Facile Approach to C-Glucosides by Using a Protecting-Group-Free Hiyama Cross-Coupling Reaction: High-Yielding Dapagliflozin Synthesis

Article abstract of DOI:10.1002/chem.202101052

Access to unprotected (hetero)aryl pseudo-C-glucosides via a mild Pd-catalysed Hiyama cross-coupling reaction of protecting-group-free 1-diisopropylsilyl-d-glucal with various (hetero)aryl halides has been developed. In addition, selected unprotected pseudo-C-glucosides were stereoselectively converted into the corresponding α- and β-C-glucosides, as well as 2-deoxy-β-C-glucosides. This methodology was applied to the efficient and high-yielding synthesis of dapagliflozin, a medicament used to treat type 2 diabetes mellitus. Finally, the versatility of our methodology was proved by the synthesis of other analogues of dapagliflozin.

Full text of DOI:10.1002/chem.202101052

Accepted Article
Accepted Article
Title: Facile Approach to C-glucosides by Using a Protecting-
Group-Free Hiyama Cross-Coupling Reaction: High-Yielding
Dapagliflozin Synthesis
Authors: Karolína Vaňková, Michal Rahm, Jan Choutka, Radek Pohl,
and Kamil Parkan
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.202101052
Link to VoR: https://doi.org/10.1002/chem.202101052
01/2020

10.1002/chem.202101052
Chemistry - A European Journal
FULL PAPER
Facile Approach to C-glucosides by Using a Protecting-Group-
Free Hiyama Cross-Coupling Reaction: High-Yielding
Dapagliflozin Synthesis
Karolína Vaňková,[a] Michal Rahm,[a] Jan Choutka[a] Radek Pohl[b] and Kamil Parkan*[a]
[a]
K. Vaňková, M. Rahm, J. Choutka, Dr. K. Parkan
Department of Chemistry of Natural Compounds
University of Chemistry and Technology, Prague
Technická 5, 166 28 Prague 6 (Czech Republic)
E-mail: parkank@vscht.cz
[b]
Dr. R. Pohl,
Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences,
Gilead Sciences & IOCB Research Centre
Flemingovo nám. 2, 166 10, Prague, Czech Republic
Supporting information for this article is given via a link at the end of the document.
Abstract: Access to unprotected (hetero)aryl pseudo-C-glucosides
via a mild Pd-catalyzed Hiyama cross-coupling reaction of protecting-
group-free 1-diisopropylsilyl-D-glucal with various (hetero)aryl halides
has been developed. In addition, selected unprotected pseudo-C-
glucosides were stereoselectively converted into the corresponding α-
and β-C-glucosides, as well as 2-deoxy-β-C-glucosides. This
methodology was applied to the efficient and high-yielding synthesis
of dapagliflozin, a medicament used to treat type 2 diabetes mellitus.
Finally, the versatility of our methodology was proved by the synthesis
of other analogues of dapagliflozin.
aryl halides with glycosyl stannanes, described by Walczak et
al.[13] The use of protecting groups suppresses problems with
solubility in organic solvents and purification as well as side
reactivity.[13-14] Nevertheless, depending on the sensitivity of the
introduced functional group, the deprotection of the obtained
saccharide mimetics can become challenging. Therefore, a
straightforward access to unprotected aryl C-glycosides in high
yield is advantageous. In continuation of our interest in the
synthesis of C-glycosides,[4, 9a, 15] we report access to diverse
(hetero)aryl C-glycosides via a key Hiyama cross-coupling
reaction of 1-diisopropylsilyl-D-glucal 4 with different (hetero)aryl
halides under mild conditions. Herein, to the best of our
knowledge, the most suitable and practical synthesis of
dapagliflozin 8n and its derivatives 7n and 10n is also described.
Introduction
In the field of carbohydrate chemistry, aryl C-glycosides are
an important class of natural products[1] and synthetic drugs.[2]
Structurally, they can be viewed as mimetics of aryl O-glycosides,
in which their labile O-glycosidic bond is replaced with a stable
C-C bond. This modification increases their stability towards
chemical and enzymatic hydrolysis substantially, providing them
with great potential as drug candidates. Various C-glycosides
such as papulacandin D,[3] bergenin,[4] thailanstatin A,[5] and
vicenin-2,[6] have been isolated from natural sources and show
significant biological activities. The major use of aryl C-glycosides
includes the inhibition of the sodium-glucose cotransporter-2
(SGLT-2). In recent years, these inhibitors have attracted much
attention due to their effective, safe and well-tolerated control and
regulation of type 2 diabetes.[2b] Accordingly, the FDA has
approved SGLT-2 inhibitors, such as dapagliflozin (Forxiga),
empagliflozin (Jardiance), and canagliflozin (Invokana).[7]
Consequently, many successful strategies to provide a
synthetic access to aryl C-glycosides in a direct or de novo
manner have been already developed.[1] Notable strategies to
construct these saccharide mimetics by transition-metal-mediated
Figure 1. Cross-coupling approaches for the synthesis of aryl C-glycosides.
coupling reactions include Heck,[8] Stille,[9] Suzuki-Miyaura,[4, 9a, 10] Results and Discussion
Negishi[10b,
and Hiyama-Denmark cross-coupling reactions
11]
(Figure 1).[12] All of these methodologies were applied almost
exclusively to O-protected (mostly benzylated and silylated) 1-
substituted glycals except for the Stille cross-coupling reaction of
The first step was the preparation of 1-diisopropylsilyl-D-glucal 4
from commercially available D-glucal 1, which was fully protected
with 2-methoxyprop-2-yl (MOP) groups (Scheme 1). The reaction
1
This article is protected by copyright. All rights reserved.

10.1002/chem.202101052
Chemistry - A European Journal
FULL PAPER
of 2-methoxypropene in the presence of pyridinium tosylate
afforded MOP-D-glucal 2 in an excellent yield of 84%. As we
reported before,[9a] MOP acetal protecting groups represent an
efficient strategy for transient protection of glycals. Their use is
favourable for several reasons, such as their easy introduction
and removal, compatibility with silyl groups, low cost, and stability
under harsh basic conditions, which are required for C-1
derivatization. In the next step, MOP-D-glucal 2 was converted
into a 1-lithiated intermediate by treatment with t-BuLi (3.5 equiv.)
in THF at −78 °C and then this intermediate was reacted with
(iPr)2SiClH to give compound 3. The subsequent mildly acidic
hydrolysis (THF/1% aqueous AcOH in the ratio 1:1, v/v) of 3
resulted in the formation of bench-stable solid 1-diisopropylsilyl-
D-glucal 4 in a notable yield of 78% over three steps.
cross-coupling product, D-glucal 1 was identified as a by-product.
Its origin can be explained by competitive protodesilylation,[18] in
which the diisopropylsilyl group is replaced by a hydrogen atom.
In an attempt to prevent its formation, several reactions with
different timing t2 between the addition of CsF and the addition of
1-iodonaphthalene along with the catalyst were performed.
Unfortunately, the presence of D-glucal 1 was still apparent.
Table 1. The optimisation of the Hiyama cross-coupling reaction.
Entry
R+F-
(equiv.)
Solvent
tR (h)/
t2(min)
T
(°C)
Yield
of 1/5a
(%)[a]
Conversion
of 4
(%)[a]
1
2d
TBAF (2.2)
KF (2.2)
THF
THF
1/10
4/10
0-rt
rt
0/80b
ND/24
0/36
100
ND
ND
63
3d
KF (4.4)
THF
4/10
rt
4d
KF (6.6)
THF
4/10
rt
0/17
Scheme 1. The preparation of 1-diisopropylsilyl-D-glucal
D-glucal 1.
4
from
5d
CsF (2.2)
CsF (4.4)
CsF (4.4)
CsF (4.4)
CsF (4.4)
CsF (4.4)
CsF (4.4)
CsF (4.4)
NH4F (2.2)
DMF
DMF
DMF
DMF
DMF
DMSO
DMSO
DMSO
THF
16/0
rt
53/20
52/14
68/16
ND/38
26/40
ND/ND
26/40
ND/ND
0/0
100
100
100
100
100
ND
100
ND
0
6d
16/0
rt
The reaction of the unprotected 1-diisopropylsilyl-D-glucal 4
with 1-iodonaphthalene was chosen as a model reaction for the
Hiyama cross-coupling.[16] To optimize the reaction conditions, the
fluoride source and its quantity was screened, as summarized in
the Table 1. Other parameters such as the temperature and the
reaction time were varied. The formation of the desired 1-
7d
16/0
60
60
60
60
60
60
rt
8d
16/20
16/60
16/0
9d
1
10d
11d
12d
13
naphthyl-D-glucal 5a was monitored by H NMR with an internal
standard (1,3,5-trimethoxybenzene).
16/20
16/60
16/10
The initial study began with tetrabutylammonium fluoride
(TBAF) as a fluoride source in anhydrous THF at 0 °C (Table 1,
entry 1). A complete conversion of the starting material 4 to
product 5a was observed within 1 h and the reaction conditions
appeared to be satisfactory. Unfortunately, the purification of the
product became problematic because of the presence of
tetrabutylammonium salts, which we were not able to remove by
column chromatography, ion-exchange on DOWEX 50X8 in Et3N
cycle or HPLC. For this reason, it was necessary to find a more
convenient source of fluoride ions.
TMAF•4H2O
14
15
DMF
16/10
rt
0/81c
100
(2.2)
TMAF•4H2O
THF
16/10
rt
0/40c
50
(2.2)
[a] All yields and conversions were determined by 1H NMR with an internal
standard, 1,3,5-trimethoxybenzene; [b] product contained traces of
tetrabutylammonium salts; [c] isolated yield; [d] 18-crown-6 (0.2 equiv.) was
added; tR = reaction time; t2 = time between the addition of fluoride and the
addition of 1-iodonaphthalene with a Pd catalyst; ND = not determined.
We first turned our attention to KF in the presence of a
catalytic amount of 18-crown-6 (0.2 equiv.).[17] We tested different
amounts of KF (2.2, 4.4, and 6.6 equiv.) under constant conditions
in anhydrous THF at room temperature for 4 hours. As shown in
Table 1, the product 5a was detected in all cases, although in low
yields (17–36%) (Table 1, entries 2-4). Moreover, the conversion
of the starting 1-diisopropylsilyl-D-glucal 4 was incomplete, and
the addition of a Pd catalyst, crown ether, an extension of the
reaction time or higher temperature did not lead to increased
conversion. Therefore, CsF was used as a fluoride ion (Table 1,
entries 5-12), because of its greater ionic character compare to
KF. In this case, DMF and DMSO were tested as solvents. To our
delight, the complete conversion of 4 was observed in all
reactions (Table 1, entry 5-12). However, besides the desired
Additionally, NH4F was also tested as a source of fluoride ions
(Table 1, entry 13), but no reactivity of the starting material was
observed, which was attributed to the insufficient ionic character
of this salt. Ideal conditions (Table 1, entry 14) were found using
tetramethylammonium fluoride tetrahydrate (TMAF•4H2O) in DMF.
This quaternary ammonium salt provided almost the same
reaction yield as the experiment using TBAF; in this case, the
cross-coupling product 5a was observed in 81% yield. Its main
advantage lays in the purification step, in which the quaternary
ammonium salts were easily separated from the product by
column chromatography. Based on these facts, TMAF•4H2O
could have become an alternative source of fluoride ions to the
2
This article is protected by copyright. All rights reserved.

10.1002/chem.202101052
Chemistry - A European Journal
FULL PAPER
more problematic TBAF. It is also important to note that the use
of THF as a solvent (Table 1, entry 15) instead of DMF decreased
the yield to 40%. This could be explained by the lower solubility
of TMAF•4H2O in THF.
and 5c) in excellent yields (89% and 94%). Compared to aromatic
iodides, the identical bromides provided lower yields as
demonstrated by the yields of products 5a (81% vs. 51%) and 5b
(89% vs. 54%). Similarly, aryl triflates, which are also commonly
used as suitable donors for cross-coupling reactions, are not able
to compete with identical iodides as is illustrated by no formation
of the product 5a.
Monitoring of the reaction progress by NMR revealed, that
the addition of hydrated TMAF to the solution of 4 leads to
dehydrogenative hydrolysis to silanol
6 (Figure 2). This
Table 2. The substrate scope of the Pd-catalyzed Hiyama cross-coupling
reaction of 1-diisopropylsilyl-D-glucal 4 with different aryl halides.
observation confirms the role of silyl hydride 4 as a masked silanol,
as described before for silacyclobutanes[19] as well as for 2-
thienyl[20] and 2-pyridyl[21] silanes. This hydrolysis is likely
promoted by water present in the TMAF hydrate, as the cross-
coupling reactions were otherwise carried out under anhydrous
conditions. The in situ formation of silanol 6 from 4 was evidenced
by upfield shift of the singlet in decoupled 29Si NMR spectra (from
-1.42 to -3.40 ppm) and the disappearance of 1JSi-H coupling in the
silyl hydride signal in proton-coupled 29Si spectra (Figure 2B). The
formation of silanol 6 was also observed in ESI-MS of the reaction
mixture after the addition of TMAF•4H2O ([M+Na]+ 299.1287).
Reaction conditions: 4 (0.6 mmol, 1.2 equiv.), Aryl-X (0.5 mmol, 1 equiv.),
[PdCl(allyl)]2 (2.5 mol%), TMAF•4H2O (1.1 mmol, 2.2 equiv.), 25 °C, DMF (5
mL), 16 h. Isolated yields. n.d. = not detected.
Electron-poor aromatic iodides, such as 1-iodo-4-
nitrobenzene and methyl 4-iodobenzoate were also well tolerated
substrates and their products 5d and 5e were isolated in high
yields (68% and 89%). Thereafter, we probed whether the C-
glycosidic bond could be formed with heteroaromatic compounds
containing nitrogen or sulphur. The utilisation of 2-iodo-5-
methylthiophene provided 5f in 70% yield. However, the reaction
of the starting silylhydride 4 with 3-iodopyridine afforded 5g in
lowered 25% yield. Further study of ortho-substituted 1-fluoro-2-
iodobenzene with 4 gave the corresponding coupling product 5h
in good 64% yield, whereas bulky 2-iodo-1,3-dimethylbenzene
failed to provide any product. To expand the substrate scope, a
reaction with meta-substituted arene was performed. 1-Fluoro-3-
iodobenzene underwent the coupling reaction to give 5j in 38%
yield. Due to the lower reactivity of aryl bromides, we also
performed a reaction with 1-bromo-4-iodobenzene. As expected,
the reaction resulted in a mixture of mono- and bis-glycosylated
products 5k and 5m, but the major product 5k was successfully
isolated in 56% yield. Besides, we also wondered whether our
method would be suitable for the synthesis of glycopeptides C-
analogues, as the attachment of saccharide to peptides and
Figure 2. Reaction monitoring by NMR. A: Reaction scheme B: Decoupled 29Si
NMR spectra of 4 before (upper spectrum) and after (lower spectrum) the
addition of TMAF•4H2O C: Kinetics of the arylation step from 1H (left) and 29Si
(right) NMR data. Displayed are relative integral intensities after fitting to
pseudo-first order kinetic equation.
The addition of 1-iodonaphthalene and [PdCl(allyl)]2 led to
the conversion of silanol 6 to the arylated product 5a overnight,
as observed by the decay of the silanol signal at -3.40 ppm in 29Si
1
NMR and corresponding arylation kinetics in H NMR (pseudo-
first order rate, k = 1.8 × 10-4 s-1) (Figure 2C, see Supporting
Information (chapter S3) for details).
As shown in Table 2, various aromatic partners were tested
with 1-diisopropylsilyl-D-glucal 4 under optimized conditions. In
general, electron-rich aryl iodides were found to provide better
results. Especially, 5-iodo-1,2,3-trimethoxybenzene and 1-
benzyloxy-4-iodobenzene provided the expected products (5b
3
This article is protected by copyright. All rights reserved.

10.1002/chem.202101052
Chemistry - A European Journal
FULL PAPER
proteins is a common post-translational modification. The reaction
of 4 with N-(tert-butoxycarbonyl)-4-iodo-L-phenylalanine methyl
ester afforded the glycoconjugate 5l in a high yield of 85%. Lastly,
we showed that 1,4-diiodobenzene can be employed in this
transformation with 2.2 equivalents of 4, affording the bis-
glycosylated product 5m in 70% yield.
pseudo-C-glycoside 5n (Table 2) was obtained in excellent 96%
yield. For the synthesis of dapagliflozin, 5n was simply used for
hydroboration-oxidation, which resulted in
a molecule of
dapagliflozin 7n in 82% isolated yield with exclusive β-anomeric
stereoselectivity (Scheme 2). Next, the pseudo-C-glycoside 5n
was subjected to catalytic hydrogenation in the presence of
Adams’ catalyst. As we assumed, the syn-addition of hydrogen
successfully led to the desired 2-deoxy analogue of dapagliflozin
8n in 88% yield (Scheme 2).
9a, 15a]
As we published before,[4,
the protected unsaturated
pseudo-C-glycosides are suitable substrates for further
stereoselective transformations, which produced α- or β-C-
glycosides and 2-deoxy-β-C-glycosides in high yields. As most of
the naturally occurring aryl C-glycosides exist with β-configuration
at the pseudoanomeric centre, selected unprotected cross-
coupling products 5a–d were directly converted to β-C-glucosides
7a–d (Table 3. Route A).
Table 3. A. The preparation of aryl β-C-glucosides 7a-d. B. The preparation of
aryl 2-deoxy-β-C-glucosides 8b and 8h.
Scheme 2. The preparation of dapagliflozin 7n and its analogues 8n and 11n.
For efficiency reasons, we aimed for a one-pot sequence[9a]
of hydroboration followed by oxidation with alkaline hydrogen
peroxide. The resulting products 7a–d were obtained in moderate
yields (46–66%) with only β-selectivity. Moreover, the direct
transformation of unprotected cross-coupling products to aryl 2-
deoxy-β-C-glucosides was examined. In this case, the choice of
the catalyst played a crucial role in the reactivity. The unprotected
pseudo-C-glycosides 5b and 5h were used as the starting
materials. Unfortunately, the hydrogenation of their endocyclic
double bonds in the presence of 10% Pd/C in ethanol resulted in
a mixture of the expected products and sugar ring-opened product.
Due to their very similar RF values, these products could not be
separated. Therefore, several conditions were tested (Pd/C in
THF, methanol and ethyl acetate,[10a] Pd(OH)2/C in ethanol[22] and
Pt2O in ethanol[4]). It turned out that the hydrogenation of the
double bond of 5b and 5h with Adams’ catalyst in ethanol
successfully led to the desired aryl 2-deoxy-β-C-glucosides 8b
and 8h in high yields (83% and 79%) (Table 3. Route B). These
results might be explained by the coordination of palladium
catalysts to endocyclic oxygen and the subsequent reductive
sugar-ring opening.
For completeness, we probed previously published two-step
synthetic procedure[15a, 24] for the first stereoselective preparation
of the α-anomer analogue of dapagliflozin 11n. Firstly, the
pseudo-C-glycoside 5n was benzylated to prevent side reaction
of free hydroxyl groups during the oxidation in the following step.
The benzylation of 5n with BnBr and NaH in the presence of
tetrabutylammonium iodide (TBAI) in THF provided the
benzylated derivative 9n in high 85% yield. The corresponding
pseudo-α-anomer 10n was obtained by the epoxidation of the
endocyclic double bond with dimethyldioxirane (DMDO)[25]
,
followed by the subsequent cleavage of the formed epoxide ring
with lithium triethylborohydride (Super-Hydride®). The α-C-
glycoside 10n was successfully isolated in 80% yield as the only
isomer. The final catalytic debenzylation of 10n in the presence of
Pd(OH)2/C furnished a free analogue of dapagliflozin 11n with α-
D-gluco configuration. The advantage of using Pd(OH)2/C was
that no dehalogenation product was observed and 11n was
obtained in high 86% yield (Scheme 2).
Conclusion
In summary, we have developed a new strategy for the synthesis
of (hetero)aryl C-glycosides via the Pd-catalyzed Hiyama reaction
of unprotected 1-diisopropylsilyl-D-glucal 4 with iodo- or bromo-
(hetero)arenes under mild conditions. It enables the
straightforward synthesis of a variety of (hetero)aryl C-glycosides,
providing access to products that are otherwise difficult to prepare
by known methods and offers a tool for late-stage modifications
of drug molecules. We also report a new practical two-step
The compatibility with various aromatic electrophiles
enabled us to apply this Hiyama coupling reaction to the synthesis
of dapagliflozin, a worldwide approved inhibitor for the treatment
of type 2 diabetes. The required acceptor for the cross-coupling
reaction was prepared via a published three-step synthesis.[23]
According to the general procedure, 1-diisopropylsilyl-D-glucal 4
underwent the Hiyama cross-coupling reaction with the prepared
1-chloro-2-(4-ethoxybenzyl)-4-iodobenzene and the desired
4
This article is protected by copyright. All rights reserved.

10.1002/chem.202101052
Chemistry - A European Journal
FULL PAPER
[1]
[2]
K. Kitamura, Y. Ando, T. Matsumoto, K. Suzuki, Chem.
Rev. 2018, 118, 1495-1598.
a) E. De Clercq, J. Med. Chem. 2016, 59, 2301-2311; b)
D. S. Hsia, O. Grove, W. T. Cefalu, Curr. Opin. Endocrinol.
Diabetes Obes. 2017, 24, 73-79.
S. E. Denmark, C. S. Regens, T. Kobayashi, J. Am.
Chem. Soc. 2007, 129, 2774-2776.
K. Parkan, R. Pohl, M. Kotora, Chem.-Eur. J. 2014, 20,
4414-4419.
strategy for the synthesis of pharmaceutically important
dapagliflozin in the overall yield of 79%. Next, we have verified
that our stereoselective procedure is also useful for the
preparation of dapagliflozin analogues with 2-deoxy-β-D-gluco
and α-D-gluco configuration in good overall yields (84% and 56%)
from 1-diisopropylsilyl-D-glucal 4.
[3]
[4]
[5]
A. K. Ghosh, A. M. Veitschegger, S. Nie, N. Relitti, A. J.
MacRae, M. S. Jurica, J. Org. Chem. 2018, 83, 5187-
5198.
Experimental Section
[6]
[7]
T. C. Ho, H. Kamimura, K. Ohmori, K. Suzuki, Org. Lett.
2016, 18, 4488-4490.
General procedure A for Hiyama cross-coupling reaction
A. R. Aguillón, A. Mascarello, N. D. Segretti, H. F. Z. de
Azevedo, C. R. W. Guimaraes, L. S. M. Miranda, R. O. M.
A. de Souza, Org. Process Res. Dev. 2018, 22, 467-488.
a) K. W. Wellington, S. A. Benner, Nucleosides
Nucleotides Nucleic Acids 2006, 25, 1309-1333; b) F.
Otte, B. Schmidt, J. Org. Chem. 2019, 84, 14816-14829.
a) J. Choutka, R. Pohl, K. Parkan, ACS Omega 2018, 3,
7875-7887; b) E. Dubois, J.-M. Beau, Carbohydr. Res.
1992, 228, 103-120.
a) L. Gong, H.-B. Sun, L.-F. Deng, X. Zhang, J. Liu, S.
Yang, D. Niu, J. Am. Chem. Soc. 2019, 141, 7680-7686;
b) R. W. Friesen, R. W. Loo, J. Org. Chem. 1991, 56,
4821-4823.
TMAF•4H2O (2.2 equiv.) was added to
a 0.24M solution of 1-
diisopropylsilyl-D-glucal (1.2 equiv.) in anhydrous DMF at room
4
[8]
temperature under an argon atmosphere. After 15 min, [PdCl(allyl)]2
catalyst (0.025 equiv.) and the corresponding aryl halide (1 equiv.) were
subsequently added, and the reaction mixture was stirred for 16 h. After
consumption of the starting material, the reaction mixture was
concentrated in vacuo and the resulting residue was purified by column
chromatography on silica to afford the desired product 5a-n. In some
cases, the obtained compounds were additionally purified by prep-HPLC
on Phenomenex (Luna C18) column.
[9]
[10]
[11]
[12]
[13]
M. A. Tius, X. Q. Gu, J. Gomez-Galeno, J. Am. Chem.
Soc. 1990, 112, 8188-8189.
S. E. Denmark, T. Kobayashi, C. S. Regens, Tetrahedron
2010, 66, 4745-4759.
F. Zhu, J. Rodriguez, T. Yang, I. Kevlishvili, E. Miller, D.
Yi, S. O’Neill, M. J. Rourke, P. Liu, M. A. Walczak, J. Am.
Chem. Soc. 2017, 139, 17908-17922.
K. H. Shaughnessy, Molecules 2015, 20, 9419-9454.
a) B. Oroszova, J. Choutka, R. Pohl, K. Parkan, Chem.-
Eur. J. 2015, 21, 7043-7047; b) I. Snajdr, K. Parkan, F.
Hessler, M. Kotora, Beilstein J. Org. Chem. 2015, 11,
1392-1397.
a) F. Foubelo, C. Nájera, M. Yus, Chem. Rec. 2016, 16,
2521-2533; b) S. E. Denmark, L. Neuville, Org. Lett. 2000,
2, 3221-3224.
D. J. Sam, H. E. Simmons, J. Am. Chem. Soc. 1974, 96,
2252-2253.
I. Fleming, J. A. Langley, J. Chem. Soc.-Perkin Trans. 1
1981, 1421-1423.
S. E. Denmark, D. Wehrli, J. Y. Choi, Org. Lett. 2000, 2,
2491-2494.
K. Hosoi, K. Nozaki, T. Hiyama, Proc. Jpn. Acad. Ser. B-
Phys. Biol.Sci. 2002, 78, 154-160.
K. Itami, T. Nokami, Y. Ishimura, K. Mitsudo, T. Kamei, J.-
i. Yoshida, J. Am. Chem. Soc. 2001, 123, 11577-11585.
B. Oroszova, J. Choutka, R. Pohl, K. Parkan, Chem.-Eur.
J. 2015, 21, 7043-7047.
S. Hu, W. Sun, Y. Wang, H. Yan, Med. Chem. Res. 2019,
28, 465-472.
D. C. Koester, M. Leibeling, R. Neufeld, D. B. Werz, Org.
Lett. 2010, 12, 3934-3937.
General procedure B for hydroboration-oxidation
To a 0.025M solution of corresponding coupling product 5a-d and 5n (1
equiv.) in anhydrous THF, BH3•THF complex solution (1.0 M in THF, 10
equiv.) was added dropwise at 0°C. The resulting mixture was stirred for
10 min at this temperature and then warmed to room temperature and
stirred for next 16 h (overnight). Then the mixture was cooled to 0 °C again
and 30% solution of H2O2 and 30% solution of NaOH (2 mL, 1:1) were
[14]
[15]
added simultaneously dropwise. After
a slow transition to room
temperature (approx. 20 min), the reaction mixture was filtered through
Celite and concentrated in vacuo. The residue was purified by column
chromatography on silica to provide the desired aryl β-C-glycoside 7a-d
and 7n. In some cases, the obtained compounds were additionally purified
prep-HPLC on Phenomenex (Luna C18) column.
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
General procedure C for hydrogenation
To a 0.016M solution of corresponding coupling products 5b, 5h, and 5n
(1 equiv.) in ethanol PtO2 (0.05 mmol, 0.5 equiv.) was added. The resulting
mixture was stirred for 40 min under a hydrogen atmosphere for 2 h. The
mixture was filtered through Celite and concentrated in vacuo. The
residue was purified by column chromatography on silica to provide the
desired aryl 2-deoxy-β-C-glycosides 8b, 8h, and 8n.
Acknowledgements
H. Mikula, D. Svatunek, D. Lumpi, F. Glöcklhofer, C.
Hametner, J. Fröhlich, Org. Process Res. Dev. 2013, 17,
313-316.
This work was financially supported from specific university
research (grant No. A1_FPBT_2021_002). We also appreciate
the support from Gilead Sciences, Inc., provided under the
program ‘Molecules for Life’ at the Gilead Sciences & IOCB
Prague Research Centre.
Conflict of interest
The authors declare no conflict of interest.
Keywords: protecting-group-free • Hiyama reaction • cross-
coupling • aryl C-glycosides • dapagliflozin
5
This article is protected by copyright. All rights reserved.

10.1002/chem.202101052
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
A mild palladium-catalyzed Hiyama cross-coupling of 1-diisopropylsilyl-D-glucal with various (hetero)aryl halides has been developed.
This arylation strategy enables the synthesis of a variety of (hetero)aryl pseudo-C-glucosides in good to excellent yields. Without the
need for protecting groups, selected pseudo-C-glucosides were stereoselectively converted into the corresponding C-glucosides, as
well as 2-deoxy-β-C-glucosides. This methodology was applied to the efficient and high-yielding synthesis of dapagliflozin and its
uncommon -isomer.
Institute researcher Twitter usernames: @VSCHT
6
This article is protected by copyright. All rights reserved.