Borinic acid-catalyzed stereo- and site-selective synthesis of β-glycosylceramides

Article abstract of DOI:10.1039/c7cc01673e

A method for activation of unprotected ceramides towards stereo- and site-selective glycosylation is described. Two-point binding of a diarylborinic acid catalyst to the ceramide accelerates its reactions with 'armed' glycosyl methanesulfonate donors, resulting in the formation of a β-glycosidic linkage at the primary OH group.

Full text of DOI:10.1039/c7cc01673e

ChemComm
View Article Online
View Journal
COMMUNICATION
Borinic acid-catalyzed stereo- and site-selective
synthesis of b-glycosylceramides†
Cite this:DOI: 10.1039/c7cc01673e
Kyan A. D’Angelo and Mark S. Taylor
*
Received 3rd March 2017,
Accepted 9th May 2017
DOI: 10.1039/c7cc01673e
rsc.li/chemcomm
A method for activation of unprotected ceramides towards stereo-
and site-selective glycosylation is described. Two-point binding of a
diarylborinic acid catalyst to the ceramide accelerates its reactions
with ‘armed’ glycosyl methanesulfonate donors, resulting in the
formation of a b-glycosidic linkage at the primary OH group.
Glycosphingolipids are a complex class of biomolecules composed
of a ceramide (N-acylated sphingosine) lipid linked to a mono- or
oligosaccharide, which may be modified with sialic acid or sulfate
groups (Scheme 1). They are ubiquitous in mammalian cells, with
the ceramide moiety generally residing in the outer leaflet of the
plasma membrane, displaying the glycan at the cell surface. Their
diverse roles in physiology, human health and disease continue to
be elucidated.1,2 The laboratory synthesis of glycosphingolipids
and their analogs has been pursued intensively as a way to access
homogeneous material for use in biochemical studies,3–5 biological
probe compounds6–9 or potential therapeutic agents.10–12
One of the challenging aspects of glycosphingolipid synthesis is
the selective construction of the glycosidic bond that links the
glycan to the primary OH group of the ceramide. Direct couplings
of glycosyl donors with ceramides are not straightforward: the
reactivity of the primary OH group is attenuated by intramolecular
NHꢀꢀꢀO hydrogen bonding,13 and aggregation of these amphiphilic
acceptors further reduces their reactivity. Azidosphingosine14
and other N-protected sphingosine derivatives15 show improved
reactivity as glycosyl acceptors, and have been employed in
numerous studies, often with protection of the secondary OH
group. Couplings of unprotected ceramide and phytoceramide
with per-O-silylated glycosyl iodides have been shown to provide
efficient and direct access to a-configured glycolipids.16 Other
Scheme 1 Envisioned borinic acid-catalyzed glycosylation of unpro-
tected ceramide, and examples of b-configured glycosphingolipids.
approaches include glycosylations of ceramide-derived dialkyl- of the ceramide moiety.18 Mutant glycosidase enzymes capable of
stannylene acetals17 and intramolecular glycosylation through accelerating the coupling of glycosyl fluorides with sphingosine
tethering of a glucopyranosyl donor to the secondary OH group have been used to generate lyso analogs of glycosphingolipids.19,20
Here, we show that diphenylborinic acid can be employed as
a catalyst for site- and b-selective couplings of an unprotected
ceramide with glycosyl methanesulfonates (mesylates). This work
Department of Chemistry, University of Toronto, Toronto, ON, M5S 3H6, Canada.
E-mail: mtaylor@chem.utoronto.ca
builds on our recent discovery that diarylborinic acid catalysts
are able to activate acceptors containing 1,2- or 1,3-diol groups
† Electronic supplementary information (ESI) available: Synthetic protocols,
characterization data and copies of NMR spectra. See DOI: 10.1039/c7cc01673e
This journal is ©The Royal Society of Chemistry 2017
Chem. Commun.

View Article Online
Communication
ChemComm
towards reactions with glycosyl mesylates.21 These glycosylations acceptors having free primary or secondary OH groups, giving
gave high regioselectivity for the primary position of a 1,3-diol group, a modest preference for a-glycoside products. The limited
generated 1,2-trans-configured linkages using mesylate donors solubility of 1 in CH2Cl2, as well as deactivation of the OH
lacking protective groups capable of anchimeric assistance, and groups by intramolecular hydrogen bonding, likely contribute
took place under mild, homogeneous reaction conditions. Each of to its low reactivity in the absence of catalyst.
these factors suggested that borinic acid catalysis with glycosyl
In addition to the rate acceleration relative to the uncatalyzed
mesylate electrophiles could be useful for the challenging couplings reaction, the b-selectivity of the borinic acid-promoted coupling of 1
of unprotected ceramides, as depicted in Scheme 1. This approach with an armed glycosyl donor is a noteworthy observation. With
differs from the arylborinic acid-catalyzed coupling of glycal epoxides very few exceptions,24 couplings of sphingosine and phytosphingo-
and protected phytosphingosine derivatives that was reported sine derivatives with benzyl ether-protected glycosyl donors yield
recently by Takahashi and Toshima:22 the latter reaction generates primarily a-glycosidic products. Ester protection of the 2-OH group
a-configured glycolipids through intramolecular aglycon delivery of the glycosyl donor is the primary strategy for obtaining
mediated by the boron catalyst, whereas the present method b-glycosides. This approach requires that the amount of ortho-
involves activation of the unprotected acceptor through two-point ester side product be minimized, either by judicious selection of
binding, and was expected to deliver products of b-configuration carbonyl protective groups (e.g., pivaloyl,25 chloroacetyl26) or by
through an associative mechanism.
employing conditions that promote the rearrangement of ortho-
The coupling of N-palmitoyl-D-erythro-sphingosine (1) and donor esters to glycosides.17 The ability to generate b-glycosylceramides
2a was attempted under the previously optimized conditions for without relying on neighboring group participation may offer
borinic acid-catalyzed reactions of glycosyl mesylates (Scheme 2). increased flexibility in the design of protective group strategies
para-Methoxybenzyl (PMB) ethers were chosen as ‘arming’23 protec- to access this class of targets. Deprotection of the PMB groups of
tive groups that could be removed without concomitant reduction of compound 3a was accomplished by treatment with trifluoroacetic
the carbon–carbon double bond of the ceramide. Mesylate 2a was acid in the presence of anisole as a carbocation scavenger,27
generated in situ from the corresponding hemiacetal by treatment yielding b-glucosylceramide 4a.
with methanesulfonic anhydride (Ms2O) in the presence of 1,2,2,6,6-
b-Galactosylceramide 4b and b-lactosylceramide 4c were
pentamethylpiperidine (PMP). We have shown that under these prepared in a similar way, by diphenylborinic acid-promoted
conditions, MsOꢁ-promoted anomerization is relatively rapid, with coupling of 1 with the corresponding glycosyl mesylates, followed by
the a-anomer being favored at equilibrium.21 Using diphenylborinic acid-mediated cleavage of the PMB ether groups (Scheme 3). The
anhydride‡ (0.25 equiv., corresponding to a 50 mol% loading of separation of protected b-lactosylceramide derivative 3c from the
diphenylborinic acid) as catalyst in dichloromethane at 23 1C, the hemiacetal resulting from donor hydrolysis was challenging, and so
protected b-glucosylceramide derivative 3a was obtained in 77% the overall yield of 4c over the two-step process is reported. It should
yield (Scheme 2). The stereoselectivity of this glycosylation was at be noted that the cleavage of the PMB groups was quite efficient:
least 19 : 1 b:a, as judged by analysis of the unpurified reaction the modest yields of the unprotected glycosylceramides reflect
1
mixture by H NMR spectroscopy. In the absence of catalyst, the the challenging purification of these amphiphilic glycolipids.
yield of 3a was less than 5%: the reaction mixture consisted
In conclusion, we have shown that diarylborinic acid catalysis
primarily of unreacted 1 and the hemiacetal resulting from hydro- can be used to effect the otherwise challenging direct glycosylation
lysis of mesylate 2a. The small amounts of glycoside product formed of N-palmitoyl-D-erythro-sphingosine. The organoboron-catalyzed
in the uncatalyzed reaction were primarily a-configured, as judged couplings with PMB-protected glycosyl mesylate donors take place
by analysis of the HSQC and COSY NMR spectra of the unpurified at room temperature under mild, homogeneous conditions. They
reaction mixture (see the ESI†). We have previously found that deliver the b-1,10-linkages characteristic of the major classes of
glycosyl mesylates show appreciable uncatalyzed reactivity with mammalian glycosphingolipids, without protection of the cera-
mide moiety and using armed glycosyl donors lacking protective
groups capable of anchimeric assistance. The catalyst loadings are
relatively high, resulting in fewer than two turnovers, but Ph2BOH
is significantly less valuable than the substrates (and the products)
and its separation from the products is generally straightforward.
The ability to conduct direct glycosylations of unprotected ceramides
may lead to new opportunities to develop convergent syntheses of
glycolipids and their derivatives. These results further illustrate the
utility and versatility of borinic acids as promoters for glycosylation
reactions of complex partners.28,29
This work was supported by NSERC (Discovery Grant and
Canada Research Chairs Programs), the Canada Foundation for
Innovation (projects #17545 and #19119), and the Ontario
Ministry of Research and Innovation. Dr Sean Liew (Yudin
group, University of Toronto) is acknowledged for contribu-
Scheme 2 Synthesis of b-glucosylceramide.
tions to preliminary experiments related to this study. Dr Darcy
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2017

View Article Online
ChemComm
Communication
´
4 J. A. Morales-Serna, O. Boutureira, Y. Dıaz, M. I. Mateu and
´
S. Castillon, Carbohydr. Res., 2007, 342, 1595–1612.
5 M. Kiso, H. Ishida, H. Ando and A. Imamura, in Glycoscience: Biology
and Medicine, ed. N. Tanigushi, T. Endo, G. W. Hart, P. H. Seeberger
and C.-H. Wong, Springer, Japan, 2015, pp. 331–338.
6 A. K. Yadav, D. L. Shen, X. Shan, X. He, A. R. Kermode and
D. J. Vocadlo, J. Am. Chem. Soc., 2015, 137, 1181–1189.
7 G.-Y. Yang, C. Li, M. Fischer, C. W. Cairo, Y. Feng and S. G. Withers,
Angew. Chem., Int. Ed., 2015, 54, 5389–5393.
8 N. Komura, K. G. N. Suzuki, H. Ando, M. Konishi, M. Koikeda,
A. Imamura, R. Chadda, T. K. Fujiwara, H. Tsuboi, R. Sheng,
W. Cho, K. Furukawa, K. Furukawa, Y. Yamaguchi, H. Ishida,
A. Kusumi and M. Kiso, Nat. Chem. Biol., 2016, 12, 402–410.
9 M. Budani, M. Mylvaganam, B. Binnington and C. Lingwood, J. Lipid
Res., 2016, 57, 1728–1736.
10 J. S. Schneider, A. Pope, K. Simpson, J. Taggart, M. G. Smith and
L. DiStefano, Science, 1992, 256, 843–846.
11 Y. Matsuoka, M. Saito, J. LaFrancois, M. Saito, K. Gaynor, V. Olm,
L. Wang, E. Casey, Y. Lu, C. Shiratori, C. Lemere and K. Duff,
J. Neurosci., 2003, 23, 29–33.
12 A. Di Pardo, V. Maglione, M. Alpaugh, M. Horkey, R. S. Atwal,
J. Sassone, A. Ciammola, J. S. Steffan, K. Fouad, R. Truant and
S. Sipione, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 3528–3533.
13 M. A. Peterson and R. Polt, J. Org. Chem., 1993, 58, 4309–4314.
14 R. R. Schmidt and P. Zimmermann, Angew. Chem., Int. Ed., 1986, 25,
725–726.
15 R. Di Benedetto, L. Zanetti, M. Varese, M. Rajabi, R. Di Brisco and
L. Panza, Org. Lett., 2014, 16, 952–955.
16 M. Schombs, F. E. Park, W. Du, S. S. Kulkarni and J. Gervay-Hague,
J. Org. Chem., 2010, 75, 4891–4898.
´
´
17 J. A. Morales-Serna, Y. Dıaz, M. I. Matheu and S. Castillon, Eur.
J. Org. Chem., 2009, 3849–3852.
18 K. Fujikawa, T. Nohara, A. Imamura, H. Ando, H. Ishida and
M. Kiso, Tetrahedron Lett., 2010, 51, 1126–1130.
19 M. D. Vaughan, K. Johnson, S. DeFrees, X. Tang, R. A. J. Warren and
S. G. Withers, J. Am. Chem. Soc., 2006, 128, 6300–6301.
20 J. R. Rich, A.-M. Cunningham, M. Gilbert and S. G. Withers, Chem.
Commun., 2011, 47, 10806–10808.
Scheme 3 Synthesis of b-galactosylceramide (4b) and b-lactosylceramide (4c).
Burns is gratefully acknowledged for assistance with NMR
spectroscopy.
21 K. A. D’Angelo and M. S. Taylor, J. Am. Chem. Soc., 2016, 138,
11058–11066.
22 M. Tanaka, D. Takahashi and K. Toshima, Org. Lett., 2016, 18,
5030–5033.
Notes and references
23 D. R. Mootoo, P. Konradsson, U. Ududong and B. Fraser-Reid,
J. Am. Chem. Soc., 1988, 110, 5583–5584.
24 C. Xia, Q. Yao, J. Schu¨mann, E. Rossy, W. Chen, L. Zhu, W. Zhang,
G. De Libero and P. G. Wang, Bioorg. Med. Chem. Lett., 2006, 16,
2195–2199.
25 Q. Yao, J. Song, C. Xia, W. Zhang and P. G. Wang, Org. Lett., 2006, 8,
911–914.
26 M. Seki, A. Kayo and K. Mori, Tetrahedron Lett., 2001, 42, 2357–2360.
‡ In a previous study, we found that (Ph2B)2O is the optimal catalyst for
activation of 1,3-diol groups in carbohydrate-derived substrates,
whereas an oxaboraanthracene-derived borinic acid provides superior
results for activation of cis-1,2-diol groups (ref. 21). Consistent with this
trend, (Ph2B)2O provided higher activity that the oxaboraanthracene-
derived catalyst for activation of the 1,3-diol group in a ceramide
acceptor.
1 T. Wennekes, R. J. B. H. N. van den Berg, R. G. Boot, G. A. van der 27 E. F. De Medeiros, J. M. Herbert and R. J. K. Taylor, J. Chem. Soc.,
Marel, H. S. Overkleeft and J. M. F. G. Aerts, Angew. Chem., Int. Ed.,
2009, 48, 8848–8869.
Perkin Trans. 1, 1991, 2725–2730.
28 T. M. Beale and M. S. Taylor, Org. Lett., 2013, 15, 1358–1361.
2 C. A. Lingwood, Cold Spring Harbor Perspect. Biol., 2011, 3, a004788. 29 S. O. Bajaj, E. U. Sharif, N. G. Akhmedov and G. A. O’Doherty,
3 Y. D. Vankar and R. R. Schmidt, Chem. Soc. Rev., 2000, 29, 201–216.
Chem. Sci., 2014, 5, 2230–2234.
This journal is ©The Royal Society of Chemistry 2017
Chem. Commun.