The stereoselectivities of tributyltin hydride-mediated reductions of 5-bromo-d-glucuronides to l-iduronides are dependent on the anomeric substituent: Syntheses and DFT calculations

Article abstract of DOI:10.1039/c6ob00283h

One of the shortest synthetic routes to l-iduronic acid derivatives is via free radical reduction of the C-5 bromide of the corresponding protected d-glucuronic acid derivative. The epimerization of such C-5 bromides to the l-ido derivatives via reaction with tributyltin hydride was investigated. It was found that the stereoselectivity of the reaction was dependent on the anomeric substituent. If the substituent was fluoride the l-ido product was obtained exclusively in 65-72% yield whereas the O-methyl or O-acetyl derivatives led to isomeric mixtures of both the l-ido and d-gluco products in different ratios depending on the reaction conditions. DFT calculations were performed to determine the stereoelectronic factors that favour formation of the l-ido isomer from the fluoride and suggest the selectivity is due to a transition state gauche effect and an Sn-F interaction.

Full text of DOI:10.1039/c6ob00283h

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
The stereoselectivities of tributyltin hydride-
mediated reductions of 5-bromo-D-glucuronides
to L-iduronides are dependent on the anomeric
substituent: syntheses and DFT calculations†
Cite this: DOI: 10.1039/c6ob00283h
Shifaza Mohamed, Elizabeth H. Krenske* and Vito Ferro*
One of the shortest synthetic routes to L-iduronic acid derivatives is via free radical reduction of the C-5
bromide of the corresponding protected D-glucuronic acid derivative. The epimerization of such C-5 bro-
mides to the L-ido derivatives via reaction with tributyltin hydride was investigated. It was found that the
stereoselectivity of the reaction was dependent on the anomeric substituent. If the substituent was
fluoride the L-ido product was obtained exclusively in 65–72% yield whereas the O-methyl or O-acetyl
derivatives led to isomeric mixtures of both the L-ido and D-gluco products in different ratios depending
on the reaction conditions. DFT calculations were performed to determine the stereoelectronic factors
that favour formation of the L-ido isomer from the fluoride and suggest the selectivity is due to a transition
state gauche effect and an Sn–F interaction.
Received 3rd February 2016,
Accepted 9th February 2016
DOI: 10.1039/c6ob00283h
www.rsc.org/obc
advantages and disadvantages, and these have been compre-
hensively reviewed recently.16–18 Amongst these method-
Introduction
The monosaccharide L-iduronic acid (IdoA) is a key component ologies,
a common approach involves the inversion of
of the glycosaminoglycans (GAGs) heparin, heparan sulfate configuration at C-5 of the more abundant D-gluco configured
(HS) and dermatan sulfate (DS). GAGs are complex, highly sul- derivatives, which differ from L-idose only by the stereo-
fated polysaccharides composed of repeating disaccharide sub- chemistry at this carbon. Commonly used methods for inver-
units of a uronic acid, either IdoA or D-glucuronic acid (GlcA), sion of configuration at C-5 include SN2 displacement of
(1 → 4)-linked to D-glucosamine (for heparin/HS)1–3 or (1 → 3)- sulfonates,19–25 base catalysed epimerization of GlcA deriva-
linked to N-acetyl-D-galactosamine (for DS).4,5 GAGs mediate tives,26,27 and radical induced epimerization at C-5.28–32 The
numerous biological processes via their interactions with a latter method involves the tributyltin hydride-mediated
range of structurally diverse proteins such as growth factors, reduction of the readily derived C-5 bromide33,34 of a protected
enzymes, morphogens, cell adhesion molecules and GlcA derivative, which usually results in an isomeric mixture
cytokines.4–8 GAGs are therefore key players in various disease of both L-ido and D-gluco products requiring chromatographic
processes such as angiogenesis, metastasis, inflammation and separation.
viral infections. The synthesis of well-defined GAG oligo-
The isomeric ratios for the C-5 epimerization of protected
saccharides is thus of great interest for identifying therapeuti- GlcA derivatives using tributyltin hydride that have been
cally useful sequences and elucidating structure–activity reported in the literature show significant variation depending
relationships.9–12 The main challenge in the synthesis of GAG on the reaction conditions and the nature of the anomeric sub-
oligosaccharides is the efficient preparation of IdoA building stituent. In 1986, Chiba and Sinaÿ reported28 that when
blocks since neither IdoA nor L-idose are commercially avail- methyl 1,2,3,4-tetra-O-acetyl-5-C-bromo-β-D-glucopyranosyluro-
able or readily accessible from natural sources.
nate 1 was reduced with tributyltin hydride in toluene at reflux
Numerous synthetic approaches for the preparation of IdoA the L-ido to D-gluco product ratio was 1 : 2.3, with the L-idopyra-
and/or L-idose have been reported,13–15 each with their own nuronate 2 isolated in only 27% yield (Scheme 1). When the
β-anomeric substituent was changed from acetyl to methoxy,
reduction of the derived C-5 bromide 7 gave an improvement
The University of Queensland, School of Chemistry and Molecular Biosciences,
in selectivity to 1 : 1.2, with L-idopyranuronate 8 isolated in
Brisbane, QLD 4072, Australia. E-mail: v.ferro@uq.edu.au, e.krenske@uq.edu.au
38% yield.28 Medaković29 subsequently reported that the
†Electronic supplementary information (ESI) available: Computational details,
computational data, and copies of 1H and 13C NMR spectra of compounds 1–12
and 14–15. See DOI: 10.1039/c6ob00283h
selectivity in favour of the L-ido product could be significantly
improved by changing the anomeric configuration of the start-
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 1 Tributyltin hydride mediated reduction of 5-C-bromides of D-glucuronic acid derivatives to give L-ido and D-gluco configured products.
ing material from β to α. A lowering of the reaction tempera- the synthesis of iduronate-2-sulfatase substrates for newborn
ture also resulted in a higher yield of the desired product with screening of MPS II (Hunter syndrome), an important lyso-
less decomposition. Thus, reduction of bromide 4 with tri- somal storage disease.32 This route was thus chosen for further
butyltin hydride in benzene at reflux gave a ratio of 3.3 : 1 in investigation as part of a program aimed at the synthesis of
favour of the L-ido product which was isolated in 67% yield. orthogonally protected IdoA derivatives. Herein are described
More recently, Wong and co-workers30 reported the same iso- our own studies of the tributyltin hydride mediated reduction
meric ratio of 3.3 : 1 in favour of the L-ido isomer upon of GlcA C-5 bromides, which show that a β-F substituent at the
reduction of 4 in benzene and an improved ratio of 2 : 1 in anomeric carbon leads to high selectivity in favour of the
favour of the L-ido isomer upon reduction of 1 under the same desired L-ido product. Density functional theory (DFT) calcu-
conditions. There have also been reports of the reduction of lations are performed to understand the role of the β-F
the 5-C-bromo-β-fluoride 10 with tributyltin hydride in both substituent.
toluene and benzene.31,32 Voznyi et al.31 carried out the reac-
tion in toluene and obtained a product ratio of 2.6 : 1 in favour
of the L-idopyranosyl fluoride 11 which was isolated in 67%
Results and discussion
yield. Interestingly, Gelb and coworkers32 obtained 11 exclu-
sively in 64% yield when the reaction was carried out in
benzene and did not report any evidence of the formation of
the D-glucopyranosyl fluoride 12. The above ratios of L-ido to
D-gluco products are based on isolated yields. The results indi-
cate that both the nature and the stereochemistry of the
anomeric substituent as well as the temperature play a role in
the stereochemical outcome of the radical reduction reaction.
Despite the limitations of the above route, it remains one of
the most direct methods to access IdoA building blocks and
the fluoride 11, prepared in this fashion, has been utilized for
In order to study the influence of various anomeric substitu-
ents on the stereochemical outcome of C-5 epimerization by
tributyltin hydride reduction, several derivatives of GlcA were
prepared, photobrominated and then reduced with tributyltin
hydride. Firstly, D-glucuronolactone 13 was initially treated
with NaOMe and then acetylated under acidic conditions
using perchloric acid in acetic anhydride, according to the pro-
cedure reported by Bollenback et al.,35 to give methyl tetra-O-
acetyl-D-glucuronolactone as a mixture of anomers. Samples
of the pure anomers 3 and 6 were then isolated by fractional
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 2 Synthesis of C-5 bromides of D-glucuronic acid derivatives. Reagents and conditions: (a) i. NaOMe/MeOH, ii. Ac2O/HClO4; (b) NBS/CCl4,
light, reflux, 2 h, 75% for 1, 2 h, 39% for 4, 1.5 h, 49% for 7, 24 h, 44% for 10; (c) 30% HBr-HOAc, 2 h, 62%; (d) AgCO3, MeOH, 91% (e) AgF, MeCN,
94%; (f) HF·Pyr, 19%.
crystallization and flash chromatography (Scheme 2) and their spectra were all in agreement with the proposed structure.
structures confirmed by NMR spectroscopy. The β-anomer 3 Both 1 and 7 were then converted into mixtures of their
was then utilized for the synthesis of the other GlcA derivatives respective starting materials (3 and 9) and their L-ido config-
9, 12 and 15. Treatment of 3 with HF.pyridine complex gave ured counterparts (2 and 8) upon reduction with tributyltin
the α-fluoride 15 in low yield (19%).36 Reaction of 3 with HBr hydride (see below).
in acetic acid gave the bromide 1435 which was glycosylated
Tributyltin hydride mediated reduction of the C-5 bromide
with methanol under Koenigs–Knorr conditions in the pres- derivatives 1, 4, 7 and 10 yielded isomeric mixtures of D-gluco
ence of Ag2CO3 to give the β-methyl glycoside 9 in 91% yield.37 and L-ido configured products (Table 1). The reductions for all
Bromide 14 was also converted into the β-glycosyl fluoride 12 the C-5 bromides were carried out at a concentration of
in 94% yield via treatment with silver fluoride in dry aceto- 60–80 mM in benzene at reflux with 0.1 mol% AIBN unless
nitrile.32 Free radical bromination of the D-glucuronic acid otherwise stated. Reduction of bromide 1 gave a 1 : 1.8 ratio of
derivatives (3, 6, 9 and 12) using NBS and UV light in refluxing L-idopyranuronate 2 to D-glucopyranuronate 3 in good overall
CCl4 as described by Ferrier and Furneaux,33 gave the known yield (70%). This was significantly worse than the 2 : 1 ratio
C-5 bromides 1,30,33 4,29,30 728 and 1031,32 in acceptable yield reported by Wong and coworkers30 but similar to the 1 : 2.3
(39–75%). Treatment of the α-fluoride 15 under the same con- ratio obtained by Chiba and Sinaÿ in toluene at reflux.29 When
ditions, however, resulted in a complex mixture from which no the reaction was carried out in toluene the overall yield was
desired product could be isolated. The radical bromination similar but the product ratio worsened to only 1 : 5. The
reaction is a homolytic process in which the methoxycarbonyl decreased fraction of L-ido isomer in refluxing toluene is un-
group at C-5 provides stabilisation of the intermediate radical expected. It is possible that the higher temperature accelerates
leading to selective bromination of C-5.34 On the whole the C-5 the formation of both isomers, but also accelerates degra-
bromides gave satisfactory analytical and spectroscopic data. dation of the L-ido isomer, leading to an apparent decrease in
Bromide 1 gave a low m.p. (103–105 °C) compared with that the proportion of L-ido product isolated. Reduction of the
reported by Ferrier and Furneaux33 (159–161 °C), however, the α-anomer 4 gave an isomeric ratio of 3.7 : 1 of L-idopyranuro-
1H NMR spectrum was in accord with that reported and the nate 5 to D-glucopyranuronate 6, which was a slight improve-
13C NMR and mass spectra were consistent with the proposed ment over the 3.3 : 1 ratio previously reported.29,30 However,
1
structure. The H NMR spectrum of bromide 7 was also not in the overall yield was 78% compared with 87–91%. It is note-
complete agreement with the (90 MHz) spectrum reported by worthy that the 13C NMR data reported in the literature for 2
Chiba and Sinaÿ.28 However, the 1H and 13C NMR and mass by Whitfield et al.38 matches that obtained for the β-L-anomer
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 1 Reduction of C-5 bromides with tributyltin hydride in benzene (or toluene)
This study
Radical initiator
AIBNa
Et3Bb,c
Literatured
C-5 bromide
L-ido : D-gluco
% L-ido (% D-gluco)
L-ido : D-gluco
L-ido : D-gluco
% L-ido (% D-gluco)
Ref.
1
1 : 1.8
1 : 5
17% (53%)
14% (69%)
62% (16%)
2 : 1e
1 : 2.3
3.3 : 1
3.3 : 1e
53% (27%)
27% (63%)
67% (20%)
70% (21%)
30
28
29
30
1 (toluene)
4
4 (toluene)
7
7 (toluene)
10
10 (toluene)
1 : 3.3
3.4 : 1
1 : 1.8
1 : 0
3.7 : 1c
1 : 1.1
1 : 0
43% (48%)
65%
1 : 1.2
1 : 0
2.6 : 1
38% (44%)
64%
67% (25%)
28
32
31
a Reactions were performed under nitrogen or argon in benzene or toluene at reflux at a concentration of 60–80 mM. b Reactions were performed
under argon in toluene at r.t. at a concentration of 40 mM. c Product ratios were determined by 1H NMR analysis of crude product mixtures.
d Except where indicated, reactions were conducted in benzene or toluene at reflux at various concentrations (up to 190 mM) without the use of a
radical initiator. e AIBN used as a radical initator.
5 obtained here and also reported by Wong and co-workers,30 To begin, calculations were performed on a model system con-
although in the latter case the product from the reduction of 4 sisting of the reduction of pyranosyl radical A (Fig. 1a) by
was mislabelled as being the “α-L-idopyranuronate.”
Me3SnH. Radical A lacks the OAc groups of 10 but contains the
Reduction of β-methyl glucopyranuronate 7 resulted in a β-F substituent and the ester group. Transition states for
similar (1 : 1.1) isomeric ratio of products as reported by Chiba hydrogen atom transfer to A from Me3SnH were computed.
and Sinaÿ while reduction of the fluoride 10 gave exclusively The computational procedure involved geometry optimizations
L-ido isomer 11 in 65% yield, as also reported by Gelb and co- in implicit benzene at the B3LYP/6-31G(d)-LANL2DZ-SMD
workers.32 No trace of the D-gluco isomer was detected by (benzene) level of theory, followed by single-point energy calcu-
1H NMR spectroscopy. Overall, the isomeric ratios obtained lations with B3LYP-D3(BJ)/Def2-TZVPP-SMD(benzene). Based
were similar to the literature, and show a dependence upon on a preliminary conformational search of bromide 10, six
the nature and configuration of the substituent at the anome- transition state geometries were considered, as shown in
ric centre. The isomer ratios reported in Table 1, including Fig. 1b. In three of the transition states (TS1–TS3), Me3SnH
both our own data and those from previously published work, delivers a hydrogen atom to the β face of A, leading to L-ido
are calculated based on the yields of isolated products. Owing configured product B. In the other three transition states (TS4–
to possible losses during workup, these ratios are not necess- TS6), Me3SnH delivers a hydrogen atom to the α face of A
arily equivalent to the relative rates at which the L-ido and leading to the D-gluco configured product C. Transition states
D-gluco products are generated. To examine the stabilities of TS1 and TS4 have a 4C1 pyranose chair conformation, with
the reduced products, NMR spectra were recorded of the crude fluorine equatorial. TS2 and TS5 have a 1C4 chair confor-
2
1
reaction mixtures following reductions of bromides 1 and 10. mation, with fluorine axial. TS3 and TS6 are S0 and S5 skew-
For 1, the ratio of 2 to 3 in the crude mixture was 1 : 2.6, which boats, respectively.
is lower than the final ratio based on isolated yields (1 : 1.8).
The calculated free energy barriers of TS1–TS6 span a range
This suggests that some loss of 3 occurs during workup. On of 4.4 kcal mol−1. The lowest-energy transition state is TS2. Its
the other hand, the crude NMR spectrum of the reduction of structure is shown in Fig. 1c. TS2 features a 1C4 chair pyranose
10 showed no trace of D-gluco product 12, suggesting that ring, an axial β-F substituent, and an axial approach trajectory
either the D-gluco product 12 degraded before the spectrum for the delivery of the hydrogen atom to C-5 by the stannane.
could be measured, or the reaction is completely selective for The conformation of the pyranose ring in TS2 is similar to that
L-ido isomer 11. The reductions of all four bromide substrates in the most stable conformer of radical A (Fig. 1a), which is
were also performed in toluene at room temperature using the stabilized by negative hyperconjugation between the axial lone
low temperature radical initiator Et3B.39 The reactions pro- pair on oxygen and the C–F σ* orbital. The reaction of Me3SnH
ceeded cleanly and rapidly (complete after only 30 minutes) with this radical occurs preferentially via an axial trajectory
and gave improved yields of crude products (72–84%). (TS2) rather than an equatorial trajectory (TS5). This prefer-
1
However, H NMR analysis showed that the ratios of products ence may be attributed to two factors, which are depicted
were similar to, or slightly worse, than those obtained in schematically in Fig. 2. The first factor is a stereoelectronic
benzene at reflux (Table 1).
effect recently termed the ‘transition state gauche effect’ by
Density functional theory calculations were carried out to Houk, Hsung and co-workers.40 This effect depends on hyper-
explore the stereoselectivity of the reduction of β-fluoride 10. conjugative interactions between the σ* orbital associated with
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 1 (a) Reductions of β-F-substituted pyranosyl radical A by Me3SnH, leading to L-ido product B or D-gluco product C. (b) Relative energies of
different conformers of the transition states for the reaction of radical A with Me3SnH in benzene, computed with B3LYP-D3(BJ)/Def2-TZVPP-SMD//
B3LYP/6-31G(d)-LANL2DZ-SMD. (c) Geometry of the favoured transition state TS2. Bond lengths in Å; ΔH‡rel and ΔGr‡el in kcal mol−1
.
stereoelectronic factors that determine the transition-state
energies of this reaction. Transition-state calculations for the
reaction of the C-5 radical derived from 10 with Me3SnH are
shown in Fig. 3. Transition states TS1′–TS6′ have the same
pyranose conformations and Me3SnH attack trajectories as
model transition states TS1–TS6, respectively, except for tran-
sition state TS6′, which has a 1S3 skew-boat pyranose ring as
opposed to the 1S5 conformation of TS6.
The most obvious difference between the reactions of the
model and acetylated radicals is that, for the acetylated
radical, no transition state analogous to TS2 (which was the
lowest-energy TS in the model system) could be found. Such a
transition state is disfavored because the F substituent,
C3-acetyl group, and incoming stannane occupy
a very
Fig. 2 Factors affecting the relative energies of axial and equatorial
approach of Me3SnH to radical A.
crowded 1,3,5-triaxial arrangement. Attempts to locate TS2′ on
the potential energy surface led to collapse to the reactants.
Interestingly, it proved possible to locate the D-gluco config-
ured transition state TS5′, which is also 1,3,5-triaxial but
slightly less crowded. The transition states TS1′, TS5′, and
TS6′, are high in energy (≥2.4 kcal mol−1) and do not contri-
bute significantly to the reaction. Only transition states TS3′
and TS4′ are low enough in energy to make significant con-
tributions to product formation.
the newly-forming C–H bond, and groups on adjacent atoms.
The C–H σ* orbital acts as a σ acceptor. In the axial TS2, the
forming C–H bond is antiperiplanar to one of the oxygen lone
pairs, which is an electron donor, and this arrangement is
stabilized by hyperconjugation. In contrast, in equatorial TS5
the forming C–H bond is antiperiplanar to the O–C bond,
which is a poorer σ donor than an oxygen lone pair. This
arrangement is less stabilized by hyperconjugation. The second
factor contributing to the stability of TS2 is a non-covalent inter-
action between Sn and F. These two atoms lie 3.39 Å apart in
TS2, well within the sum of the van der Waals radii of Sn and F
(3.64 Å),41 whereas in TS5 there is no Sn–F interaction (>6 Å).
Compared with TS2, the other transition states for the reac-
tion of radical A with Me3SnH are all at least 2 kcal mol−1
higher in energy. At the temperature of refluxing benzene
(80 °C), TS2, which leads to L-ido configured product B, would
be expected to account for >92% of product formation.
Favored transition state TS3′ (L-ido) is 1.0 kcal mol−1 lower
in energy than TS4′ (D-gluco). Based on the calculated ΔG‡
values of all five transition states, the predicted isomer ratio
(11 : 12) resulting from the reduction of bromide 10 at 80 °C is
77 : 23. This is lower than the experimental selectivity (100 : 0,
Table 1). Some of the discrepancy between theory and experi-
ment may stem from inaccuracies in the energies of transition-
state conformers resulting from the use of Me3SnH as a model
for Bu3SnH. Furthermore, as described above, the possibility
that some 12 is formed but degrades prior to workup cannot
be ruled out. Nonetheless, the calculations do correctly predict
that the reduction of 10 favors the L-ido product.
Although the calculations on the model radical A match the
selectivity observed experimentally in the reduction of 10,
which favors the L-ido product, conformational effects associ-
ated with the acetyl groups of 10 give rise to a different set of
1
Transition state TS3′ features a S3 skew-boat conformation
of the pyranose ring, while second-lowest transition state TS4′
is a 4C1 chair. In the absence of Me3SnH, these two radical
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 3 (a) Relative energies of different conformers of the transition states for the reaction of Me3SnH with the radical derived from 10 in benzene,
leading to L-ido product 11 or D-gluco product 12. (c) Geometry of the two lowest-energy transition states, TS3’ and TS4’. Bond lengths in Å; ΔH‡rel
and ΔG‡rel in kcal mol−1
.
conformations are almost equienergetic. Both TSs contain
antiperiplanar arrangements between the C–H forming bond
and oxygen lone pair orbitals, allowing hyperconjugative
Experimental section
General methods
stabilization. The lower energy of TS3′, relative to TS4′, reflects Melting points were determined on a DigiMelt MSRS appar-
a stronger interaction between the radical and the stannane. atus. Optical rotations were determined on a JASCO P-2000
While the forming and breaking bond lengths are the same in polarimeter at ambient temperature and are given in units of
each TS, TS3′ derives additional stabilization from a Sn–F 10−1 deg cm2 g−1. 1H and 13C NMR spectra were recorded on a
interaction (3.65 Å), whereas no such interaction is present Bruker Avance 500 MHz spectrometer at 20 °C. The residual
in TS4′.
solvent peaks (CDCl3: δH 7.24 and δC 77.0) served as internal
standard. Coupling constants in Hz were measured from one-
dimensional spectra. Low resolution mass spectra were
acquired on
a Bruker HCT 3D mass spectrometer. All
Conclusions
chemicals were purchased as reagent grade and used without
further purification. Carbon tetrachloride, hexane, and EtOAc
were distilled before use. Reactions were monitored by ana-
lytical thin layer chromatography (TLC) on silica gel 60 F254
plates and visualized by charring with 10% sulfuric acid in
ethanol.
In conclusion, a series of D-glucuronides bearing different sub-
stituents at the anomeric centre were photobrominated to give
the respective 5-C-bromo-D-glucuronides. The bromides were
subjected to tributyltin hydride-mediated radical reduction to
give a mixture of L-ido and D-gluco configured products. All
final compounds and synthetic intermediates were fully
characterized by NMR spectroscopy. In keeping with previous
literature reports, it was found that the diastereomeric product
ratio was dependent on the reaction temperature (reactions
conducted in refluxing benzene versus toluene) and, in particu-
lar, upon the nature and configuration of the anomeric substi-
tuent. Particularly noteworthy was the reduction of the 5-C-
bromo-D-glucuronyl β-fluoride 10 which gave exclusively the
L-ido product, making this an attractive small-scale route to
IdoA building blocks. DFT calculations show that the L-ido
selectivity obtained with the β-F derivative originates from a
combination of a transition state gauche effect and an Sn–F
interaction.
Methyl 1,2,3,4-tetra-O-acetyl-D-glucopyranuronate (3, 6). To
a
stirred solution of D-glucurono-6,3-lactone 13 (17.0 g,
78 mmol) in MeOH (50 mL) was added a catalytic amount of
sodium methoxide (50 mg). The mixture was stirred at r.t. for
1 h and then concentrated under reduced pressure. The
residue was dissolved in acetic anhydride (50 mL) and a
mixture of HClO4 (1.0 mL) in acetic anhydride (2 mL) was
added dropwise such that the reaction temperature did not
rise above 40 °C. The mixture was then stirred at r.t. for 24 h
and then kept at 5 °C for 24 h leading to separation of
crystalline material. The crystalline material was isolated by
filtration, washed with ether and recrystallized from ethanol to
give pure methyl 1,2,3,4-tetra-O-acetyl-β-D-glucopyranuronate 3
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
(3.0 g, 10%) as colourless crystals, m.p. 176–177 °C (lit.28,42 75 °C. Subsequently silver carbonate (0.50 g, 1.8 mmol) was
177–178 °C; lit.35 176.5–178 °C). [α]D +7.6 (c 1.0, CHCl3; lit.35 added in three portions over a period of 3 h. The reaction
1
+7.4, lit.42 +8.6). H and 13C NMR spectra were in accord with mixture was stirred at this temperature for 18 h. The reaction
the literature.42,43 1H NMR (500 MHz, CDCl3): δ 5.74 (d, 1H, mixture was then cooled to r.t., filtered and the filtrate was
J1,2 = 7.7 Hz, H-1), 5.29 (dd, 1H, J2,3 = 9.1 Hz, J3,4 = 9.5 Hz, concentrated under reduced pressure. Flash chromatography
H-3), 5.22 (dd, 1H, J3,4 = J4,5 = 9.5 Hz, H-4), 5.12 (dd, 1H, H-2), (EtOAc/hexane 1 : 1) of the crude mixture gave the methyl
4.15 (d, 1H, H-5), 3.74 (s, 3H, OCH3), 2.15 (s, 3H, CH3), 2.02 (s, glycoside 9 (0.41 g, 91%), m.p. 154–155 °C (hexane/EtOAc; lit.37
6H, 2 × CH3), 2.01 (s, 3H, CH3); 13C NMR (125 MHz, CDCl3): δ 154.1–154.4 °C; lit.49 151–152 °C; lit.50 148–150 °C). [α]D −31.3
1
169.9, 169.4, 169.1, 168.8, 166.8 (5 × CvO), 91.4 (C-1), 73.0 (c 0.3, CHCl3; lit.49 −28.1). The H and 13C NMR spectra were
(C-5), 71.8 (C-4), 70.2 (C-2), 68.9 (C-3), 53.0 (OCH3), 20.7, 20.5, in accord with the literature.37,51 1H NMR (500 MHz, CDCl3):
20.5, 20.4 (4 × CH3); ESMS: m/z 399.1 [M + Na]+. The mother δ 5.26–5.18 (m, 2H, H-3,4), 5.00–4.96 (m, 1H, H-2), 4.46 (d, 1H,
liquors were cooled to 0 °C, MeOH (10 mL) was added and the J1,2 = 7.7 Hz, H-1), 4.04–4.00 (m, 1H, H-5), 3.74 (s, 3H, OCH3),
mixture concentrated under reduced pressure (×2 with 3.50 (s, 3H, OCH3), 2.03 (s, 3H, CH3), 2.00 (s, 3H, CH3), 1.99 (s,
toluene). The residue was dissolved in EtOAc, washed with 3H, CH3); 13C NMR (125 MHz, CDCl3): δ 170.1, 169.5, 169.2,
water (×1) and saturated aq. NaHCO3 (×1), decolourized, dried 167.3 (4 × CvO), 101.8 (C-1), 72.6 (C-5), 72.1 (C-2), 71.2 (C-3),
(MgSO4) and concentrated. The syrup obtained (8.0 g) was dis- 69.5 (C-4), 57.3 (OCH3), 52.9 (OCH3), 20.7, 20.6, 20.5 (3 × CH3).
solved in hot isopropyl alcohol. Upon cooling a semi solid ESMS: m/z 386.9 [M + Na]+.
material separated out. The liquid layer was then decanted
Methyl
(2,3,4-tri-O-acetyl-α-D-glucopyranosyl
fluoride)
and the semi solid was purified by flash chromatography uronate (15). The peracetate 3 (0.40 g, 1.1 mmol) was dis-
(hexane/EtOAc 3 : 1) to give methyl 1,2,3,4-tetra-O-acetyl-α-D- solved in 30% HF/pyridine (10 mL) at −20 °C under an argon
glucopyranuronate 6 as colourless needles, (1.0 g, 3%), m. atmosphere. The mixture was then allowed to warm up to r.t.
p. 109–115 °C (EtOAc/hexane; lit.44 114–114.5 °C; lit.45 and stirred for 18 h. The reaction mixture was then diluted
108–120 °C). [α]D +89.0 (c 0.7, CHCl3; lit.45 +92.6; lit.46 +98.0). with DCM (20 mL) and washed with water, saturated aq.
The 1H NMR spectrum was in accord with the literature.29,47 NaHCO3, dried (MgSO4) and concentrated under reduced
1H NMR (500 MHz, CDCl3): δ 6.40 (d, 1H, J1,2 = 3.7 Hz, H-1), pressure. The crude material was then purified by flash
5.52 (dd, 1H, J2,3 = 10.3 Hz, J3,4 = 9.6 Hz, H-3), 5.22 (dd, 1H, J4,5 column chromatography (hexane/EtOAc 1 : 1) to give the fluor-
= 10.3 Hz, H-4), 5.12 (dd, H-2), 4.41 (d, 1H, H-5), 3.75 (s, 3H, ide 15 (70 mg, 19%) as colourless oil. The 1H and 13C NMR
OCH3), 2.17 (s, 3H, CH3), 2.05 (s, 3H, CH3), 2.04 (s, 3H, CH3), spectra were in accord with the literature.36 1H NMR
2.02 (s, 3H, CH3); 13C NMR (125 MHz, CDCl3): δ 170.0, 169.5, (500 MHz, CDCl3): δ 5.79 (dd, 1H, J1,2 = 2.6 Hz, J1,F = 52.5 Hz,
169.4, 168.5, 167.2 (5 × CvO), 88.8 (C-1), 70.4 (C-5), 69.1 (C-4), H-1), 5.53 (dd, 1H, J2,3 = J3,4 = 9.9 Hz, H-3), 5.21 (dd, 1H, J4,5
=
69.0 (C-2), 68.9 (C-3), 53.0 (OCH3), 20.8, 20.6, 20.5, 20.4 (4 × 9.9, H-4), 4.95 (ddd, J2,F = 23.9 Hz, H-2), 4.44 (d, 1H, H-5), 3.74
CH3); ESMS: m/z 399.1 [M + Na]+.
Methyl (2,3,4-tri-O-acetyl-α-D-glucopyranosyl
(s, 3H, OCH3), 2.08 (s, 3H, CH3), 2.02 (s, 3H, CH3), 2.02 (s, 3H,
bromide) CH3); 13C NMR (125 MHz, CDCl3): δ 169.9, 169.8, 169.4, 167.0
uronate (14). The tetracetate 3 (2.0 g, 5.3 mmol) was sus- (C-6), 103.5 (d, J1,F = 231.4 Hz, C-1), 69.9 (d, J2,F = 24.0 Hz, C-2),
pended at 0 °C in 33% HBr in acetic acid (27 mL) under a N2 69.8 (d, J5,F = 4.3 Hz, C-5), 53.1 (OCH3), 20.4, 20.5, 20.6
atmosphere. After stirring for 15 min at 0 °C, the reaction (3 × CH3); ESMS: m/z 359.0 [M + Na]+.
mixture was allowed to warm up to r.t. and stirred for a
Methyl
(2,3,4-tri-O-acetyl-β-D-glucopyranosyl
fluoride)
further 2 h. The acetic acid was then removed by evaporation uronate (12). To a solution of the bromide 14 (0.50 g,
with toluene under reduced pressure. The crude oil was 1.3 mmol) in dry acetonitrile (20 mL) under a N2 atmosphere
diluted with EtOAc (50 mL) and washed with cold sat. NaHCO3 was added silver fluoride (0.60 g, 4.8 mmol). The reaction
(4 × 20 mL) until bubbling ceased. The organic layer was dried mixture was then stirred overnight in the dark at r.t. The reac-
(MgSO4) and concentrated under reduced pressure to give the tion mixture was then filtered and concentrated under reduced
bromide 14 (1.3 g, 62%) as an oil. The crude material was used pressure. The crude material was purified by flash chromato-
1
in the next step without further purification. The H and 13C graphy (hexane/EtOAc 2 : 1) to give the fluoride 12 (0.40 g,
NMR spectra were in accord with the literature.48 1H NMR 94%) as a solid, m.p. 104–107 °C (EtOAc/hexane; lit.52
1
(500 MHz, CDCl3): δ 6.62 (d, 1H, J1,2 = 4.0 Hz, H-1), 5.59 (dd, 108–109 °C). [α]D +13.3 (c 1.2, CHCl3; lit.52 +18). The H NMR
1H, J2,3 = 9.7 Hz, J3,4 = 9.5 Hz, H-3), 5.22 (dd, J4,5 = 10.3 Hz, 1H, spectrum was in accord with the literature.52,53 1H NMR
H-4), 4.84 (dd, 1H, H-2), 4.56 (d, 1H, H-5), 3.75 (s, 3H, OCH3), (500 MHz; CDCl3): δ 5.42 (dd, 1H, J1,2 = 5.1 Hz, J1,F = 51.0 Hz,
2.08 (s, 3H, CH3), 2.04 (s, 3H, CH3), 2.03 (s, 3H, CH3); 13C NMR H-1), 5.40 (ddd, 1H, J3,4 = J4,5 = 8.8 Hz, J4,F = 0.7 Hz, H-4),
(125 MHz, CDCl3): δ 169.6, 169.6, 169.4, 166.6 (4 × CvO), 85.3 5.26–5.18 (m, 1H, H-3), 5.08 (ddd, J2,3 = 7.6 Hz, J2,F = 22.2 Hz
(C-1), 72.0 (C-5), 70.3 (C-2), 69.2 (C-3), 68.4 (C-4), 53.1 (OCH3), 1H, H-2), 4.45 (d, 1H, H-5), 3.76 (s, 3H, OCH3), 2.08 (s, 3H,
20.5, 20.4 (3 × CH3).
Methyl (methyl
CH3), 2.03 (2s, 6H, 2 × CH3); 13C NMR (125 MHz, CDCl3):
2,3,4-tri-O-acetyl-β-D-glucopyranosid) δ 169.8, 169.3, 169.1, 166.9 (4 × CvO), 105.8 (d, J1,F = 223.8 Hz,
uronate (9). To a stirred solution of the bromide 14 (0.50 g, C-1), 72.2, (d, J5,F = 3.5 Hz, C-5), 70.6 (d, J2,F = 31.0 Hz, C-2),
1.3 mmol) in dry toluene (10 mL) under an Ar atmosphere, dry 70.4 (d, J3,F = 6.7 Hz, C-3), 67.9 (C-4), 53.1 (OCH3), 20.6, 20.5,
MeOH (1 mL) was added and the solution was warmed to 20.5 (3 × CH3); ESMS: m/z 338.9 [M + Na]+.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
General procedure for the radical bromination reaction
J2,F = 23.9 Hz, C-2), 69.9 (d, J3,F = 10.6 Hz C-3), 69.4 (C-4),
54.5 (OCH3), 20.7, 20.6, 20.5 (3 × CH3); ESMS: m/z 415.8.8
A suspension of glycosyl uronate (1 eq.) and N-bromosuccin- [M + Na]+.
amide (2 eq.) in dry CCl4 (10 mL per 1.0 mmol) was heated at
reflux for 4 h under irradiation with a 500 W tungsten lamp.
The reaction mixture was then cooled to r.t. and filtered over
General procedure for AIBN-initiated tributyltin hydride
reduction
celite, and concentrated under reduced pressure. The pure The D-glucuronic acid 5-C-bromide (1 eq.) was dissolved in
bromide was isolated by flash chromatography (hexane : anhydrous benzene (60–80 mM) and stirred under nitrogen.
EtOAc : toluene 1 : 1 : 0.1).
Tributyltin hydride (1.5 eq.) and AIBN (0.1 mol%) were added,
Methyl 1,2,3,4-tetra-O-acetyl-5-C-bromo-β-D-glucopyranosyl- and the reaction mixture was heated at reflux for 2 h. The
uronate (1). Bromination of uronate 3 (0.47 g, 1.2 mmol) via mixture was cooled to r.t., and the solvent was removed under
the general procedure gave the bromide 1 (0.41 g, 75%) as reduced pressure. The residue was purified by flash chromato-
colourless needles, m.p. 103–105 °C (EtOAc/hexane; lit.33 graphy on silica gel (EtOAc : hexane : toluene 1 : 1 : 0.5) to give
159–161 °C). The 1H NMR spectrum was in accord with the the product(s).
literature.30,33 1H NMR (500 MHz, CDCl3): δ 6.26 (d, 1H, J1,2
=
Methyl 1,2,3,4-tetra-O-acetyl-α-L-idopyranuronate (2). The
8.8 Hz, H-1), 5.52 (dd, 1H, J2,3 = J3,4 9.5 Hz, H-3), 5.30 (d, 1H, bromide 1 (0.38 g, 0.8 mmol) was reduced with tributyltin
H-4), 5.22 (dd, H-2), 3.81 (s, 3H, OCH3), 2.11 (s, 3H, CH3), 2.08 hydride according to the general procedure to give 2 (55 mg,
(s, 3H, CH3), 2.03 (s, 3H, CH3), 2.00 (s, 3H, CH3); 13C NMR 17%) and 3 (100 mg, 53%). Data for 2: m.p. 125–126 °C
(125 MHz, CDCl3): δ 169.6, 169.1, 169.0, 168.2, 164.2 (5 × (EtOAc/hexane; lit.28 118–119 °C). [α]D −60.7 (c 0.5, CHCl3).
CvO), 90.8 (C-1), 89.0 (C-5), 70.7 (C-4), 69.8 (C-2), 69.2 (C-3), The 1H and 13C NMR spectra were in accord with the litera-
54.1 (OCH3), 20.6, 20.6, 20.5, 20.5 (4 × CH3); ESMS: m/z 476.9 ture.28,30,38 1H NMR (500 MHz, CDCl3): δ 6.27–6.25 (m, 1H,
[M + Na]+.
H-1), 5.16–5.11 (m, 2H, H-3, H-4), 4.84 (d, J4,5 = 2.6 Hz, H-5),
Methyl 1,2,3,4-tetra-O-acetyl-5-C-bromo-α-D-glucopyranosyl- 4.83 (ddd, 1H, J1,2 = 0.9, J2,3 = 2.2, J2,4 = 3.3 Hz, H-2), 3.78 (s,
uronate (4). Bromination of uronate 6 (0.53 g, 1.4 mmol) via 3H, OCH3), 2.10 (s, 3H, CH3), 2.09 (s, 3H, CH3), 2.08 (s, 3H,
the general procedure gave the bromide 4 (0.25 g, 39%) as a CH3), 2.06 (s, 3H, CH3); 13C NMR (125 MHz, CDCl3): δ 169.3,
colourless oil. 1H and 13C NMR spectra were in accord with the 169.0, 168.6, 168.0, 167.6 (5 × CvO), 90.4 (C-1), 68.2 (C-5), 66.7
literature.29,30 1H NMR (500 MHz, CDCl3): δ 6.51 (d, 1H, J1,2
4.1 Hz, H-1), 5.78 (dd, 1H, J2,3 = 10.1 Hz, J3,4 = 9.8 Hz, H-3), (4 × CH3). ESMS: m/z 399.1 [M + Na]+.
=
(C-3), 66.2 (C-4), 65.9 (C-2), 52.7 (OCH3), 20.8, 20.7, 20.6, 20.5
5.23 (d, 1H, H-4), 5.17 (dd, H-2), 3.80 (s, 3H, OCH3), 2.19 (s,
Methyl 1,2,3,4-tetra-O-acetyl-β-L-idopyranuronate (5). The
3H, CH3), 2.09 (s, 3H, CH3), 2.03 (s, 3H, CH3), 2.02 (s, 3H, bromide 4 (100 mg, 0.22 mmol) was reduced with tributyltin
CH3); 13C NMR (125 MHz, CDCl3): δ 169.5, 169.4, 168.9, 168.5, hydride according to the general procedure to give a mixture of
164.8 (5 × CvO), 89.5 (C-1), 87.1 (C-5), 69.8 (C-3), 67.9 (C-4), 5 and 6 (600 mg, 78%) in a ratio of 3.7 : 1, as determined by 1H
67.0 (C-2), 54.1 (OCH3), 21.2, 20.6, 20.5, 20.3 (4 × CH3); ESMS: NMR spectroscopy. Purification by flash chromatography pro-
m/z 476.9 [M + Na]+.
vided analytical data for 5: colourless oil. [α]D +8.1 (c 0.3,
Methyl (methyl 2,3,4-tri-O-acetyl-5-C-bromo-β-D-glucopyrano- CHCl3; lit.29 +12.3). The 1H and 13C NMR spectra were in
sid)uronate (7). Bromination of uronate 9 (0.41 g, 1.2 mmol) accord with the literature.29,30 1H NMR (500 MHz, CDCl3):
via the general procedure gave the bromide 7 (0.25 g, 49%) as δ 6.03 (d, 1H, J1,2 = 1.9 Hz, H-1), 5.26 (dd, 1H, H-3), 5.08 (ddd,
a colourless oil.28 1H NMR (500 MHz, CDCl3): δ 5.46 (dd, 1H, J = 1.0, 4.0, H-4), 4.99 (ddd, 1H, J = 2.5 Hz, H-2), 4.69 (d, 1H,
J2,3 = J3,4 = 9.5 Hz, H-3), 5.27 (d, 1H, H-4), 5.10 (dd, 1H, J1,2
=
J = 3.5, Hz, H-5), 3.78 (s, 3H, OCH3), 2.12 (s, 3H, CH3), 2.11 (s,
8.3 Hz, H-2), 5.00 (d, H-1), 3.82 (s, 3H, OCH3), 3.53 (s, 3H, 3H, CH3), 2.10 (s, 3H, CH3), 2.09 (s, 3H, CH3); 13C NMR
OCH3), 2.06 (s, 3H, CH3), 2.04 (s, 3H, CH3), 1.98 (s, 3H, CH3); (125 MHz, CDCl3): δ 169.4, 169.2, 168.6, 168.1, 166.8 (5 ×
13C NMR (125 MHz, CDCl3): δ 169.7, 169.2, 168.9, 164.6 (4 × CvO), 89.8 (C-1), 73.3 (C-2), 67.2 (C-5), 66.3 (C-3), 65.3 (C-4),
CvO), 101.6 (C-1), 90.1 (C-5), 70.8 (C-3), 70.1 (C-4), 70.0 (C-2), 52.7 (OCH3), 20.7(4), 20.7(3), 20.7(0), 20.5 (4 × CH3); ESMS: m/z
57.8 (OCH3), 54.0 (OCH3), 20.6, 20.6, 20.5 (3 × CH3); ESMS: m/z 399.1 [M + Na]+.
448.9 [M + Na]+.
Methyl (methyl 2,3,4-tri-O-acetyl-α-L-idopyranoside) uronate
Methyl
(2,3,4-tri-O-acetyl-5-C-bromo-β-D-glucopyranosyl (8). The bromide 7 (30 mg, 0.07 mmol) was reduced with tri-
fluoride) uronate (10). Bromination of uronate 12 (0.37 g, butyltin hydride according to the general procedure to give 8
1.1 mmol) via the general procedure gave the bromide 10 (10 mg, 43%) and 9 (11 mg, 48%). Data for 8: colourless oil.
(0.20 g, 44%) as a colourless oil.31,32 The 13C NMR spectrum [α]D −43.1 (c 0.2, CHCl3; lit.28 −56.5). The 1H and 13C NMR
was in accord with the literature.31 1H NMR (500 MHz, CDCl3): spectra were in accord with the literature.28,38 1H NMR
δ 5.70 (dd, 1H, J1,F 51.2 Hz, J1,2 7.3 Hz, H-1), 5.47 (ddd, 1H, (500 MHz, CDCl3): δ 5.11 (br ddd, 1H, J3,4 = 3.1 Hz, J4,5
=
J3,4 = 9.5 Hz, J2,3 = 10.3 Hz, J3,F = 0.7 Hz, H-3), 5.35 (d, 1H, H-4), 2.6 Hz, H-4), 5.03 (dd, 1H, J3,4 = 3.1 Hz, J2,3 = 4.4 Hz, H-3), 4.88
5.22 (ddd, 1H, J2,F = 23.2 Hz, H-2), 3.85 (s, 3H, OCH3), 2.09 (s, (dd, J1,2 = 1.8 Hz, J1,3 = 1.0 Hz 1H, H-1), 4.80 (d, 1H, J4,5
=
3H, CH3), 2.08 (s, 3H, CH3), 2.00 (s, 3H, CH3); 13C NMR 2.6 Hz, H-5), 4.78–4.76 (m, 1H, H-2), 3.77 (s, 3H, OCH3), 3.42
(125 MHz; CDCl3): δ 169.6, 169.0, 168.8, 163.8 (4 × CvO), (s, 3H, OCH3), 2.09 (s, 3H, CH3), 2.07 (s, 3H, CH3), 2.06 (s, 3H,
106.3 (d, J1,F = 225 Hz, C-1), 87.7 (d, J5,F = 8.0 Hz, C-5), 70.0 (d, CH3); 13C NMR (125 MHz; CDCl3): δ 169.5, 169.4, 169.0, 168.4
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
(4 × CvO), 99.2 (C-1), 67.2 (C-4), 66.9 (C-3), 66.7 (C-2), 66.3 basis set consisting of LANL2DZ on Sn and 6-31G(d) on all
(C-5), 56.3 (OCH3), 52.6 (OCH3), 20.8, 20.6 (3 × CH3); ESMS: other atoms. The optimizations were conducted in implicit
m/z 371.0 [M + Na]+.
benzene, as modeled with the SMD implicit solvent model.62
Methyl (2,3,4-tri-O-acetyl-α-L-idopyranosylfluoride) uronate Harmonic vibrational frequency calculations were performed
(11). The bromide 10 (0.19 g, 0.4 mmol) was reduced with to confirm whether stationary points were ground states (zero
tributyltin hydride according to the general procedure to give imaginary frequencies) or transition states (one imaginary fre-
11 (100 mg, 65%) as a colourless oil. [α]D −37.1 (c 0.1, CHCl3; quency), and also to compute zero-point energies and thermo-
lit.31 −39) The 1H and 13C NMR spectra were in accord with chemical corrections. Errors in computed entropies,
the literature.31 1H NMR (500 MHz, CDCl3): δ 5.67 (d, 1H, J1,F
=
introduced by the treatment of low frequency modes as harmo-
47.6 Hz, H-1), 5.15–5.12 (m, 1H, H-4), 5.04 (br s, 1H, H-3), 4.91 nic motions, were minimized by use of Truhlar’s approxi-
(d, 1H, J4,5 = 2.0 Hz, H-5), 4.85–4.82 (m, 1H, H-2), 3.76 (s, 3H, mation,63 in which all harmonic frequencies below 100 cm−1
OCH3), 2.10 (s, 3H, CH3), 2.06 (s, 3H, CH3), 2.04 (s, 3H, CH3); were raised to exactly 100 cm−1 before evaluation of the
13C NMR (125 MHz; CDCl3): δ 169.2, 168.9, 168.6, 167.1 (4 × vibrational component of the thermal contribution to entropy.
CvO), 104.7 (d, J1,F = 226.6 Hz, C-1), 67.5 (d, J5,F = 2.5 Hz, C-5), Subsequently, single-point energy calculations were performed
65.8 (C-4), 65.1 (C-3), 64.2 (d, J2,F = 37.9 Hz, C-2), 52.7 (OCH3), at the B3LYP-D3(BJ)64,65/Def2-TZVPP level of theory, again
20.6, 20.5, 20.4 (3 × CH3); ESMS: m/z 359.0 [M + Na]+.
using SMD benzene. Free energies in solution were computed
by adding the B3LYP zero-point energy and thermochemical
corrections to the B3LYP-D3(BJ) solution-phase electronic ener-
gies. A standard state of 298.15 K and 1 mol L−1 was used.
General procedure for triethylborane initiated tributyltin
hydride reduction
The D-glucuronic acid 5-C-bromide (1 eq.) was dissolved in
anhydrous toluene (40 mM) and stirred under argon. Tributyl-
tin hydride (1.1 eq.) and triethylborane (1 M in hexanes,
0.1 eq.) was added, and the reaction mixture was stirred at r.t.
Acknowledgements
for 30 min. The reaction mixture was then concentrated under We thank the University of Queensland, the National MPS
reduced pressure. The crude residue was then dissolved in Society (USA) and the Australian Research Council Future
acetonitrile and washed with hexane (×3). The acetonitrile Fellowships scheme (FT120100632 to E. H. K.) for funding.
layer was then concentrated under reduced pressure and a Computational resources were provided by the Australian
1H NMR spectrum of the crude mixture was obtained.
National Computational Infrastructure National Facility and
Methyl 1,2,3,4-tetra-O-acetyl-α-L-idopyranuronate (2). The the University of Queensland Research Computing Centre.
bromide 1 (200 mg, 0.4 mmol) was reduced with tributyltin
hydride according to the general procedure to give a mixture of
2 and 3 (1 : 3.3, 120 mg, 82%) as a colourless oil.
Methyl 1,2,3,4-tetra-O-acetyl-β-L-idopyranuronate (5). The
References
bromide 4 (250 mg, 0.7 mmol) was reduced with tributyltin
hydride according to the general procedure to give a mixture of
5 and 6 (3.5 : 1, 210 mg, 84%) as a colourless oil.
1 B. Casu and U. Lindahl, Adv. Carbohydr. Chem. Biochem.,
2001, 57, 159–206.
2 D. L. Rabenstein, Nat. Prod. Rep., 2002, 19, 312–331.
3 B. Casu, A. Naggi and G. Torri, Carbohydr. Res., 2015, 403,
60–68.
Methyl (methyl 2,3,4-tri-O-acetyl-α-L-idopyranoside) uronate
(8). The bromide 7 (300 mg, 0.7 mmol) was reduced with tri-
butyltin hydride according to the general procedure to give a
mixture of 8 and 9 (1 : 1.8, 190 mg, 79%) as a colourless oil.
Methyl (2,3,4-tri-O-acetyl-α-L-idopyranosylfluoride) uronate
(11). The bromide 10 (100 mg, 0.2 mmol) was reduced with
tributyltin hydride according to the general procedure to give
11 (50 mg, 72%) as a colourless oil.
4 C. Malavaki, S. Mizumoto, N. Karamanos and K. Sugahara,
Connect. Tissue Res., 2008, 49, 133–139.
5 M. A. Thelin, B. Bartolini, J. Axelsson, R. Gustafsson,
E. Tykesson, E. Pera, A. Oldberg, M. Maccarana and
A. Malmstrom, FEBS J., 2013, 280, 2431–2446.
6 I. Capila and R. J. Linhardt, Angew. Chem., Int. Ed., 2002,
41, 391–412.
7 J. M. Whitelock and R. V. Iozzo, Chem. Rev., 2005, 105,
2745–2764.
8 D. Xu and J. D. Esko, Annu. Rev. Biochem., 2014, 83, 129–
157.
Theoretical calculations
Density functional theory calculations were performed using
Gaussian 09.54 Prior to performing geometry optimizations
with DFT, the conformations of bromide 10 were explored by
means of a conformational search in MacroModel 10.6,55,56
using the MCMM torsional sampling algorithm in conjunction
9 S. B. Dulaney and X. Huang, Adv. Carbohydr. Chem.
Biochem., 2012, 67, 95–136.
with the OPLS_2005 forcefield57 with a dielectric constant set 10 E. Grand, J. Kovensky, G. Pourceau, S. Toumieux and
to ε = 2.3 to mimic the dielectric constants of benzene (ε =
2.27) and toluene (ε = 2.37). DFT geometry optimizations of
A. Wadouachi, in Carbohydrate Chemistry, The Royal Society
of Chemistry, 2014, vol. 40, pp. 195–235.
transition states derived from the significant conformers were 11 M. M. L. Zulueta, S.-Y. Lin, Y.-P. Hu and S.-C. Hung, Curr.
then performed using the B3LYP functional58–61 and a mixed
Opin. Chem. Biol., 2013, 17, 1023–1029.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
12 M. M. L. Zulueta, S.-Y. Lin and S.-C. Hung, Trends Glycosci. 40 A. Patel, J. R. Vella, Z.-X. Ma, R. P. Hsung and K. N. Houk,
Glycotechnol., 2013, 25, 141–158.
J. Org. Chem., 2015, 80, 11888–11894.
13 H. Pellissier, Org. Prep. Proced. Int., 2002, 34, 441–465.
14 H. H. A. M. Hassan, Mini-Rev. Org. Chem., 2007, 4, 61–74.
15 D. D’Alonzo, A. Guaragna and G. Palumbo, Curr. Org.
Chem., 2009, 13, 71–98.
16 M. M. L. Zulueta, Y.-Q. Zhong and S.-C. Hung, Chem.
Commun., 2013, 49, 3275–3287.
17 S. Mohamed and V. Ferro, Adv. Carbohydr. Chem. Biochem.,
2015, 72, 21–61.
18 T. G. Frihed, M. Bols and C. M. Pedersen, Chem. Rev., 2015,
115, 3615–3676.
41 Computations provide estimates of the stabilization arising
from the transition-state gauche effect and the Sn–F inter-
action in TS2. Firstly, replacement of the pyranose ring
oxygen in TS2 and TS5 with a CH2 group leads to transition
states differing in energy by only 2.0 kcal mol−1, compared
to the 3.3 kcal mol−1 difference in energy between TS2 and
TS5 themselves. Secondly, replacement of the fluorine
atom in TS2 and TS5 with a hydrogen atom leads to a
0.5 kcal mol−1 reduction in the difference in energy
between TS2 and TS5.
19 J. C. Jacquinet, M. Pettitou, P. Duchaussoy, I. Lederman, 42 M. M. Cavalluzzi, G. Lentini, A. Lovece, C. Bruno,
J. Choay, G. Torri and P. Sinay, Carbohydr. Res., 1984, 130,
221–241.
A. Catalano, A. Carocci and C. Franchini, Tetrahedron Lett.,
2010, 51, 5265–5268.
20 C. A. A. van Boeckel, T. Beetz, J. N. Vos, A. J. M. de Jong, 43 Y. Y. Root, T. R. Wagner and P. Norris, Carbohydr. Res.,
S. F. Van Aelst, R. H. van den Bosch, J. M. R. Mertens and
F. A. van der Vlugt, J. Carbohydr. Chem., 1985, 4, 293–321.
21 R. Ojeda, J. L. de Paz, M. Martín-Lomas and
J. M. Lassaletta, Synlett, 1999, 1316–1318.
22 W. Ke, D. M. Whitfield, M. Gill, S. Larocque and S.-H. Yu,
Tetrahedron Lett., 2003, 44, 7767–7770.
23 J.-C. Lee, X.-A. Lu, S. S. Kulkarni, Y.-S. Wen and S.-C. Hung,
J. Am. Chem. Soc., 2004, 126, 476–477.
24 J. Tatai, G. Osztrovszky, M. Kajtár-Peredy and P. Fügedi,
Carbohydr. Res., 2008, 343, 596–606.
2002, 337, 2343–2346.
44 W. Stoecklin, E. Weiss and T. Reichstein, Helv. Chim. Acta,
1967, 50, 474–490.
45 F. Micheel and R. Habendorff, Chem. Ber., 1957, 90, 1590–
1595.
46 W. F. Goebel and F. H. Babers, J. Biol. Chem., 1934, 106,
63–69.
47 R. Nakajima, M. Ono, S. Aiso and H. Akita, Chem. Pharm.
Bull., 2005, 53, 684–687.
48 X.-R. Zhu, Y. Zheng, P. Wang, S.-B. Yang, R. Chen and
Y.-Q. Zhu, J. Chin. Pharm. Sci., 2010, 19, 307–311.
25 S.-C. Hung, X.-A. Lu, J.-C. Lee, M. D.-T. Chang, S.-l. Fang,
T.-c. Fan, M. M. L. Zulueta and Y.-Q. Zhong, Org. Biomol. 49 B. Helferich and A. Berger, Chem. Ber., 1957, 90, 2492–
Chem., 2012, 10, 760–772. 2498.
26 P. Schell, H. A. Orgueira, S. Roehrig and P. H. Seeberger, 50 E. Gemma, A. N. Hulme, A. Jahnke, L. Jin, M. Lyon,
Tetrahedron Lett., 2001, 42, 3811–3814.
R. M. Müeller and D. Uhrín, Chem. Commun., 2007, 2686–
27 X. Cao, Q. Lv, D. Li, H. Ye, X. Yan, X. Yang, H. Gan,
2688.
W. Zhao, L. Jin, P. Wang and J. Shen, Asian J. Org. Chem., 51 R. N. Rej, J. N. Glushka, W. Chew and A. S. Perlin, Carbo-
2015, 4, 899–902.
hydr. Res., 1989, 189, 135–148.
28 T. Chiba and P. Sinaÿ, Carbohydr. Res., 1986, 151, 379–389.
29 D. Medaković, Carbohydr. Res., 1994, 253, 299–300.
52 Y. V. Voznyi, I. S. Kalicheva and A. A. Galoyan, Bioorg.
Khim., 1987, 13, 1655–1658.
30 H. N. Yu, J. Furukawa, T. Ikeda and C.-H. Wong, Org. Lett., 53 B. Ernst and T. Winkler, Tetrahedron Lett., 1989, 30, 3081–
2004, 6, 723–726. 3084.
31 Y. V. Voznyi, I. S. Kalicheva, A. A. Galoyan and 54 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
N. B. Gusina, Bioorg. Khim., 1989, 15, 1411–1415.
32 S. Blanchard, M. Sadilek, C. R. Scott, F. Turecek and
M. H. Gelb, Clin. Chem., 2008, 54, 2067–2070.
33 R. J. Ferrier and R. H. Furneaux, J. Chem. Soc., Perkin Trans.
1, 1977, 1996–2000.
34 L. Somsák and R. J. Ferrier, Adv. Carbohydr. Chem.
Biochem., 1991, 49, 37–92.
35 G. N. Bollenback, J. W. Long, D. G. Benjamin and
J. A. Lindquist, J. Am. Chem. Soc., 1954, 77, 3310–3315.
36 A. G. Pearson, M. J. Kiefel, V. Ferro and M. von Itzstein,
Carbohydr. Res., 2005, 340, 2077–2085.
37 B. Mönch, A. Gebert, F. Emmerling, R. Becker and I. Nehls,
Carbohydr. Res., 2012, 352, 186–190.
38 D. M. Whitfield, G. I. Birnbaum, H. Pang, J. Baptista and
B. Sarkar, J. Carbohydr. Chem., 1991, 10, 329–348.
39 K. Miura, Y. Ichinose, K. Nozaki, K. Fugami, K. Oshima
and K. Utimoto, Bull. Chem. Soc. Jpn., 1989, 62, 143–147.
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand,
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox,
Gaussian 09, Revision D.01, 2009.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
55 MacroModel, version 10.6, Schrödinger, LLC, New York, NY, 60 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens.
2014. Matter Mater. Phys., 1988, 37, 785–789.
56 Maestro, version 9.9, Schrödinger, LLC, New York, NY, 61 P. J. Stephens, F. J. Devlin, C. F. Chabalowski and
2014. M. J. Frisch, J. Phys. Chem., 1994, 98, 11623–11627.
57 J. L. Banks, H. S. Beard, Y. Cao, A. E. Cho, W. Damm, 62 A. V. Marenich, C. J. Cramer and D. G. Truhlar, J. Phys.
R. Farid, A. K. Felts, T. A. Halgren, D. T. Mainz, Chem. B, 2009, 113, 6378–6396.
J. R. Maple, R. Murphy, D. M. Philipp, M. P. Repasky, 63 Y. Zhao and D. G. Truhlar, Phys. Chem. Chem. Phys., 2008,
L. Y. Zhang, B. J. Berne, R. A. Friesner, E. Gallicchio 10, 2813–2818.
and R. M. Levy, J. Comput. Chem., 2005, 26, 1752– 64 S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem.
1780.
Phys., 2010, 132, 154104.
58 A. D. Becke, J. Chem. Phys., 1993, 98, 1372–1377.
59 A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
65 S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem.,
2011, 32, 1456–1465.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.