Stereoselective β-Mannosylation by Neighboring-Group Participation

Article abstract of DOI:10.1002/anie.201604358

The stereoselective synthesis of glycosidic bonds is the main challenge of oligosaccharide synthesis. Neighboring-group participation (NGP) of C2 acyl substituents can be used to provide 1,2-trans-glycosides. Recently, the application of NGP has been extended to the preparation of 1,2-cis-glycosides with the advent of C2 chiral auxiliaries. However, this methodology has been strictly limited to the synthesis of 1,2-cis-gluco-type sugars. Reported herein is the design and synthesis of novel mannosyl donors which provide 1,2-cis-mannosides by NGP of thioether auxiliaries. A key element in the design is the use of¹C₄locked mannuronic acid lactones to enable NGP of the C2 auxiliary. In addition to C2 participation a new mode of remote participation of the C4 benzyl group was identified and provides 1,2-cis-mannosides.

Full text of DOI:10.1002/anie.201604358

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201604358
German Edition: DOI: 10.1002/ange.201604358
Glycosylation
Stereoselective b-Mannosylation by Neighboring-Group Participation
Hidde Elferink+, Rens A. Mensink+, Paul B. White, and Thomas J. Boltje*
Abstract: The stereoselective synthesis of glycosidic bonds is
the main challenge of oligosaccharide synthesis. Neighboring-
group participation (NGP) of C2 acyl substituents can be used
to provide 1,2-trans-glycosides. Recently, the application of
NGP has been extended to the preparation of 1,2-cis-glycosides
with the advent of C2 chiral auxiliaries. However, this method-
ology has been strictly limited to the synthesis of 1,2-cis-gluco-
type sugars. Reported herein is the design and synthesis of
novel mannosyl donors which provide 1,2-cis-mannosides by
NGP of thioether auxiliaries. A key element in the design is the
use of 1C4 locked mannuronic acid lactones to enable NGP of
the C2 auxiliary. In addition to C2 participation a new mode of
remote participation of the C4 benzyl group was identified and
provides 1,2-cis-mannosides.
T
he most challenging aspect of oligosaccharide synthesis is
the stereoselective synthesis of glycosidic bonds.[1] For oligo-
saccharide synthesis to become a routine process, broadly
applicable and highly reliable glycosylation methods are
needed. In this respect, the most promising methodology is
based on the neighboring-group participation (NGP) of C2
acyl substituents to provide 1,2-trans-glycosides.[1b] This
methodology is broadly applicable to manno- and gluco-
type sugars and has been applied to (automated) solid-phase
oligosaccharide synthesis (SPOS).[2]
Scheme 1. General overview of NGP in manno- and glucotype donors:
a) Neighboring-group participation by a chiral auxiliary in gluco- and
galacto-type donors resulting in the 1,2-cis product. b) C2 participation
4
of the auxiliary in the C1 conformation resulting in the cis-decalin
1
system and hence 1,2-trans-mannosides. c) A ring flip into the C4
conformation favors the trans-decalin system. Attack of the acceptor at
the b position (axial) results in the b-mannoside. LG=leaving group.
Herein we report a new strategy to enable NGP for the
1
synthesis of 1,2-cis-mannosides by use of the C4 conforma-
Recently, the application of NGP has been extended to
the preparation of 1,2-cis-glycosides with the advent of C2
chiral auxiliaries (Scheme 1a).[3] In this approach, a C2 (S)-
(phenylthiomethyl)benzyl ether is used to trap the oxocarbe-
nium ion from the b-face, thus resulting in the formation of
1,2-cis-glycosides. This methodology holds great promise to
become a generally applicable principle for the synthesis of
1,2-cis-glycosides.[4] However, this methodology has been
strictly limited to the synthesis of 1,2-cis-gluco-type sugars.
The trans-decalin sulfonium ion intermediate is unlikely
formed in mannose and the most likely intermediate would
be a cis-decalin system, which would provide the 1,2-trans-
mannoside instead (Scheme 1b). Conversely, an alternative
method to prepare b-mannosides has been developed by
Crich and co-workers, but this method is mostly limited to the
synthesis of b-mannosides.[5] We therefore explored the
possibility of preparing b-mannosides using NGP since this
method is applicable to other classes of carbohydrates as well.
tion. In this conformation, and using an R-configured
auxiliary, the formation of a trans-decalin intermediate is
possible (Scheme 1c). We designed three mannose donors,
9a–c (see Scheme 2) to study the effect of conformation on
NGP and the stereoselectivity of the ensuing glycosylation.[6]
To obtain the desired 1C4 conformation, we used a 3,6-lactone
bridge which can be introduced by oxidation of the 3,6-diol
precursor using TEMPO/BAIB.[7] Critically, this mild oxida-
tion method is chemoselective and compatible with thioeth-
ers.[8] Typically, the sulfonium ions are obtained after pre-
activation of glycosyl imidates. However, early attempts to
prepare glycosyl imidates of glycosyl lactones failed because
the intermediate lactol underwent ring opening. Hence, we
selected the carboxybenzyl leaving group as it is compatible
with thioethers and can be introduced at an early stage.[9]
Synthesis of 9a–c started from the known benzylidene
protected thiomannosides 1a/b (Scheme 2).[10] The benzyli-
denes in 1a/b were reductively opened at C4 to give the
corresponding 3,6-diols 2a/b, which, after acetylation,
afforded 3a/b in good yields. The thiomannosides 3a/b were
used to glycosylate 2-(allylcarboxy)benzyl (ACB) alcohol
using NIS/TfOH[11] to give 4a/b. Next, the 2-methylnaphthyl
ether of 4a was cleaved by DDQ oxidation to give the alcohol
5. The (R)-(phenylthiomethyl)benzyl ether was introduced
with retention of stereochemistry by BF3-Et2O-promoted
activation of (R)-1-phenyl-2-(phenylthio)ethyl acetate.[3a] To
prepare the donor 9a, 6 was deallylated with [Pd(PPh3)4].
[*] H. Elferink,[+] R. A. Mensink,[+] Dr. P. B. White, Dr. T. J. Boltje
Institute for Molecules and Materials
Heyendaalseweg 135, 6525 AJ, Nijmegen (The Netherlands)
E-mail: t.boltje@science.ru.nl
[+] These authors contributed equally to this work.
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under http://dx.doi.org/10.
1002/anie.201604358.
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 1: Glycosylation results for the donors 9a–c.
a/b[b] 1JC1-H1
[c]
Entry Donor
Acceptor
Yield
[%][a]
[Hz]
1
2
82[d]
71[d]
10:1
170
>20:1
175
172
61[d]
52[e]
3
4
>1:20
>1:20
Scheme 2. Reagents and conditions: a) BH3·THF, Bu2BOTf, THF; 2a,
97%; 2b, 70%; b) Ac2O, pyridine; 3a, 95%; 3b 95%; c) ACB, NIS,
TfOH, DCM; 4a, 82%; 4b, 82%; d) DDQ, DCM/H2O (7.5:1); 5, 75%;
e) (R)-PhCH(OAc)CH2SPh, BF3·Et2O, DCM; 6, 70%; f) AllylOH,
KOtBu; 7a, 85%; 7b, 87% g) TEMPO, BAIB, H2O/DCM (1:10); 8a,
60%; 8b, 80%; h) [Pd(PPh3)4], AcOH/DCM (1:5); 9a, 90%; 9b, 90%,
9c, 85%. ACB=allylcarboxybenzyl, BIAB=bis(acetoxy)iodobenzene,
CB=carboxybenzyl, DCM=dichloromethane, DDQ=2,3-dichloro-5,6-
dicyano-1,4-benzoquinone, Nap=2-methylnaphthyl, NIS=N-iodosuc-
cinimide, (R)-Aux=(R)-CH(Ph)CH2SPh, TEMPO=2,2,6,6,-tetramethyl-
piperidine-N-oxyl, Tf=trifluoromethanesulfonyl.
58[d]
176
87[d,f]
70[e]
5
6
>1:20
>1:20
172
176
37[d]
[a] Yield of isolated product. [b] Ratios determined by integration of key
NMR signals of in the spectra of the crude reaction mixture. [c] The
stereochemistry was determined by the 1JC1,H1 coupling constant: axial
H1 (ꢁ160 Hz) or equatorial H1 (ꢁ170 Hz)[13] [d] Used 2.0 equiv of
glycosyl donor. [e] Used 2.0 equiv of acceptor. [f] Inseparable mixture
with 14c. Bz=benzoyl, DTBMP=2,6-di-tert-butyl-4-methylpyridine.
Next, 4b and 6 were deacetylated using KOtBu in allyl
alcohol to obtain the diols 7a and 7b, respectively. Because
the thioether moiety is readily oxidized using most traditional
oxidation methods, the aforementioned chemoselective
TEMPO/BAIB method was chosen.[7] Oxidation of 7a/b
(60–80%) proceeded with moderate yields to provide 1C4
mannolactones 8a/b. Finally, the compounds 8a/b were
deallylated to afford glycosyl donors 9b/c.
mixed with 2.0 equivalents of DTBMP in CD2Cl2 at ꢀ208C
and activated with 1.0 equivalents of Tf2O in an NMR tube.
1
1
1
1
Next, we glycosylated the donors 9a–c with glycosyl
acceptors 10 and 11 (Table 1, entries 1–6). The donors 9a–c
were preactivated at low temperature (ꢀ788C, ꢀ408C
respectively) with 1.0 equivalents of Tf2O in the presence of
2,6-di-tert-butyl-4-methylpyridine (DTBMP) as a base. All
donors were activated almost instantaneously and reactions
with 9a and 9b were warmed to ꢀ208C before addition of the
acceptor. As expected, 9a produced mainly a-mannosides,
presumably via a cis-decalin sulfonium ion intermediate
(entries 1 and 2). In contrast, glycosylation with 9b showed
excellent b-selectivity in respectable yields (entries 3 and 4).
The b-selectivity observed in these glycosylations supports
the hypothesis that the reaction proceeds by an SN2-like
displacement of the proposed equatorial sulfonium ion.[12]
However, 9c, which because of the high reactivity was
activated at lower temperature (ꢀ788C), produced exclu-
sively b-mannosides with both the primary and secondary
acceptor (entries 5 and 6).These results indicate that reactive
intermediates other than the a-sulfonium ion are also
important for the observed b-selectivity.
At this temperature H, H-TOCSY, H-13C-HSQC, and H-
13C-HMBC NMR spectra were recorded to confirm the
identity of the intermediates present. HMBC analysis
showed a C1–H8ax correlation, which confirmed the presence
of the b-sulfonium ion as expected (see the Supporting
Information).
As depicted in Figures 1a–d, the 1C4 donor 9b was
activated at ꢀ308C under the aforementioned reaction
conditions. At ꢀ208C a clean spectrum of a single species
was observed (Figure 1b). A one-dimensional 1H-TOCSY
NMR spectra (Figure 1c) with irradiation of H2 and H4
revealed a major downfield shift of H1 (d = 5.06 ppm!d =
7.46 ppm). Furthermore, a strong correlation between C1 and
H8ax (d = 4.23 ppm), and to a minor extent with H8eq (d =
1
4.82 ppm) in the H-13C-HMBC spectrum (Figure 1d) indi-
cated formation of a ring system. The 3JH1,H2 coupling constant
of 6.0 Hz supports a diaxial orientation and suggests an a-
sulfonium ion.
The observation of the b-sulfonium ion after activation of
9a and the a-sulfonium ion found following activation of 9b
support the hypothesis that these intermediates are displaced
by the alcohol to provide a- and b-mannosides, respec-
tively.[12] However, glycosylation with the mannosyl donor 9c
To explain the results of the glycosylations shown in
Table 1, VT-NMR studies were performed on the donors 9a–c
to identify reaction intermediates. The 4C1 donor 9a was
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
HMBC spectra characterized these species as an anomeric
triflate and to our surprise, benzyltriflate. To observe the
stability of the triflate, the sample was heated gradually until
one species remained at ꢀ208C (Figure 1 f). Extensive NMR
studies (see the Supporting Information) showed a strong
H1–C4 correlation which, together with the formation of
benzyl triflate, led us to conclude that the tricyclic compound
14c had formed.
We propose that upon activation of 9c, the intermediate
oxocarbenium ion 12c is stabilized by the C4 benzyl ether to
form 13c (Scheme 3). In the absence of the acceptor, this
Scheme 3. Proposed remote participation in the mannuronic acid
lactones 9b and 9c.
intermediate ultimately results in the formation of 14c and
benzyl triflate. The formation of 14c and benzyl triflate
indicate NGP of the C4 benzyl group, and may also explain
the high b-selectivity of 9b.[17] Careful examination of the
glycosylation mixture indeed showed the presence of 14c as
a side product in moderate quantities. A batch experiment to
reproduce 14c from 9c was performed and gave 14c in good
yield (79%). Unlike earlier reported 1,4-anhydro sugars, the
1,4-anhydro-3,6-lactone 14c was stable and neither polymer-
ization nor furanose formation was observed.[18] In addition to
remote participation of the C4 benzyl ether, the expected 3H4
conformation of the oxocarbenium ion may also lead to the b-
product.[15] In principle, although not observed in the NMR
study at ꢀ208C, C4 benzyl participation could also occur in
9b and indeed the tricyclic compound 14b could be retrieved
as a side product in the glycosylation reaction. The formation
of 14b as a side product may explain the moderate yields
observed (Table 1, entries 1–4). Although 14b is ultimately
formed, VT-NMR experiments show that sulfonium ion
formation is highly favored at ꢀ208C. Therefore, in case of
9b, an additional pathway via the sulfonium ion 15 may lead
to b-mannosides but the observed selectivity is more likely to
be a result of the aforementioned alternative pathways.
Figure 1. VT-NMR study of donors 9b,c. Reaction conditions:
i) 2.0 equiv DTBMP+1.0 equiv Tf2O; ii) 10 or 11; iii) Raising temper-
ature. a) 9b pre-activation at ꢀ308C. b) Activated 9b at ꢀ208C. c) 1H-
TOCSY spectrum of the activated 9b. d) HMBC spectrum of 9b post-
activation. e) 1H NMR spectrum of activated 9c at ꢀ808C. f) H NMR
spectrum of activated 9c at ꢀ208C. g) 1H-19F HOESY spectrum of
activated 9c.
was also highly b-selective. Previously, high a-selectivity was
observed in glycosylations with galacturonic acid lactones.[7]
The selectivity was attributed to the disarmed nature of the
acceptor although exemptions were reported in a later
study.[14] The observed b-selectivity could also be explained
by pseudoaxial attack on the 3H4 oxocarbenium ion to provide
the observed 1,2-cis addition.[15] Finally, anomeric triflate
species may also be important intermediates.[16]
To investigate the reaction intermediates, 9c was activated
at ꢀ788C. A VT-NMR study at ꢀ788C showed a mixture of
several species (Figure 1e). Close examination with 19F-NMR
spectroscopy (see the Supporting Information) showed tri-
flated intermediates were present. 1H-19F-HOESY (Fig-
ure 1g) in combination with 1H-13C-HSQC and 1H-13C-
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 2: Opening of the dissacharide lactones.
Entry
Lactone
Product
Yield
[%][a]
1JC1-H1
[Hz]
[1] a) T. J. Boltje, T. Buskas, G. J. Boons, Nat. Chem. 2009, 1, 611 –
622; b) X. M. Zhu, R. R. Schmidt, Angew. Chem. Int. Ed. 2009,
48, 1900 – 1934; Angew. Chem. 2009, 121, 1932 – 1967.
[2] O. J. Plante, E. R. Palmacci, P. H. Seeberger, Science 2001, 291,
1523.
[3] a) J. H. Kim, H. Yang, J. Park, G. J. Boons, J. Am. Chem. Soc.
2005, 127, 12090 – 12097; b) T. J. Boltje, J.-H. Kim, J. Park, G.-J.
Boons, Org. Lett. 2011, 13, 284 – 287.
1
2
90
91
156
155
[4] S. S. Nigudkar, A. V. Demchenko, Chem. Sci. 2015, 6, 2687 –
2704.
[5] a) D. Crich, S. X. Sun, Tetrahedron 1998, 54, 8321 – 8348; b) D.
Crich, S. X. Sun, J. Am. Chem. Soc. 1997, 119, 11217 – 11223;
c) M. Huang, G. E. Garrett, N. Birlirakis, L. Bohꢀ, D. A. Pratt, D.
Crich, Nat. Chem. 2012, 4, 663 – 667.
3
4
89[b]
80
160
156
[6] C. M. Pedersen, L. G. Marinescu, M. Bols, C. R. Chim. 2011, 14,
17 – 43.
[7] L. J. van den Bos, R. E. J. N. Litjens, R. J. B. H. N. van den Berg,
H. S. Overkleeft, G. A. van der Marel, Org. Lett. 2005, 7, 2007 –
2010.
5
90
155
[8] L. J. van den Bos, J. D. C. Codee, J. C. van der Toorn, T. J. Boltje,
J. H. van Boom, H. S. Overkleeft, G. A. van der Marel, Org.
Lett. 2004, 6, 2165 – 2168.
[a] Yield of isolated product. [b] An inseparable mixture with 14c.
[9] K. S. Kim, J. H. Kim, Y. J. Lee, Y. J. Lee, J. Park, J. Am. Chem.
Soc. 2001, 123, 8477 – 8481.
[10] a) T. J. Boltje, C. Li, G.-J. Boons, Org. Lett. 2010, 12, 4636 – 4639;
b) H. A. T. M. Huang, L. Bohe, D. Crich, Carbohydr. Chem.
Proven Synth. Methods 2014, 2, 175 – 181.
[11] G. H. Veeneman, S. H. van Leeuwen, J. H. van Boom, Tetrahe-
dron Lett. 1990, 31, 1331 – 1334.
[12] T. Fang, Y. Gu, W. Huang, G.-J. Boons, J. Am. Chem. Soc. 2016,
138, 3002 – 3011.
[13] K. Bock, C. Pedersen, J. Chem. Soc. Perkin Trans. 2 1974, 293 –
297.
[14] A. E. Christina, L. J. van den Bos, H. S. Overkleeft, G. A.
van der Marel, J. D. C. Codꢀe, J. Org. Chem. 2011, 76, 1692 –
1706.
[15] a) L. Ayala, C. G. Lucero, J. A. C. Romero, S. A. Tabacco, K. A.
Woerpel, J. Am. Chem. Soc. 2003, 125, 15521 – 15528; b) J. D. C.
Codee, L. J. van den Bos, A. R. de Jong, J. Dinkelaar, G. Lodder,
H. S. Overkleeft, G. A. van der Marel, J. Org. Chem. 2009, 74,
38 – 47.
[16] D. Crich, Acc. Chem. Res. 2010, 43, 1144 – 1153.
[17] a) Y. Ma, G. Lian, Y. Li, B. Yu, Chem. Commun. 2011, 47, 7515 –
7517; b) J. Kalikanda, Z. T. Li, J. Org. Chem. 2011, 76, 5207 –
5218.
Finally, the opening of mannuronic acid lactones 16–19
4
with Dowex(H+) in MeOH restored the conformation to C1
and afforded the corresponding methyl esters 20–23 in high
yield.[15b,19] The orientation of H1 was confirmed using the
1JC1-H1 coupling constants, again confirming the 1,2-cis sub-
stitution (Table 2, entries 1–4).[13] Additionally, the mannur-
onic acid lactone 14 was selectively reduced to its corre-
sponding mannose analogue using K-selectride to obtain 24 in
90% yield (entry 5).[20] This approach may be an attractive
way to prepare b-mannosides found in the human core N-
glycan.
In conclusion, we have shown that chiral auxiliaries can be
used to prepare b-mannosides by NGP. To make trans-decalin
1
sulfonium ion formation possible it is imperative to use C4
mannosides. As a result of this conformation, the C4 benzyl
substituent is also able to engage in remote NGP and provide
1,2-cis-mannosides. Although good alternatives for b-man-
nosylation are available, the method described herein relies
on NGP, which is a generally applicable method. Hence, the
use of NGP to prepare 1,2-cis-glycosides of gluco- and
manno-type sugars is now possible, thus bringing us one
step closer to a unified procedure for 1,2-cis glycosylation.
[18] a) F. Micheel, O.-E. Brodde, K. Reinking, Justus Liebigs Ann.
Chem. 1974, 124 – 136; b) T. Uryu, C. Yamaguchi, K. Morikawa,
K. Terui, T. Kanai, K. Matsuzaki, Macromolecules 1985, 18, 599 –
605.
[19] L. J. van den Bos, J. Dinkelaar, H. S. Overkleeft, G. A. van der
Marel, J. Am. Chem. Soc. 2006, 128, 13066 – 13067.
[20] N. M. Yoon, K. E. Kim, J. Kang, J. Org. Chem. 1986, 51, 226 –
229.
Acknowledgments
This work was supported by a NWO-VENI grant.
Received: May 4, 2016
Revised: June 2, 2016
Keywords: chiral auxiliaries · glycosylation ·
neighboring-group effects · stereoselectivity · sulfur
Published online: && &&, &&&&
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Glycosylation
H. Elferink, R. A. Mensink, P. B. White,
T. J. Boltje*
&&&& — &&&&
Stereoselective b-Mannosylation by
Neighboring-Group Participation
Borrowing from next door: Reported
herein is the design and synthesis of
novel mannosyl donors which provide
1,2-cis-mannosides by neighboring-group
participation of thioether auxiliaries. By
using conformationally locked mannur-
onic acid lactone donors, neighboring-
group participation from C2 and C4 could
be exploited to prepare b-mannosides.
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!