W. J. Chem. Inf. Model. 2006, 46, 1194.
14. Hricoví, M. Carbohydr. Res. 2006, 341, 2575.
15. Hricoví, M.; Bízik, F. Carbohydr. Res. 2007, 342, 779.
40. Marquez, B. L.; Gerwick, W. H.; Williamsson, R. T. Magn. Reson. Chem. 2001, 39,
499.
16. Hricoví, M.; Scholtzova, E.; Bízik, F. Carbohydr. Res. 2007, 342, 1350.
17. McCasland, G. E.; Furuta, S.; Durham, L. J. J. Org. Chem. 1966, 31, 1516.
18. Ogawa, S. Trends Glycosci. Glycotechnol. 2004, 16, 33.
19. Arjona, O.; Gómez, A. M.; López, J. C.; Plumet, J. Chem. Rev. 2007, 107, 1919.
20. Ogawa, S.; Ara, M.; Kondoh, T.; Saitoh, M.; Masuda, R.; Toyokuni, T.; Suami, T.
Bull. Chem. Soc. Jpn. 1980, 53, 1121.
21. Barton, D. H. R.; Géro, S. D.; Cléophax, J.; Salustiano Machado, A.; Quiclet-Sire,
B. J. Chem. Soc., Chem. Commun. 1988, 1184.
22. Yoshikawa, M.; Cha, B. C.; Nakae, T.; Kitagawa, I. Chem. Pharm. Bull. 1988, 36,
3714.
41. Nishida, T.; Widmalm, G.; Sandor, P. Magn. Reson. Chem. 1995, 33, 596.
42. Vidal, P.; Esturau, N.; Parella, T.; Espinosa, J. F. J. Org. Chem. 2007, 72, 3166.
43. Koz´min´ ski, W.; Nanz, D. J. Magn. Reson. 2000, 142, 294.
44. Meissner, A.; Sørensen, O. W. Magn. Reson. Chem. 2001, 39, 49.
45. Ottiger, M.; Delaglio, F.; Bax, A. J. Magn. Reson. 1998, 131, 373.
46. Thrippleton, M. J.; Keeler, J. Angew. Chem., Int. Ed. 2003, 42, 3938.
47. Wasylishen, R.; Schaefer, T. Can. J. Chem. 1973, 51, 961.
48. Stott, K.; Keeler, J.; Van, Q. N.; Shaka, A. J. J. Magn. Reson. 1997, 125, 302.
49. Kjellberg, A.; Widmalm, G. Biopolymers 1999, 50, 391.
50. Carpintero, M.; Fernández-Mayoralas, A.; Jiménez-Barbero, J. Eur. J. Org. Chem.
2001, 681.
23. Larsen, D. S.; Lins, R. J.; Stoodley, R. J.; Trotter, N. S. Org. Biomol. Chem. 2004, 2,
1934.
24. Yu, S.-H.; Chung, S.-K. Tetrahedron: Asymmetry 2005, 2729, 16.
25. Ramana, C. V.; Chaudhuri, S. R.; Gurjar, M. K. Synthesis 2007, 523.
26. Sollogoub, M.; Pearce, A. J.; Hérault, A.; Sinaÿ, P. Tetrahedron: Asymmetry 2000,
11, 283.
51. Carpintero, M.; Bastida, A.; García-Junceda, E.; Jiménez-Barbero, J.; Fernández-
Mayoralas, A. Eur. J. Org. Chem. 2001, 4127.
52. Jiménez-Barbero, J.; Díaz, M. D.; Nieto, P. M. Anti-Cancer Agents Med. Chem.
2008, 8, 52.