4′-Methoxyphenacyl-Assisted Synthesis of β-Kdo Glycosides

Article abstract of DOI:10.1021/acs.joc.6b01431

3-Deoxy-β-d-manno-oct-2-ulosonic acid (β-Kdo) glycosides are mainly found in capsular polysaccharides and extracellular exopolysaccharides from Gram-negative bacteria. These compounds have profound biological implications in immune response and act as virulence factors. We have developed a novel methodology for the stereoselective synthesis of β-Kdo glycosides via the use of a 4′-methoxyphenacyl (Phen) auxiliary group at the C1 position of a peracetylated β-Kdo thioglycoside. Under the promotion of NIS/AgOTf in acetonitrile, a series of Kdo glycosides was synthesized in good yield and β-selectivity while minimizing the formation of undesirable glycals. Stereoselectivity of the glycosylation was shown to be modulated by various factors such as promotor, solvent, anomeric ratio of donor, nature of acceptor, and Phen substitution. Chemoselective cleavage of the Phen group was performed under the action of Zn/HOAc. DFT calculations together with experimental results suggested that α-triflate and a six-membered α-spiroPhen are plausible intermediates of the reaction, accounting for the enhanced formation of β-Kdo glycosides. The developed methodology could be applied to the synthesis of β-Kdo-containing glycans from pathogenic bacteria.

Full text of DOI:10.1021/acs.joc.6b01431

Subscriber access provided by University of Alabama
Featured Article
4#-Methoxyphenacyl-Assisted Synthesis of beta-Kdo Glycosides
Marcelina Mazur, Barbara Barycza, Hanitra Andriamboavonjy, Serge Lavoie,
Marielle Tamigney Kenfack, Anaïs Laroussarie, Yves Bleriot, and Charles Gauthier
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.6b01431 • Publication Date (Web): 14 Sep 2016
Downloaded from http://pubs.acs.org on September 16, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Organic Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 57
The Journal of Organic Chemistry
1
2
3
4
4′ꢀMethoxyphenacylꢀAssisted Synthesis of βꢀKdo Glycosides
5
6
7
8
9
,
,
Marcelina Mazur,† ‡ Barbara Barycza,† ‡ Hanitra Andriamboavonjy,† Serge Lavoie,§ Marielle
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
, , ,
Tamigney Kenfack,† Anaïs Laroussarie,† Yves Blériot,† and Charles Gauthier* † §
ꢀ
†Institut de Chimie IC2MP, CNRSꢀUMR 7285, Équipe Synthèse Organique, Université de
Poitiers, 4 rue Michel Brunet, 86073 Poitiers Cedex 9, France
‡Department of Chemistry, Wroclaw University of Environmental and Life Sciences, Norwida
25, 50ꢀ375 Wroclaw, Poland
§Laboratoire LASEVE, Département des Sciences Fondamentales, Université du Québec à
Chicoutimi, 555 boul. de l’Université, Chicoutimi (Québec), Canada G7H 2B1
ꢀ
INRSꢀInstitut ArmandꢀFrappier, Université du Québec, 531 boul. des Prairies, Laval (Québec),
Canada H7V 1B7
TABLE OF CONTENTS/ABSTRACT GRAPHIC
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 57
1
2
3
4
ABSTRACT: 3ꢀDeoxyꢀβꢀ
D
ꢀmannoꢀoctꢀ2ꢀulosonic acid (βꢀKdo) glycosides are mainly found in
5
6
7
8
capsular polysaccharides (CPS) and extracellular exopolysaccharides (EPS) from Gramꢀnegative
bacteria. These compounds have profound biological implications in immune response and act as
virulence factors. We have developed a novel methodology for the stereoselective synthesis of βꢀ
Kdo glycosides via the use of a 4′ꢀmethoxyphenacyl (Phen) auxiliary group at the C1 position of
a peracetylated βꢀKdo thioglycoside. Under the promotion of NIS/AgOTf in acetonitrile, a series
of Kdo glycosides was synthesized in good yield and βꢀselectivity while minimizing the
formation of undesirable glycals. Stereoselectivity of the glycosylation was shown to be
modulated by various factors such as promotor, solvent, anomeric ratio of donor, nature of
acceptor, and Phen substitution. Chemoselective cleavage of the Phen group was performed
under the action of Zn/HOAc. DFT calculations together with experimental results suggested that
αꢀtriflate and a sixꢀmembered αꢀspiroPhen are plausible intermediates of the reaction, accounting
for the enhanced formation of βꢀKdo glycosides. The developed methodology could be applied to
the synthesis of βꢀKdoꢀcontaining glycans from pathogenic bacteria.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 57
The Journal of Organic Chemistry
1
2
3
4
INTRODUCTION
5
6
7
8
9
3ꢀDeoxyꢀDꢀmannoꢀoctꢀ2ꢀulosonic acid (Kdo) glycosides are mainly found in the surface
polysaccharides of bacteria.1 Kdo glycosides with the αꢀconfiguration are present in virtually all
of the lipopolysaccharide (LPS) core regions of Gramꢀnegative bacteria, playing a crucial role in
the structural integrity of bacterial membranes.2 In contrast, βꢀKdo glycosides occur far less
frequently within LPS. Rare occurrences include: 1) LPS core regions from Alteromonas
macleodii;3 2) LPS Oꢀantigen (OAg) from Providencia alcalifaciens;4 and 3) nonꢀreducing end of
LPS OAg from Klebsiella pneumoniae serotype O12.5 The most frequent occurrence of βꢀKdo
glycosides is within the repeating unit of capsular polysaccharides (CPS),1 which are known as
virulence factors and are involved in protection from host immune mechanisms.6 For instance,
Kingella kingae, a Gramꢀnegative bacteria causing septic arthritis, osteomyelitis, and bacteremia
in young children, produces a CPS featuring a repeating disaccharide comprised of a Kdo residue
in the βꢀconfiguration (Figure 1).7 βꢀKdo glycosides have also been found in extracellular
exopolysaccharides (EPS), such as the one expressed by the ʻTier 1 Select Agentʼ Burkholderia
pseudomallei, the causative agent of melioidosis.8ꢀ10 Cytidine monophosphate (CMP)ꢀKdo,11,12
the activated sugar nucleotide processed by Kdo glycosyltransferases, is another important
example of a naturally occurring compound bearing a βꢀKdo unit. Recently, Whitfield and coꢀ
workers13 have highlighted the presence of polyꢀKdo linkers, containing alternating βꢀ(2→4) and
βꢀ(2→7) linkages, at the reducing end of CPS from various Gramꢀnegative pathogens including
Escherichia coli, Campylobacter jejuni, Haemophilus influenza, Neisseria meningitides, and
Pasteurella multocida. Enzymes involved in the biosynthesis of these polyꢀKdo linkers have been
characterized as novel retaining Kdo transferases (KpsC and KpsS).14,15 Owing to the structural
significance and biological importance of βꢀKdo residues in bacterial polysaccharides,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 57
1
2
3
4
5
6
7
8
9
straightforward synthetic routes towards βꢀKdo glycosides are needed.16,17 Access to these
compounds in pure and homogeneous forms would further the development of vaccines,
diagnostics and therapeutics against some clinically relevant bacterial pathogens.1,2,18
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Naturally occurring βꢀKdoꢀcontaining glycans from bacteria.
The synthesis of Kdo glycosides is not a trivial task, and shares similarities with the glycosylation
of Nꢀacetylneuraminic acid (Neu5Ac).19 The lack of a hydroxyl group at the C3 position, which
hampers the use of the neighboring group participation effect, the presence of an electron
withdrawing carboxylic acid at C1, which deactivates and hinders the anomeric position, and the
undesirable formation of 2,3ꢀglycals are the main issues regarding the synthesis of Kdo
glycosides.16,17 In the last few years, novel methodologies have been implemented allowing
access to αꢀKdo glycosides in excellent yields and stereoselectivity. In this respect, it is worth
mentioning the use of 5,7ꢀOꢀdiꢀtertꢀbutylsilylene protected thioglycoside20 and 3ꢀiodo fluoride21ꢀ
23 donors. Yet, the synthetic chemistry of Kdo glycosides having the opposite thermodynamically
less stable βꢀconfiguration (OR group in equatorial rather than axial position) still requires
improvements.16
4
ACS Paragon Plus Environment

Page 5 of 57
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
Towards this aim, Ling and coꢀworkers,24 relying on the pioneering work of Takahashi,25
developed a novel class of 4,5;7,8ꢀdiꢀOꢀisopropylidene protected 1ꢀCꢀarylglycal donors, which,
upon treatment with Nꢀiodosuccinimide (NIS), led to the stereoselective formation of βꢀKdo
glycosides. Yet, their approach required a supplemental reductive deiodination step followed by
an oxidative transformation in order to provide the carboxylic acid at C1. Recently, Mong and coꢀ
workers26 partially resolved the latter issue by preparing Kdo glycal donors bearing a preinstalled
carboxylate at C1. NISꢀmediated glycosylation of these glycals in a DCM/CH3CN mixture led to
βꢀKdo glycosides in a β/α ratio of up to 20:1 following radical deiodination. In both previous
cases, the presence of a 4,5ꢀOꢀisopropylidene group locking the pyranose ring into a skewꢀboat
conformation was found to be essential for providing high βꢀstereoselectivity. Using unlocked
perbenzylated or peracetylated Kdo glycals led to the opposite transꢀdiaxial selectivity with
regard to the C3 iodine atom and C2 OR group.25,27
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
There have been few reports regarding the synthesis of βꢀKdo glycosides with ‘nonꢀglycal’
donors. van Boom and coꢀworkers28,29 were the first to show that reacting peracetylated βꢀKdo
thioglycosides with NIS/TfOH could provide Kdo glycosides in the major βꢀconfiguration when
3ꢀaminoꢀNꢀbenzyloxycarbonylꢀ1ꢀpropanol was used as an acceptor. More recently, an interesting
study by the group of Oscarson27 revealed that peracylated βꢀKdo thioglycosides were suitable
donors for the formation of βꢀKdo glycosides bearing 2ꢀ(4ꢀtrifluoroacetamidophenyl)ethyl as a
spacer when DMTST or IBr/AgOTf were used as promoters. In all these studies, however, no
systematic evaluations of the glycosylation conditions were performed and no mechanistic details
were provided. Herein, we report a novel approach for the stereoselective synthesis of βꢀKdo
glycosides involving the use of a longꢀrange participating 4′ꢀmethoxyphenacyl (Phen) auxiliary
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 57
1
2
3
4
5
6
7
8
9
group at the C1 position. Glycosylation conditions were thoroughly investigated both in the
presence and in the absence of the Phen group, e.g. promoters, leaving groups, anomeric
configuration of donors, solvents, nature of acceptors, addition order of reagents, and counterꢀ
anions. On the basis of DFT calculations and experimental details, we also propose plausible
intermediates accounting for the formation of both αꢀ and βꢀKdo glycosides under the optimized
reaction conditions.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
Synthetic Approach. Longꢀrange participating effects through the use of an auxiliary group have
been previously described for the synthesis of αꢀNeu5Ac glycosides, which display structural
similarities with βꢀKdo glycosides. The presence of ester chains at the C1 position of Neu5Ac,
such as 2ꢀmethylthioethyl, 2ꢀphenylthioethyl,30 and N,Nꢀdimethylglycolamide,31,32 was shown to
enhance the αꢀselectivity of the glycosylation reaction via the stabilization of the oxocarbenium
ion from the βꢀaxial orientation, thereby favoring the attack of the nucleophile from the αꢀface.
Enhanced αꢀselectivities were also observed for Neu5Ac thioesters33 and 2ꢀcyanoethyl esters34 in
conjunction with CH3CN, presumably through a mechanism involving stabilization of the βꢀ
oriented nitrilium ion.
Inspired by these previous studies, we have devised an analogous approach to tackle the problem
of βꢀKdo glycosides synthesis via the use of a 4′ꢀmethoxyphenacyl (Phen) auxiliary group at the
C1 position of peracetylated Kdo donors. Our choice was driven by two important factors: 1) the
enhanced electronic density of the ketone functionality that would be likely to participate
favorably in the course of the glycosylation reaction, and 2) the orthogonality of phenacyl
6
ACS Paragon Plus Environment

Page 7 of 57
The Journal of Organic Chemistry
1
2
3
4
5
groups35 with several baseꢀ and acidꢀsensitive protecting groups that would be an asset over
previously reported auxiliary groups.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Our working hypothesis is depicted in Figure 2. Once the Kdo donor is activated by a suitable
electrophilic promoter, the oxocarbenium ion (glycosyl cation)36 will be formed. According to
5
Woerpel,37ꢀ40 the attack of the electronꢀrich ketone from the αꢀface of the H4 halfꢀchair
conformer would be favored in order to minimize the destabilizing 1,3ꢀdiaxial interactions. The
resulting sixꢀmembered αꢀspiro compound, which could be found either as a covalent or contact
ion pair intermediate, would then be attacked by the acceptor preferentially from the opposite βꢀ
face leading to enhanced βꢀselectivity for the formation of Kdo glycosides. DFT calculations at
the B3LYP/6ꢀ311++G(2d,2p) level of theory tend to support this hypothesis since the αꢀspiro
intermediate was found to be energetically favored compared to the βꢀspiro intermediate by 13.0
kJ⋅mol−1 (see Figure S1 and Table S1). Moreover, these two intermediates were at least 36.8
kJ⋅mol−1 more stable than the free oxocarbenium ion. Nevertheless, it is important to point out
that, in the case of a CurtinꢀHammett scenario in which there is a rapid exchange between both
intermediates, the major product could also arise from the higher energy ground state
intermediate.41,42 Furthermore, it has to be stressed out that a conformational change from the
more stable Zꢀester to the less stable Eꢀester must occur in order to allow the phenacyl ketone to
approach the glycosyl cation. According to the literature, the energy difference for the Z/E ester
isomerization is about 12.5 kJ⋅mol−1.43 This energy barrier can be lowered by using polar
solvents, such as acetonitrile, or when an electronꢀwithdrawing group is attached to the R′
position of a RCO2R′ ester, such as a phenacyl group.44 In the case of our work, the Z/E ester
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 57
1
2
3
4
5
6
isomerization would be beneficial in terms of energy because it opens the way for the formation
of the spiro intermediates, which are more stable than the free oxocarbenium ion.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OAc
AcO
AcO
OMe
O
O
AcO
O
A
O
E
4'ꢀmethoxyphenacyl
(Phen)
Z- to E-ester
isomerization
–AE
OAc
OAc
OAc
O
O
MeO
OAc
AcO
AcO
O
AcO
O
O
O
OAc
O
O
oxocarbenium
oxocarbenium
4H5
5H4
OMe
unfavored
favored
OAc
OAc
O
AcO
AcO
AcO
AcO
OMe
O
O
O
AcO
AcO
O
O
O
O
-spiro
-spiro
ꢀH+
ꢀH+
ROH
ROH
G = +13 kJ.mol–1
G = 0 kJ.mol–1
OMe
OAc
OAc
AcO
AcO
AcO
AcO
O
O
<
CO2Phen
OR
AcO
AcO
OR
CO2Phen
ꢀKdo
(major)
ꢀKdo
(minor)
Figure 2. Proposed approach for the synthesis of βꢀKdo glycosides through the use of a 4′ꢀ
methoxyphenacyl (Phen) C1ꢀauxiliary group. A = activating group; E = electrophile. The 3D
structures were obtained by DFT geometrical optimization at the B3LYP/6ꢀ311++G(2d,2p) level
of theory (hydrogen atoms have been omitted for the sake of clarity).
8
ACS Paragon Plus Environment

Page 9 of 57
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Synthesis of Kdo Donors. The synthesis of peracetylated Kdo thioglycoside and fluoride donors
2ꢀ5 bearing participating (Phen) or nonꢀparticipating (Bn) ester groups was investigated first
(Scheme 1). Crystalline ammonium Kdo was obtained through the modified Cornforth
procedure45ꢀ47 using an optimized methodology recently reported by Kosma.48 This allowed us to
prepare gram quantities of pure Kdo in a reliable manner. Ammonium Kdo was subjected to
acetylation under standard conditions (Ac2O, py, DMAP, rt) leading to peracetylated 149 with
nearly quantitative yield. Performing the reaction at more elevated temperatures (>30 °C)
generated substantial amounts of Kdo lactone.50 Then, two different routes were studied for the
synthesis of thioglycosides 2 and 3. Esterification with 4′ꢀmethoxyphenacyl bromide (PhenBr) or
BnBr in the presence of Cs2CO3 followed by glycosylation with EtSH under the action of
BF3⋅OEt2 produced donors 2 and 3 with 15 and 67% yield, respectively, both predominantly
featuring the βꢀconfiguration as expected27 (β/α 9:1 for 2, and 7:1 for 3). The low yield obtained
for the Phen derivative 2 was due to the formation of an undesirable byꢀproduct, presumably a
dithioketal coming from the addition of two EtSH molecules on the activated ketone (LRMS: m/z
[M + Na] calcd for C31H44O12S3 727.2; found 727.7). Another route was then investigated in
which free acid 1 was first refluxed in DCE with EtSH and BF3⋅OEt2 followed by esterification of
the resulting thioglycoside. Using this route, donors 2 and 3 were obtained in convenient yields
(45 and 50%, respectively), but with different anomeric ratios than route A (β/α ∼1:1). Fluoride
donors 4 and 5 were synthesized via a similar approach. Regioselective fluorination at the
anomeric position was performed by treatment of peracetylated 1 with HF⋅py 7:3 followed by
standard esterification to provide donors 4 and 5 with 49 and 45% yield, respectively.51 The
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
exclusive αꢀconfiguration of these fluoride donors was confirmed by 19F NMR analysis (3JF,H3ax
34.6 Hz, 3JF,H3eq = 6.0 Hz).
=
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 57
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10
ACS Paragon Plus Environment

Page 11 of 57
The Journal of Organic Chemistry
1
2
3
4
Scheme 1. Synthesis of Kdo Thioglycoside and Fluoride Donors
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In order to provide donor 2 with the same anomeric ratio as 3 (β/α 7:1, route A) and to improve
the yield, an additional methodology was investigated for comparison purposes (Scheme 2).
Therefore, hydrogenolysis of thioglycoside 3 followed by esterification under the aboveꢀ
mentioned conditions provided donor 2 with a good yield (76%) without erosion of
diastereoselectivity (β/α 7:1). Furthermore, this approach allowed us to synthesize Kdo
thioglycoside donors bearing 3′ꢀmethoxy (6), 2′ꢀmethoxy (7) as well as unsubstituted (8)
phenacyl groups all having the same β/α ratio (7:1).
Scheme 2. Synthesis of βꢀKdo Thioglycoside Bearing Diversely Substituted Phenacyl
Groups
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 57
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of βꢀKdo Glycosides. With Kdo donors (2ꢀ8) in hand, study of their glycosylation
behavior for the selective formation of βꢀKdo glycosides was investigated next. Using Phen
thioglycoside 2 in a ∼1:1 β/α ratio together with 5ꢀaminoꢀNꢀbenzyloxycarbonylꢀ1ꢀpentanol (9,52
Figure 3) as a model acceptor, we first screened different thiophilic promoters (Table 1).
Glycosylation reactions in entries 1 to 8 were conducted in the nonꢀparticipating solvent DCE at
–10 °C in the presence of water scavenging 4 Å molecular sieves, and the β/α selectivity ratio
1
was evaluated by H NMR analysis. At this temperature, all promoters were shown to give
excellent conversions (>95%), with the exception of MeOTf (60%), while βꢀanomeric selectivity
varied significantly.
Figure 3. Glycosyl acceptors (9ꢀ14) used in this study.
12
ACS Paragon Plus Environment

Page 13 of 57
The Journal of Organic Chemistry
1
2
3
4
Table 1. Synthesis of βꢀKdo Glycosides: Promoter and Solvent Screening
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
selectivity ratioa
15
entry
promoter
solvent
conv. (%)a
15β
1.5
2.4
3.7
3.0
5.0
5.9
6.0
4.7
5.8
7.9
nd
16
ndb
0.8
2.0
1.0
5.7
1.5
nd
0.6
nd
nd
1.0
1.0
nd
α
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IBr/AgOTf
MeOTf
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
Et2O
CH3CN
CH3CN
DCE
CH3CN
CH3CN
>95
60
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
nd
Me2S2/MeOTfc
Me2S2/Tf2O
NCS/TfOH
NIS/TfOH
>95
>95
>95
>95
41
>95
>95
>95 (84)e
>95
>95
>95
>95
NIS/TfOHd
NIS/AgOTf
NIS/AgOTf
NIS/AgOTf
NIS/AgOTff
NIS/AgOTff
NIS/AgClO4
nd
3.2
3.2
nd
1.0
1.0
NIS/AgBF4
nd
1
b
c
aDetermined by H NMR analysis of the crude reaction mixture. Not detected. In situ formed
d
e
DMTST. Performed at –40 °C. Performed at 1.25 mmol scale. Value in parentheses
f
corresponds to the isolated yield of 15β and 15
α
. Donor and reagents were premixed before the
addition of acceptor 9 (preactivation conditions).
In contrast with the results of Oscarson,27 the use of IBr/AgOTf53 gave the lowest β/α ratio
(1.5:1.0) without formation of glycal 16 (entry 1). We interpreted this result as a competitive
attack of the nucleophilic bromide anion at the anomeric center, forming a reactive βꢀbromide
species54 that can be displaced from the αꢀface by the acceptor. Using Me2S2/MeOTf (DMTST)55
or Me2S2/Tf2O56 provided better βꢀselectivity (up to 3.7:1.0) but glycal formation was observed
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 57
1
2
3
4
5
6
7
8
9
(entries 3 and 4). Enhanced formation of Kdo glycoside 15β was found with Nꢀchlorosuccinimide
(NCS)/TfOH as the promoter (β/α 5.0:1.0) but, again, substantial amounts of glycal 16 were
formed (entry 5). Switching to NIS57ꢀ59 significantly decreased the formation of 16 but kept good
βꢀselectivity (entry 6). Conducting the reaction at –40 °C with NIS/TfOH prevented glycal
formation (entry 7); however, conversion of the donor was not complete (∼41%).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We next examined the use of NIS/AgOTf59 as the promoter, which gave the best results in terms
of βꢀselectivity while minimizing glycal formation (entry 8). This result was somewhat
unexpected since Oscarson27 showed that a similar peracetylated methyl ester Kdo thioglycoside
only furnished elimination product following treatment by NIS/AgOTf in DCM. We then
explored the use of wellꢀknown participating solvents such as Et2O and CH3CN (entries 9 and
10).60 The outcome of these reactions was found to be promising: complete conversion of donor
2, enhanced βꢀselectivity, and no formation of glycal were observed. The reaction in CH3CN was
performed at 1.25 mmol scale providing Kdo glycoside 15 in 84% yield with a β/α ratio of
7.9:1.0. We hypothesized that preactivation conditions61 would be valuable in order to favor the
formation of the αꢀspiro intermediate and would thereby potentially enhance the βꢀselectivity.
Unfortunately, premixing donor 2 with NIS/AgOTf before adding acceptor 9 led exclusively to
glycal 16, either in participating (CH3CN) or nonꢀparticipating (DCE) solvents (entries 11 and
12). The effect of counterꢀanions62,63 was also studied. Therefore, promoters containing anions
lessꢀcoordinating than OTf– such as NIS/AgClO4 and NIS/AgBF4 were evaluated (entries 13 and
14). Using these promoters, βꢀselectivity decreased by more than twoꢀfold, implying that the
reaction intermediates were sensitive to the strength of the coordinating anion. On the basis of
this result, we can hypothesize that a covalentlyꢀbound (or contactꢀion pair) triflate could be one
14
ACS Paragon Plus Environment

Page 15 of 57
The Journal of Organic Chemistry
1
2
3
4
5
6
of the intermediates involved in the glycosylation reaction although this has not been
experimentally proven.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
These results were then compared to those obtained with donor 3 (β/α ratio of ∼1:1) bearing a
nonꢀparticipating benzyl ester at C1 (Table 2). The advantage of using the Phen auxiliary group
was clearly demonstrated here. Indeed, under the promotion of NIS/AgOTf, enhanced βꢀ
selectivity was obtained with donor 2 compared to donor 3 (entries 1 to 4). Notably, the β/α ratio
increased by more than twoꢀfold when CH3CN was used as the solvent (7.9:1.0). Impact of the
starting anomeric ratio of donors 2 and 3 was investigated next (entries 5 to 10). As shown in
previous studies,28,29 performing the reaction with Kdo thioglycosides as major βꢀanomer
(7.0:1.0) significantly improved βꢀselectivity (up to β/α 11.0:1.0 in CH3CN, entry 8). For all of
these reactions, donor 2 bearing a Phen group provided better βꢀselectivity than benzyl ester 3
and no glycal (16 or 18) was detected. As previously mentioned, the use of IBr instead of NIS
decreased selectivity. Next, reactions were performed with αꢀfluoride donors 4 and 5 in order to
probe the impact of the leaving group (entries 11 and 12). Six equiv. of BF3⋅OEt2 were needed to
ensure full conversion of these fluorides.64 Similarly to thioglycosides, βꢀselectivity was
enhanced when the Phen groupꢀcontaining donor 4 was used in comparison with donor 5
although the formation of glycals 16 and 18 was predominant.
15
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 57
1
2
3
4
Table 2. Synthesis of βꢀKdo Glycosides: Influence of Phenacyl Group and Anomeric Ratio
5
6
of Donors
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
selectivity ratioa
donor
(β/ ratio)
entry
solvent
conv. (%)a
glycal
ndb
0.6
nd
nd
nd
nd
nd
nd
nd
nd
α
β
α
1
2
3
4
5
6
7
8
9
3 (1:1)
2 (1:1)
3 (1:1)
2 (1:1)
3 (7:1)
2 (7:1)
3 (7:1)
2 (7:1)
3 (7:1)
2 (7:1)
5 (0:1)
4 (0:1)
DCE
DCE
CH3CN
CH3CN
DCE
90
>95
3.8
4.7
3.4
7.9
4.8
10.8
6.3
11.0
3.0
4.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
4.0
1.0
93
>95
>95
DCE
>95
CH3CN
CH3CN
CH3CNd
CH3CNd
DCMe
DCMe
>95 (94)c
>95
>95
>95
>95
>95
10
11
12
1.1
1.4
aDetermined by 1H NMR analysis of the crude reaction mixture. bNot detected. cPerformed at the
gram scale. Value in parentheses corresponds to the isolated yield of 17β and 17α
. dIBr was used
e
instead of NIS. Fluoride donors 4 and 5 were activated with BF3⋅OEt2 (6.0 equiv) and the
reaction performed at 0 °C for 2 h.
The impact of phenacyl substitution on βꢀselectivity was also examined. As depicted in Scheme
3, thioglycoside donors 6 and 7, bearing the methoxy group at positions 3′ and 2′ on the aromatic
ring, respectively, as well as previously described 2 and unsubstituted 8 were coupled with
acceptor 9 through the optimized glycosylation conditions, thereby generating Kdo glycosides 15,
and 19ꢀ21 with yields ranging from 73 to 80%. Owing to the electronꢀdonating properties of the
16
ACS Paragon Plus Environment

Page 17 of 57
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
methoxy group, it was anticipated that 4′ and 2′ꢀsubstituted Phen derivatives 2 and 7 would give
the highest β/α ratios on an electronic effect basis. Although this was true for donor 2, only
moderate βꢀselectivity (4.6:1.0) was obtained with donor 7. Moreover, similar β/α ratios were
obtained for donors 6 and 8 (∼6.4:1.0), as expected. These results mean that both electronic and
steric effects could be responsible for the formation of the plausible αꢀspiro intermediate, which
accounts for the enhanced βꢀselectivity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3. Impact of Phenacyl Substitution on β/α Ratio
Having studied various parameters modulating βꢀselectivity, we next investigated the general
scope of the glycosylation reaction. In order to do so, a series of acceptors (Figure 3) featuring
primary, secondary, or tertiary alcohols, including 1ꢀnonanol (10), cyclohexanol (11), 2ꢀ
adamantanol (12), 1ꢀadamantanol (13), and methyl 2,3ꢀdiꢀOꢀbenzylꢀαꢀDꢀglucopyranoside (14)
was reacted with thioglycoside donors 2 and 3 under the optimized conditions, i.e. NIS/AgOTf,
CH3CN, –10 °C (Scheme 4). Using 1.4 equiv. of acceptors 10ꢀ14, the reactions provided Kdo
glycosides 22ꢀ31 with fair to very good yields (40ꢀ87%) while minimizing the formation of
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 57
1
2
3
4
5
6
7
8
9
glycals 16 and 18. In all cases, βꢀKdo glycosides were formed predominantly with the exception
of the glycosylation of donor 3 with 1ꢀadamantanol (13) that was moderately αꢀselective (β/α
1.0:1.6). Once again, the effect of the Phen auxiliary group at C1 significantly enhanced βꢀ
selectivity in all cases (up to twoꢀfold) compared to the nonꢀparticipating benzyl ester. It is worth
mentioning that using Phen donor 2 with acceptor 13 led to an inversion in selectivity giving a
slight excess of βꢀKdo glycoside 28 (β/α 1.6:1.0). Moreover, glycosylation with 4,6ꢀdiol 14 was
fully regioselective at the C6 primary position.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4. Scope of the Synthesis of βꢀKdo Glycosides using Different Acceptors
Determination of Anomeric Configuration of Kdo Glycosides. Since Kdo glycosides lack an
anomeric proton at the C2 position, determination of anomeric configuration is not as
18
ACS Paragon Plus Environment

Page 19 of 57
The Journal of Organic Chemistry
1
2
3
4
5
straightforward as it is for other glycosides and cannot rely only on 1H NMR analysis. One of the
most accurate method is the determination of the coupling constant between carbonyl carbon at
6
7
8
9
5
C1 and axial proton at C3.49 For Kdo glycosides adopting a C2 conformation, which is the case
3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
for glycosides 22ꢀ31, a JC1,H3ax value of 5.0ꢀ7.0 Hz is indicative of a βꢀconfiguration while a
value ≤1.0 Hz denotes an αꢀconfiguration. Thus, the αꢀ or βꢀanomeric configuration of Kdo
glycosides 22ꢀ31 was determined via examination of this coupling constant obtained from an
undecoupled 150 MHz 13C NMR experiment. As expected, JC1,H3ax values were found to be
3
between 5.0ꢀ7.0 Hz for βꢀglycosides and ≤1.0 Hz for αꢀglycosides. Furthermore, an interesting
empirical observation was made by comparing the 1H NMR data of αꢀ and βꢀKdo glycosides. We
found that the two geminal protons at C8 were closer (or superimposed) for βꢀKdo glycosides
while the difference of the chemical shifts (ꢁδ) between Hꢀ8a and Hꢀ8b were more pronounced
for αꢀKdo glycosides (ꢁδ from 0.46 to 0.54 ppm). This statement was true for all of the
synthesized Kdo glycosides. However, ꢁδ values between Hꢀ3ax and Hꢀ3eq were not always
smaller for αꢀglycosides, especially for 1ꢀ and 2ꢀadamantyl glycosides 26ꢀ29, and thus, as
recently emphasized by Mong,26 this empirical method could not be reliably used for determining
the anomeric configuration of Kdo glycosides.
Deprotection of Phenacyl Group. As previously mentioned, one of the main advantage of using
a Phen auxiliary group lies in its possible chemoselective cleavage in the presence of other
protecting groups. As examples, Kdo glycosides 22 and 15 were reacted with activated Zn
powder in the presence of AcOH at 35 °C for 2 h (Scheme 5) to produce free carboxylic acids 32
and 33 with very good yields (85 and 89%, respectively). The Phen derivatives were thus
deprotected chemoselectively in the presence of acetyl and NHCbz groups showing the
orthogonality of this auxiliary functionality. Importantly, other reaction conditions that have not
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 57
1
2
3
4
5
6
7
8
9
been tested in the course of this study could also be suitable for the selective cleavage of
phenacyl groups including (Bu3Sn)2O in refluxing DCE; TBAF in THF; H2, Pd/C; and
photodeprotection.65
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 5. ZnꢀMediated Cleavage of Phenacyl Group
Proposed Mechanism. On the basis of the above experimental results, DFT calculations, and
literature precedent,66 reaction mechanism and plausible intermediates were proposed, accounting
for the formation of both αꢀ and βꢀKdo glycosides. As shown in Figure 4, NIS would react with
AgOTf to form an electrophilic species that would activate Kdo thioglycoside donors 2 or 3.
Following activation, oxocarbenium ion I would be formed together with EtSI and a proton
acceptor imine. DFT calculations showed that, owing to the highly unstable nature of
intermediate I, it is likely that acetyl group at C4 would stabilize ion I from the βꢀface, thereby
forming dioxalenium ion intermediate II in the B3,6 conformation (Figure S1). αꢀTriflate III
would also be a plausible intermediate, which would be stabilized by the electronꢀwithdrawing
nature of acetyl groups and carboxylate at C1. Calculations revealed that triflate III would exist
5
in the C2 conformation and be a stable intermediate. As previously discussed, αꢀspiroPhen IV
would be a plausible intermediate, which would be formed by the attack of the Phen activated
ketone on the αꢀface of oxocarbenium ion I. For Kdo donors bearing a Phen group, all these
intermediates (I to IV) would exist in equilibrium reacting in different ways with the acceptor
(R1OH). Glycosylation of oxocarbenium ion I, according to Woerpel model, as well as βꢀ
20
ACS Paragon Plus Environment

Page 21 of 57
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
dioxalenium ion II would produce αꢀKdo glycoside as the major anomer. On the other hand,
glycosylation of αꢀtriflate III and αꢀspiroPhen IV would occur from the βꢀface leading to the
preferential formation of βꢀKdo glycoside. αꢀTriflate III would thus represent a plausible
intermediate, accounting for the good βꢀselectivity obtained with Kdo donor 3 bearing a nonꢀ
participating benzyl ester. Limitations of this mechanistic pathway include difficulties in
explaining the impact of the starting anomeric ratio of donors giving enhanced βꢀselectivity for βꢀ
Kdo thioglycosides in comparison with their αꢀcounterparts (retention of configuration).
Moreover, we cannot rule out the possibility that the generated Nꢀsuccinimide (NHS) would trap
the oxocarbenium ion forming a transient Nꢀglycoside that could be displaced by the glycosyl
acceptor.67ꢀ70
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 57
1
2
3
4
5
6
7
8
9
Figure 4. Proposed mechanism and plausible intermediates for the synthesis of βꢀKdo
glycosides. Dashed lines mean that species II to IV can be found either as covalent or contactꢀion
pair intermediates.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUSIONS
In summary, a novel methodology was developed for the stereoselective formation of βꢀKdo
glycosides by taking advantage of the longꢀrange participating effect of a 4′ꢀmethoxyphenacyl
auxiliary group at the C1 position of a peracetylated Kdo thioglycoside donor. In addition to the
positive effect of the Phen group, various parameters were shown to be crucial for enhancing the
βꢀselectivity of the reaction, including NIS/AgOTf as promoter, CH3CN as solvent, βꢀanomeric
configuration of the donor, as well as paraꢀsubstitution of the Phen aromatic ring. The optimized
glycosylation conditions were applied to the synthesis of a series of Kdo glycosides, providing
good yields and βꢀselectivity while minimizing the formation of undesirable glycals.
Interestingly, chemoselective deprotection of the Phen group was achieved using activated
Zn/HOAc, which represents an advantage over previously reported C1 auxiliary groups. αꢀ
Triflate and sixꢀmembered αꢀspiroPhen were postulated as plausible intermediates, accounting
for the enhanced βꢀstereoselectivity obtained with Kdo thioglycoside donors bearing nonꢀ
participating (Bn) or participating (Phen) groups at C1. The developed methodology could find
application for the synthesis of βꢀKdoꢀcontaining oligosaccharides from pathogenic bacteria.
Work towards this aim is currently in progress in our laboratory.
EXPERIMENTAL SECTION
22
ACS Paragon Plus Environment

Page 23 of 57
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
General Methods. All starting materials and reagents were purchased from commercial sources,
and used as received without further purification. Air and water sensitive reactions were
performed in heat gunꢀdried glassware under Ar atmosphere. Moisture sensitive reagents were
introduced via a dry syringe. Anhydrous solvents were supplied over molecular sieves, and used
as received. Petroleum ether (PE) refers to the 40ꢀ60 °C boiling fraction. Powdered 4 Å
molecular sieves (MS) were activated before use by heating with a heat gun for ≥5 min under
high vacuum. Reactions were monitored by thinꢀlayer chromatography (TLC) with silica gel 60
F254 0.25 mm preꢀcoated aluminium foil plates. Compounds were visualized by using UV254
and/or orcinol (1 mg⋅mL−1) in a 10% H2SO4(aq) solution and/or Hanessian’s stain [2.5 g
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(NH4)6Mo7O24⋅4H2O, 1.0 g Ce(NH4)4(SO4)4⋅2H2O, 90 mL H2O, 10 mL H2SO4] with heating.
Normalꢀphase flash column chromatography was performed on silica gel 60 Å (15ꢀ40 µm). NMR
spectra were recorded at 297 K in the indicated solvent (CDCl3 or MeOD) with 400 or 600 MHz
1
instruments, employing standard software provided by the manufacturer. H and 13C NMR
spectra were referenced to tetramethylsilane (TMS, δH
= δC = 0.00 ppm) as internal reference for
1
13
13
spectra in CDCl3 and MeOD. Assignments were based on H, C, undecoupled C, DEPTꢀ135,
COSY, HSQC, and HMBC experiments. Highꢀresolution mass spectra (HRMS) were recorded
on an ESIꢀQꢀTOF mass spectrometer.
Computational Method. Reaction intermediates were first modeled using Sparta’10 V1.1.0
software package (Wavefunction Inc.). For each intermediate, a conformational distribution was
generated through stochastic MonteꢀCarlo guided searches at the molecular mechanics (MMFF)
level of theory.71 Conformers within 25 kJ⋅mol−1 of the most stable conformer were subjected to
geometry optimizations with Gaussian 09.E01 software.72 Calculations using DFT with the
23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 57
1
2
3
4
5
6
7
8
9
hybrid B3LYP functional73,74 and 6ꢀ31G(d,p) basis set75 were performed with IEFꢀPCM model
solvent (CH3CN).76 The most stable conformer of each intermediate was then further optimized
at the B3LYP/6ꢀ311++G(2d,2p) level of theory.77,78 In these cases, the Grimme’s empirical
dispersion correction was applied (D3 version).79 Energy minima were confirmed at 263 K by
vibrational analysis at the same level of theory, which also allowed for calculation of the Gibbs
free energies. The 3D structures were rendered using PyMOL.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ammonium
2,4,5,7,8ꢀPentaꢀOꢀacetylꢀ3ꢀdeoxyꢀ
αꢀD
ꢀmannoꢀoctꢀ2ꢀulopyranosylonate
(1).
Crystalline ammonium Kdo48 (1.39 g, 5.45 mmol, 1.0 equiv) was suspended in anhydrous py (55
mL), and then Ac2O (55 mL) followed by DMAP (6.6 mg, 54 µmol, 0.01 equiv) were added. The
suspension was stirred for 16 h at rt under Ar after which time the solution was found to be
homogeneous. The mixture was concentrated under reduced pressure, keeping the temperature
below 50 °C, and coevaporated with toluene (3×). The residue was purified by silica gel flash
chromatography (DCM/MeOH 1:0 to 6:4) to give peracetylated Kdo (1, 2.40 g, 95%, ratio α/β >
95:5) as a yellow oil. The physical and analytical data of 149 were in agreement with those
published in the literature.
4′ꢀMethoxyphenacyl
(Ethyl
4,5,7,8ꢀTetraꢀOꢀacetylꢀ3ꢀdeoxyꢀ2ꢀthioꢀα,βꢀDꢀmannoꢀoctꢀ2ꢀ
ulopyranosid)onate (2). Route A: 2ꢀBromoꢀ4′ꢀmethoxyacetophenone (733 mg, 3.20 mmol, 1.6
equiv), TBAI (111 mg, 301 µmol, 0.15 equiv) and Cs2CO3 (1.33 g, 4.08 mmol, 2.0 equiv) were
successively added to a solution of peracetylated 1 (900 mg, 2.01 mmol, 1.0 equiv) in anhydrous
DMF (20 mL). The mixture was stirred for 16 h at rt under Ar. The suspension was filtered over
Celite, rinsed, and diluted with EtOAc (50 mL). The organic phase was washed with a saturated
24
ACS Paragon Plus Environment

Page 25 of 57
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
NH4Cl(aq) solution (25 mL), and H2O (25 mL). The aqueous phase was back extracted with
EtOAc (25 mL). The combined organic layers were dried over MgSO4, and the solvents were
concentrated under reduced pressure. The residue was purified by silica gel flash chromatography
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(PE/EtOAc 8:2 to 6:4) to give 4′ꢀmethoxyphenacyl (2,4,5,7,8ꢀpentaꢀOꢀacetylꢀ3ꢀdeoxyꢀαꢀDꢀ
20
D
mannoꢀoctꢀ2ꢀulopyranosyl)onate (452 mg, 59%) as a white amorphous powder: [
α]
= +84 (c
0.54, CHCl3); 1H NMR (400 MHz, CDCl3) δ 7.91−7.86 (m, 2H, CHꢀAr), 6.99−6.95 (m, 2H, CHꢀ
Ar), 5.46 (d, J = 16.0 Hz, 1H, CHHPhen), 5.43−5.38 (m, 2H, Hꢀ4, Hꢀ5), 5.35 (d, J = 16.0 Hz, 1H,
CHHPhen), 5.25 (ddd, J6,7 = 9.8 Hz, J7,8b = 3.6 Hz, J7,8a = 2.3 Hz, 1H, Hꢀ7), 4.48 (dd, J8a,8b = 12.4
Hz, J7,8a = 2.2 Hz, 1H, Hꢀ8a), 4.18 (dd, J6,7 = 9.9 Hz, J5,6 = 1.0 Hz, 1H, Hꢀ6), 4.13 (dd, J8a,8b
=
12.4 Hz, J7,8b = 3.6 Hz, 1H, Hꢀ8b), 2.49 (dd, J3ax,3eq = 13.1 Hz, J3ax,4 = 11.8 Hz, 1H, Hꢀ3ax), 2.39
(dd, J3ax,3eq = 13.2 Hz, J3eq,4 = 5.1 Hz, 1H, Hꢀ3eq), 2.17, 2.12, 2.05, 2.02, 2.00 (all s, 15H, 5 ×
COCH3); 13C NMR (100 MHz, CDCl3) δ 189.7 (CH2CO), 170.5, 170.2, 170.1, 169.5, 167.9 (5 ×
COCH3), 165.8 (Cꢀ1), 164.3 (CꢀAr), 130.1, 130.0 (CHꢀAr), 127.0 (CꢀAr), 114.2, 114.2 (CHꢀAr),
97.3 (Cꢀ2), 69.5 (Cꢀ6), 67.3 (Cꢀ7), 66.6 (CH2CO), 65.9 (Cꢀ4), 63.9 (Cꢀ5), 62.1 (Cꢀ8), 55.6
(OCH3), 31.6 (Cꢀ3), 20.7, 20.7, 20.6, 20.6, 20.6 (5 × COCH3); HRMS (ESIꢀTOF) m/z [M +
NH4]+ calcd for C27H36NO15 614.2082; found 614.2079; m/z [M + Na]+ calcd for
C27H32NaO15 619.1633; found 619.1635. Ethanethiol (12 µL, 170 µmol, 2.0 equiv) was added to
a solution of the latter phenacyl (50 mg, 84 µmol, 1.0 equiv) in anhydrous DCM (1.7 mL). The
solution was cooled to 0 °C; then, BF3⋅OEt2 (16 µL, 130 µmol, 1.5 equiv) was added. The
mixture was stirred for 24 h under Ar, while gradually being warmed to rt. The solution was
diluted with DCM, and a saturated NaHCO3(aq) solution was added for neutralization. Then,
iodine was added until a dark color persisted. The excess of iodine was reduced by washing the
organic phase with a freshly prepared 10% Na2S2O3(aq) solution until the red color disappeared.
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 57
1
2
3
4
5
6
7
8
9
The solution was poured into a separatory funnel, and the aqueous layer was extracted with
DCM. The pooled organic phases were dried over MgSO4 and filtered, and the solvents were
concentrated under reduced pressure. The residue was purified by silica gel flash chromatography
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(PE/EtOAc 1:0 to 5:55) to give thioglycoside 2 (13 mg, 25%) in a 9.0:1.0 β/
α
anomeric mixture.
20
1
[α]
D
= +50 (c 0.83, CHCl3); H NMR (400 MHz, CDCl3, βꢀanomer) δ 7.93−7.87 (m, 2H, CHꢀ
Ar), 6.95 (m, 2H, CHꢀAr), 5.54−5.38 (m, 3H, CHHPhen, Hꢀ4, Hꢀ5), 5.37−5.28 (m, 1H,
CHHPhen), 5.28−5.20 (m, 1H, Hꢀ7), 4.36−4.33 (m, 2H, Hꢀ8a, Hꢀ8b), 4.24 (dd, J6,7 = 9.6 Hz, J5,6
= 1.1 Hz, 1H, Hꢀ6), 3.89 (s, 3H, OCH3), 2.91−2.77 (m, 1H, SCHH), 2.74−2.66 (m, 1H, SCHH),
2.63 (dd, J3eq,3ax = 12.6 Hz, J3eq,4 = 4.2 Hz, 1H, Hꢀ3eq), 2.23 (t, J3ax,3eq ≈ J3ax,4 = 12.9 Hz, 1H, Hꢀ
1
3ax), 2.13, 2.03, 1.99, 1.97 (s, 12H, 4 × COCH3), 1.26 (t, J = 7.5 Hz, 3H, SCH2CH3); H NMR
(400 MHz, CDCl3, αꢀanomer) δ 7.93−7.87 (m, 2H, CHꢀAr), 6.95 (m, 2H, CHꢀAr), 5.54−5.38 (m,
3H, CHHPhen, Hꢀ4, Hꢀ5), 5.37−5.28 (m, 1H, CHHPhen), 5.28−5.20 (m, 1H, Hꢀ7), 4.63 (dd, J8a,8b
= 12.3 Hz, J7,8a = 2.4 Hz, 1H, Hꢀ8a), 4.53 (dd, J6,7 = 9.7 Hz, J5,6 = 1.3 Hz, 1H, Hꢀ6), 4.12 (dd,
J8a,8b = 12.3 Hz, J7,8b = 3.6 Hz, 1H, Hꢀ8b), 3.89 (s, 3H, OCH3), 2.91−2.77 (m, 1H, SCHH),
2.59−2.53 (m, 1H, SCHH), 2.50 (dd, J3ax,3eq = 11.9 Hz, J3ax,4 = 10.6 Hz, 1H, Hꢀ3ax), 2.33 (dd,
J3ax,3eq = 13.7 Hz, J3eq,4 = 4.8 Hz, 1H, Hꢀ3eq), 2.10, 2.08, 2.03, 2.01, (s, 12H, 4 × COCH3), 1.26
(t, J = 7.5 Hz, 3H, SCH2CH3); 13C NMR (100 MHz, CDCl3, βꢀanomer) δ 189.3 (CO), 170.8,
170.6, 169.9, 169.7 (4 × COCH3), 167.8 (Cꢀ1), 164.3 (CꢀAr), 130.1 (CHꢀAr), 126.7 (CꢀAr), 114.2
(CHꢀAr), 83.9 (Cꢀ2), 72.1 (Cꢀ6), 67.9 (Cꢀ7), 67.2 (Cꢀ4), 66.9 (CH2), 64.4 (Cꢀ5), 62.5 (Cꢀ8), 55.6
(OCH3), 32.8 (Cꢀ3), 22.6 (SCH2β), 20.8−20.7 (4 × COCH3), 14.1 (SCH2CH3); 13C NMR (100
MHz, CDCl3, αꢀanomer) δ 189.6 (CO), 170.5, 170.4, 169.9, 169.7 (4 × COCH3), 167.7 (Cꢀ1),
164.2 (CꢀAr), 130.1 (CHꢀAr), 127.0 (CꢀAr), 114.2 (CHꢀAr), 85.1 (Cꢀ2), 68.3 (Cꢀ6), 67.6 (Cꢀ7),
67.0 (Cꢀ4), 66.6 (CH2), 64.4 (Cꢀ5), 61.9 (Cꢀ8), 55.6 (OCH3), 31.9 (Cꢀ3), 23.5 (SCH2), 20.8−20.7
26
ACS Paragon Plus Environment

Page 27 of 57
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
(4 × COCH3), 13.6 (SCH2CH3); HRMS (ESIꢀTOF) m/z [M + H]+ calcd for C27H35O13S 599.1793;
found 599.1789; m/z [M + NH4]+ calcd for C27H38NO13S 616.2058; found 616.2059; m/z [M +
Na]+ calcd for C27H34NaO13S 621.1612; found 621.1611. Route B: Ethanethiol (261 µL, 3.53
mmol, 5.0 equiv) was added to a solution of peracetylated 149 (328 mg, 705 µmol, 1.0 equiv) in
anhydrous DCE (7.1 mL) at rt under Ar. The solution was cooled to 0 °C and BF3⋅OEt2 (183 µL,
1.48 mmol, 2.1 equiv) was slowly added. The mixture was refluxed for 40 min and then cooled at
rt prior adding Et3N (206 µL, 1.48 mmol, 2.1 equiv). The solvents were concentrated under
reduced pressure to give a residue, which was used in the next step without further purification.
2ꢀBromoꢀ4′ꢀmethoxyacetophenone (322 mg, 1.41 mmol, 2.0 equiv) was added to a solution of
the crude thioglycoside in anhydrous DMF (5.6 mL) followed by Cs2CO3 (69 mg, 212 µmol, 0.3
equiv) and the mixture was stirred for 2 h at rt under Ar. The suspension was filtered over Celite,
rinsed, and diluted with EtOAc (50 mL). The organic phase was washed with a saturated
NH4Cl(aq) solution (25 mL) and H2O (25 mL). The combined organic layers were dried over
MgSO4, and the solvents were concentrated under reduced pressure. The residue was purified by
silica gel flash chromatography (PE/EtOAc 9:1 to 6:4) to give thioglycoside 6 (190 mg, 45%, two
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
steps) as a yellow oil in a ∼1:1 β/
Synthesis of Phenacyl Derivatives starting from Benzyl Ester 3. Thioglycoside 3 (965 mg, 1.79
mmol, 1.0 equiv, ratio β/ 7:1) was dissolved in anhydrous DCE/MeOH (35 mL, 1:4 v/v). The
α anomeric mixture. Route C: Representative Procedure for the
α
solution was degassed with Ar, and 10% Pd/C (965 mg) was added. The suspension was stirred
under an atmosphere of H2 at rt for 2 h. The mixture was filtered over Celite to remove the
catalyst, and the cake was rinsed with MeOH and DCM. The solvents were concentrated under
reduced pressure to give a residue (804 mg, quant.) as a yellow oil, which was used in the next
step without further purification. 2ꢀBromoꢀ4′ꢀmethoxyacetophenone (61 mg, 268 µmol, 1.5
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 57
1
2
3
4
5
6
7
8
9
equiv) was added to a solution of crude carboxylic acid (80 mg, 180 µmol) in anhydrous DMF
(0.9 mL) followed by Cs2CO3 (64 mg, 196 µmol, 1.1 equiv). The mixture was stirred for 2 h at rt
under Ar. The suspension was filtered over Celite, rinsed, and diluted with EtOAc (100 mL). The
organic phase was washed with a saturated NH4Cl(aq) solution (50 mL) and H2O (50 mL). The
combined organic layers were dried over MgSO4, and the solvents were concentrated under
reduced pressure. The residue was purified by silica gel flash chromatography (PE/EtOAc 1:0 to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6:4) to give thioglycoside 2 (81 mg, 76%) as a colorless oil in a 7:1 β/α anomeric mixture.
Benzyl (Ethyl 4,5,7,8ꢀTetraꢀOꢀacetylꢀ3ꢀdeoxyꢀ2ꢀthioꢀα,βꢀDꢀmannoꢀoctꢀ2ꢀulopyranosid)onate (3).
Route A: BnBr (848 µL, 7.09 mmol, 2.2 equiv) followed by Cs2CO3 (420 mg, 1.29 mmol, 0.4
equiv) were added to a solution of peracetylated 149 (1.5 g, 3.22 mmol, 1.0 equiv) in anhydrous
DMF (16 mL). The mixture was stirred for 2 h at rt under Ar. The suspension was filtered over
Celite, rinsed, and diluted with EtOAc (50 mL). The organic phase was washed with a saturated
NH4Cl(aq) solution (25 mL) and H2O (25 mL). The combined organic layers were dried over
MgSO4, and the solvents were concentrated under reduced pressure. The residue was purified by
silica gel flash chromatography (PE/EtOAc 8:2 to 7:3) to give benzyl (2,4,5,7,8ꢀpentaꢀOꢀacetylꢀ
3ꢀdeoxyꢀαꢀDꢀmannoꢀoctꢀ2ꢀulopyranosyl)onate (1.26 g, 73%) as a white foam. The physical and
analytical data of 380 were in agreement with those published in the literature. Ethanethiol (286
µL, 3.86 mmol, 2.0 equiv) was added to a solution of the latter benzyl ester (1.04 g, 1.93 mmol,
1.0 equiv) in anhydrous DCE (10 mL). The solution was cooled to 0 °C; then, BF3⋅OEt2 (357 µL,
2.90 mmol, 1.5 equiv) was slowly added. The mixture was stirred for 2 h under Ar, while
gradually being warmed to rt. The solution was diluted with DCM, and a saturated NaHCO3(aq)
solution was added for neutralization. Then, iodine was added until a dark color persisted. The
28
ACS Paragon Plus Environment

Page 29 of 57
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
excess of iodine was reduced by washing the organic phase with a freshly prepared 10%
Na2S2O3(aq) solution until the red color disappeared. The solution was poured into a separatory
funnel, and the aqueous layer was extracted with DCM. The pooled organic phases were dried
over MgSO4 and filtered, and the solvents were concentrated under reduced pressure. The residue
was purified by silica gel flash chromatography (PE/EtOAc 85:15 to 75:25) to give thioglycoside
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
3 (956 mg, 92%) as a 7.0:1.0 β/α
anomeric mixture. The physical and analytical data of 310 were
in agreement with those published in the literature. Route B: Ethanethiol (307 µL, 4.14 mmol, 5.0
equiv) was added to a solution a peracetylated 149 (386 mg, 829 µmol, 1.0 equiv) in anhydrous
DCE (8.0 mL) at rt under Ar. The solution was cooled to 0 °C and BF3⋅OEt2 (95 µL, 770 µmol,
2.0 equiv) was slowly added. The mixture was refluxed for 40 min and then cooled at rt prior
adding Et3N (230 µL, 4.15 mmol, 5.0 equiv). The solvents were concentrated under reduced
pressure to give a residue, which was used in the next step without further purification. Benzyl
bromide (218 µL, 1.82 mmol, 2.2 equiv) was added to a solution of the crude thioglycoside in
anhydrous DMF (4.1 mL) followed by Cs2CO3 (108 mg, 332 µmol, 0.4 equiv) and the mixture
was stirred for 2 h at rt under Ar. The suspension was filtered over Celite, rinsed, and diluted with
EtOAc (50 mL). The organic phase was washed with a saturated NH4Cl(aq) solution (25 mL) and
H2O (25 mL). The combined organic layers were dried over MgSO4, and the solvents were
concentrated under reduced pressure. The residue was purified by silica gel flash chromatography
(PE/EtOAc 9:1 to 6:4) to give thioglycoside 3 (224 mg, 50%, two steps) as a yellow oil in a ∼1:1
β/α anomeric mixture.
49
50
51
52
53
54
55
56
57
58
59
60
4′ꢀMethoxyphenacyl
(Fluoride
4,5,7,8ꢀTetraꢀOꢀacetylꢀ3ꢀdeoxyꢀαꢀDꢀmannoꢀoctꢀ2ꢀ
ulopyranosyl)onate (4). HF.py (5.0 mL, ca. 70% HF, ca. 30% py) was added to a solution of
29
ACS Paragon Plus Environment