Total Synthesis of β- D -ido-Heptopyranosides Related to Capsular Polysaccharides of Campylobacter jejuni HS:4

Article abstract of DOI:10.1021/acs.joc.7b01752

The 6-deoxy-β-d-ido-heptopyranoside related to the capsular polysaccharides of C. jejuni HS:4 is very remarkable, owing to the unique, multifaceted structural features that have been combined into one molecule, which include (1) the rare ido-configuration, (2) the unusual 7-carbon backbone, and (3) the challenging β-(1→2)-cis-anomeric configuration. Two distinct strategies toward the total synthesis of this interesting target are reported. The first involved establishment of the β-d-idopyranosyl configuration from β-d-galactopyranosides, prior to a C-6-homologation extending the d-hexose to the desired 6-deoxy-d-heptose. However, this approach encountered difficulties due to the significantly reduced reactivity of the 6-position of the β-d-idopyranosides, so instead a second strategy was employed, which involved first carrying out a 6-homologation on the less flexible d-galactopyranose, followed by a very successful conversion to the desired β-d-ido-configuration found in the target heptopyranoside (2). This report is the first successful synthesis of the 6-deoxy-β-d-ido-heptopyranoside, which could possess interesting immunological properties.

Full text of DOI:10.1021/acs.joc.7b01752

Subscriber access provided by Columbia University Libraries
Article
Total Synthesis of #-D-Ido-heptopyranosides Related to
Capsular Polysaccharides of Campylobacter jejuni HS:4
Pengfei Zhang, Rachel Hevey, and Chang-Chun Ling
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.7b01752 • Publication Date (Web): 15 Aug 2017
Downloaded from http://pubs.acs.org on August 15, 2017
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 45
The Journal of Organic Chemistry
1
2
3
4
5
6
Total Synthesis of βꢀDꢀIdoꢀheptopyranosides Related to Capsular Polysaccharides of
Campylobacter jejuni HS:4
7
8
9
Pengfei Zhang, Rachel Hevey§ and ChangꢀChun Ling*
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
Alberta Glycomics Centre, Department of Chemistry, University of Calgary, Calgary, Alberta
T2N 1N4, Canada
Corresponding author: Tel.: +1ꢀ403ꢀ220ꢀ2768; fax: +1ꢀ403ꢀ289ꢀ9488;
eꢀmail: ccling@ucalgary.ca
Table of Contents Graphic:
Abstract:
The 6ꢀdeoxyꢀβꢀDꢀidoꢀheptopyranoside related to capsular polysaccharides of C. jejuni HS:4 is
very remarkable, owing to the unique, multifaceted structural features that have been combined
into one molecule which include: (1) the rare idoꢀconfiguration, (2) the unusual 7ꢀcarbon
backbone, and (3) the challenging βꢀ(1→2)ꢀcis anomeric configuration. Two distinct strategies
toward the total synthesis of this interesting target are reported. The first involved establishment
of the βꢀDꢀidopyranosyl configuration from βꢀDꢀgalactopyranosides, prior to a Cꢀ6ꢀhomologation
§ Current address: Molecular Pharmacy, University of Basel, Klingelbergstrasse 50, CHꢀ4056 Basel, Switzerland.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 45
1
2
3
4
5
6
7
8
9
extending the
difficulties due to significantly reduced reactivity of the 6ꢀposition of the βꢀ
instead a second strategy was employed which involved first carrying out a 6ꢀhomologation on
the less flexible ꢀgalactopyranose, followed by a very successful conversion to the desired βꢀ
idoꢀconfiguration found in the target heptopyranoside (2). This report is the first successful
D
ꢀhexose to the desired 6ꢀdeoxyꢀ
D
ꢀheptose. However, this approach encountered
D
ꢀidopyranosides, so
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
D
Dꢀ
synthesis of the 6ꢀdeoxyꢀβꢀ
immunological properties.
Dꢀidoꢀheptopyranoside which could possess interesting
Keywords:
Synthesis; idopyranoside; 6ꢀhomologation; 6ꢀdeoxyꢀβꢀDꢀidoꢀheptopyranoside; 6ꢀdeoxyꢀβꢀDꢀ
galactoꢀheptopyranoside.
ACS Paragon Plus Environment

Page 3 of 45
The Journal of Organic Chemistry
1
2
3
4
1. Introduction
5
6
7
8
Campylobacter jejuni, a gramꢀnegative, spiralꢀshaped bacterial pathogen, is a leading cause
of bacterial infection worldwide.1 , 2 The incidence of C. jejuni infection is extremely high,
ranging from hyperendemic levels (40/100 children under 5 years old) in developing countries of
the world, to endemic levels (0.2–1/100 population) in more developed regions.,4 The infection is
often caused by the consumption of contaminated poultry, water, or raw milk,5,6 and symptoms
can vary from more severe (e.g. diarrhea, fever, abdominal pain), to less severe (e.g. nausea,
headache, muscle pain). Although infection is not usually associated with mortality in developed
countries, C. jejuni infection can result in a high level of morbidity. In addition, a number of
postꢀinfectious sequela can occur and can be very serious; these include the autoimmune
GuillainꢀBarré syndrome (GBS)7 or MillerꢀFisher syndrome (MFS),8 as well as reactive arthritis9
among others. 10 Therefore, the development of a vaccine against C. jejuni infection has
significant importance because of the social and economic impact of the pathogen.11
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
C. jejuni is known to produce lipooligosaccharides (LOS) which can be highly variable in
structure, but often partially resemble human gangliosides. 12 , 13 This molecular mimicry is
associated with the autoimmune disorders that are caused by C. jejuni infection. The bacterium
also produces another class of polysaccharides, called capsular polysaccharides (CPS), which
possess unique structural features. For example, Figure 1 shows the repeating structure of C.
jejuni serotype HS:4 (strain CG8486), the main cause of campylobacteriosis.14 The repeating
structure is a disaccharide that consists of a βꢀ(1→3)ꢀlinked NꢀacetylꢀDꢀglucosamine (GlcNAc)
at the reducing end and a very unusual 6ꢀdeoxyꢀDꢀidoꢀheptopyranose (6ꢀdeoxyꢀidoꢀHep) with βꢀ
anomeric configuration at the nonꢀreducing end. Interestingly, partial modification with Oꢀ
methyl phosphoramidate at the Oꢀ2 and Oꢀ7 positions of the 6ꢀdeoxyꢀidoꢀheptopyranose residues
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 45
1
2
3
4
5
6
was observed, resulting in heterogeneity in the isolated CPS structures. The prevalence of C.
jejuni infection caused by HS:4 strains, as well as the unique structure of 6ꢀdeoxyꢀidoꢀHep make
7
8
9
it an ideal antigen for the development of vaccines against campylobacter infections.15
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
Capsular Polysaccharides of C.jejuni HS:4
O
RO
P
OH
R=H or
7
NH2
OMe
6
OR
4
3
O
5
2
OH
OR
HO
O
NHAc
1
OR
NHAc
O
O
O
O
OH
HO
O
O
O
OH
6-deoxy- -D-ido-heptopyranoside
HO
HO
OH
O
OR
OH
1 R=Me
2 R=(CH2)6NH2
Ph
O
CN
OBn
O
BnO
BnO
OH
OBn
O
O
O
OMe
OMe
OMe
OBn
OBn
OBn
3
4
5
Figure 1. Structure of C. jejuni CPS containing an unusual 6ꢀdeoxyꢀβꢀ
Dꢀidoꢀheptopyranosyl
residue. Compounds 1ꢀ5 are synthetic analogs containing a βꢀ ꢀidopyranosyl configuration.
D
Thus, 6ꢀdeoxyꢀidoꢀHep derivatives such as compounds 1ꢀ2 (Figure 1) represent very
attractive synthetic targets, but their total syntheses present a significant challenge due to their
unusual 7ꢀcarbon backbone which includes a 6ꢀdeoxygenation. In addition, installation of the βꢀ
(1→2)ꢀcisꢀanomeric configuration constitutes a wellꢀrecognized difficulty in carbohydrate
chemistry. Furthermore, the idoꢀconfiguration which places the Cꢀ2, Cꢀ3, and Cꢀ4ꢀhydroxyl
4
groups in an axial position of the pyranoside C1 chair also makes them interesting molecules to
study for their unique conformational flexibility. Previously, we published a short, scalable
ACS Paragon Plus Environment

Page 5 of 45
The Journal of Organic Chemistry
1
2
3
4
synthesis of compound 3 and derivatives starting from βꢀ
Dꢀgalactopyranosides, by carrying out a
5
6
double inversion at both Cꢀ2 and Cꢀ3. 16 ꢀ 18 2,3ꢀDiꢀOꢀsulfonylꢀβꢀ
D
ꢀgalactopyranosides were
7
8
9
prepared as the key substrates, and were converted to the corresponding 2,3ꢀanhydroꢀβꢀ
talopyranoside intermediates; subsequent nucleophilic attack at Cꢀ3 by an alkoxide afforded
orthogonally protected βꢀ ꢀidopyranosides such as compound 3 in a single step from the 2,3ꢀdiꢀ
Dꢀ
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
D
Oꢀsulfonyl substrates. More recently, we succeeded in 6ꢀhomologation of the βꢀidopyranoside
via Swern oxidation and Wittig olefination.19 However, it was found that the Wittig olefination
worked well only with less sterically hindered ylides such as methylenetriphenylphosphorane, to
afford fully protected compound 4. By 1H NMR spectroscopy, it was observed that compound 4
1
preferentially adopts a C4 chair conformation in solution, which orients the vinyl group in a
more sterically hindered axial position, creating a substantial difficulty for subsequent
hydroboration and preventing the desired primary alcohol from being obtained using either
boraneꢀtetrahydrofuran (THF) or 9ꢀBBN as the reagent. We thus altered our strategy to instead
focus on synthesis of the corresponding 6ꢀnitrile derivatives, such as compound 5, since
reduction of the nitrile functionality could open an avenue to obtain the intermediate aldehyde20
which could be further reduced to the desired 6ꢀdeoxyꢀidoꢀHep (1).
2. Results and Discussion
2.1 6ꢀHomologation via nitrile substitution
Starting from the previously published 3ꢀOꢀallylated βꢀDꢀidopyranoside (6) containing a 4,6ꢀ
diol functionality, a selective Oꢀ6ꢀtosylation was carried out in a mixture of anhydrous pyridine
and dichloromethane using pꢀtoluenesulfonyl chloride; this afforded regioselectively the desired
6ꢀtosylate 7 in ~60% yield. Unfortunately, subsequent attempts at performing an SN2 substitution
with cyanide as a nucleophile in DMF at 80 °C proceeded poorly, and the best isolated yield for
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 45
1
2
3
4
5
6
7
8
9
desired compound 9 was only 13%. Attempts to improve the yield using either 18ꢀcrownꢀ6 or
tetraꢀnꢀbutylammonium iodide were both unsuccessful. In fact, when 18ꢀcrownꢀ6 was used, a
small amount of hydrolyzed 6ꢀalcohol (6, up to ~20%) was instead formed, likely the result of a
trace amount of water present in the reagents. This was surprising to observe, given the extreme
difficulty previously encountered in modifying the Cꢀ6 position directly, thus may instead result
from attack of the nucleophilic H2O (or cyanide, or both) at the S atom of the tosylate sulfonate
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
group (as observed for the double inversion of 2,3ꢀdiꢀOꢀsulfonylꢀβꢀDꢀgalactopyranosides).
Although conventionally less electrophilic, since the S center is two bonds away from the highly
sterically hindered Cꢀ6 position, it should be more sterically accessible by a nucleophile.
This difficulty in performing nitrile substitution at Cꢀ6 again confirmed the unusual lack of
reactivity of the Cꢀ6 center in βꢀDꢀidopyranosides. We therefore decided to pursue a related
substrate that instead utilized 6ꢀiodide as the leaving group, which is known to have improved
leaving group ability. Thus, starting from the same diol (6), a direct iodination was carried out in
21 , 22
anhydrous dichloromethane using I2/PPh3 as the reagent and imidazole as base.
Unfortunately, the iodination of compound 6 was found to be low yielding (26%), resulting from
both unreacted starting material (27%) and the formation of a number of unidentified byꢀ
products as indicated by thinꢀlayer chromatography (TLC).
ACS Paragon Plus Environment

Page 7 of 45
The Journal of Organic Chemistry
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
Scheme 1. 6ꢀHomologation via cyanide substitution from methyl βꢀDꢀidopyranosides 6 and 10
containing a 4,6ꢀdiol functionality.
Alternatively, we attempted to convert an analogous derivative, the 3ꢀOꢀbenzylated 4,6ꢀdiol
(10), to the corresponding 6ꢀiodide (11) as above. The reaction proceeded much better to provide
compound 11 in 63% yield. Interestingly, a 1,6ꢀanhydro byꢀproduct (12) was also formed from
the reaction, which was isolated in 26% yield. Formation of the 1,6ꢀanhydro byꢀproduct could be
explained by an intramolecular SN2 attack via an iodideꢀmediated Oꢀdemethylation at the
anomeric position (Scheme 2). The βꢀDꢀidopyranosyl configuration of the substrate may promote
this undesired intramolecular attack by facilitating a ringꢀflip to the 1C4 chair conformer. Identity
1
of the 1,6ꢀanhydro byꢀproduct 12 was confirmed by H NMR spectroscopy of the compound,
which revealed the loss of the anomeric methyl group, as well as the presence of a hydroxyl
1
group that correlates with Hꢀ4 in the Hꢀ1H COSY spectrum; additionally, large coupling
constants were observed between Hꢀ2/Hꢀ3 and Hꢀ3/Hꢀ4 (J2,3 = 8.0 Hz, J3,4 = 8.1 Hz).
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 45
1
2
3
4
Furthermore, electrospray ionization highꢀresolution mass spectrometry (HRMS) also supported
5
6
the proposed structure (calculated m/z for [M + NH4]+: 360.1805; found: 360.1802).
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
Scheme 2. Proposed mechanism for the formation of byꢀproduct 12.
The iodinated intermediate 8 was then subjected to treatment with KCN in DMF at 80 °C
(Scheme 1), but unfortunately although the substitution reaction produced higher yields than
with the 6ꢀtosylate (7), the isolated yield for the desired 6ꢀnitrile 9 was still low (~40%). The
transformation was observed to be quite sluggish, as large amounts of unreacted starting material
were often recovered. Attempts at pushing the reaction further using excess reagent, elevated
temperatures, the addition of crown ether (18ꢀcrownꢀ6), or phase transfer catalysts (Bu4NI) to aid
in solubility were all unsuccessful. Similarly, we also attempted to carry out the displacement on
compound 11 using KCN in DMF at 80 °C to obtain compound 13. However, by TLC it was
found that the expected compound 13 was formed in only low yield, and was therefore not
isolated. In an effort to increase reactivity of the substrate, the 6ꢀOꢀtriflate was also prepared and
subjected to KCN in anhydrous DMF, but unfortunately only hydrolyzed 6ꢀalcohol (6) was
recovered from the reaction mixture, similar to what had been observed previously during
reaction of the 6ꢀOꢀtosylate.
The identity of the desired product (9) could be confirmed by the significant shift of the 13C
and 1H signals associated with Cꢀ6 (20.0 ppm) and Hꢀ6 (2.86 and 2.64 ppm) in the NMR spectra,
ACS Paragon Plus Environment

Page 9 of 45
The Journal of Organic Chemistry
1
2
as well as confirmation via HRMS (calculated m/z for [M + NH4]+: 351.1914; found: 351.1917).
Despite the poor yields of nitrile derivatives 9 and 13, they were subsequently subjected to a
reduction with DIBALꢀH, as it has been well established in literature that nitrile functionalities
can be reduced selectively to an intermediate aldehyde via the imine intermediate20 and could
then be further reduced to the primary alcohol (14 or 15). Unfortunately, all attempted reduction
conditions resulted in a complex product mixture: upon purification, some unreacted starting
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
1
material was isolated together with other byꢀproducts that could not be identified by H NMR
spectroscopy. Although it has been reported that in some instances DIBALꢀH can cleave benzyl
ether groups, this did not appear to be a major pathway of the byꢀproduct formation.23,24
The substantial difficulty encountered during 6ꢀiodination and the subsequent substitution
with cyanide, as well as the later reduction of the 6ꢀnitrile functionality both appear to be
consistent with the reduced reactivity of the 6ꢀposition of βꢀDꢀidopyranosides in general.
Recently, we also reported an unsuccessful 6ꢀhomologation via Wittig olefination using a more
sterically hindered ylide.19 We attribute these unsuccessful chemical transformations to the
4
unusual conformational flexibility of βꢀ
D
ꢀidopyranosides. For example, in the C1 chair
1
conformer, the axial substituent at Cꢀ4 sterically shields the Cꢀ6 position, while in the C4 chair
conformer, the Cꢀ6 group occupies an axial position which is inherently more sterically hindered.
Thus, in either case, the functional group at the Cꢀ6 position of βꢀ
Dꢀidopyranosides experiences
reduced reactivity because of steric clashes. Previously, Streicher et al25 tried to displace the 6ꢀ
tosylate or 6ꢀmesylate of 1,2:3,4ꢀdiꢀOꢀisopropylideneꢀαꢀDꢀgalactopyranose with a cyanide, and
they also encountered significant difficulty because of the steric shielding of the axial 4ꢀ
substituent as is the case in our compounds (7 and 11); the best yield of the 6ꢀnitrile product that
they could isolate was only 4.2%, using the 6ꢀmesylate with LiCN in DMF at 110 °C for 17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 45
1
2
3
4
hours. We thus concluded that the successful synthesis of 6ꢀdeoxyꢀidoꢀHep would require
5
6
revision of our initial strategy.
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
2.2 Revised strategy to synthesize 6ꢀdeoxyꢀidoꢀheptopyranosides.
Our previous strategy provided excellent results with regards to establishing the βꢀ(1→2)ꢀcisꢀ
anomeric configuration and also the
Dꢀidoꢀconfiguration, but failed at 6ꢀhomologation. Thus, a
logical revision of the synthetic strategy would be to reverse the steps by first completing 6ꢀ
homologation on a βꢀ
heptopyranosyl intermediate; subsequent conversion of the 6ꢀdeoxyꢀβꢀ
heptopyranoside to the 6ꢀdeoxyꢀβꢀ
thought it would be advantageous to carry out the 6ꢀhomologation step on the pyranose ring of
ꢀgalactose, because of its reduced flexibility. Scheme 3 shows the retrosynthetic route of the
new strategy by using readily available 1,2:3,4ꢀdiꢀOꢀisopropylideneꢀαꢀ ꢀgalactopyranose (16) as
the starting material. We planned to attempt the 6ꢀhomologation of ꢀgalactose by Wittig
olefination to obtain structure B, carry out a subsequent glycosylation (structures C ꢀ D), and
D
ꢀgalactopyranosyl moiety, to obtain a 6ꢀdeoxyꢀβꢀ
D
ꢀgalactoꢀ
D
ꢀgalactoꢀ
D
ꢀidoꢀheptopyranoside could afford the desired target. We
D
D
D
then finally convert the heptose from Dꢀgalactoꢀ to Dꢀidoꢀconfiguration via the 2,3ꢀanhydroꢀtaloꢀ
intermediate (structure E).
ACS Paragon Plus Environment

Page 11 of 45
The Journal of Organic Chemistry
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
Scheme 3. A revised strategy to synthesize 6ꢀdeoxyꢀβꢀ
Dꢀidoꢀheptopyranosides from Dꢀ
galactopyranose by first performing the 6ꢀhomologation.
To test the feasibility of the new strategy, we first attempted to synthesize the simple 1,2:3,4ꢀ
diꢀOꢀisopropylidene αꢀ ꢀgalactoꢀheptopyranose 19 via Wittig olefination (Scheme 4). Thus,
ꢀ
D
compound 16 was first oxidized to the corresponding aldehyde 17 by pyridinium dichromate
(PDC)26 along with sodium acetate in dichloromethane, to afford the desired aldehyde in
improved yield (82%). Compound 17 was then reacted with methylidenetriphenylphosphorane
(18, Ph3P=CH2, generated in situ by the treatment of methyltriphenylphosphonium bromide with
nꢀbutyllithium, 1 equiv.) as described previously,27 and the desired alkene 19 was successfully
formed and isolated in much improved yield (74%) by column chromatography on silica gel. The
structure of compound 19 was confirmed by the three vinylic protons observed at 5.94, 5.37, and
5.28 ppm. However, when alkene 19 was subjected to a hydroboration under standard reaction
conditions, 28 , 29 no desired primary alcohol 20 was obtained. After work up, several new
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 45
1
2
3
4
compounds were indeed formed as evidenced by a number of spots observed via TLC, but upon
5
6
1
purification none of them had the characteristic H NMR signals corresponding to the
7
8
9
diastereotopic protons Hꢀ6a and Hꢀ6b in the expected heptose 20. The unsuccessful
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
hydroboration was consistent with what has been reported previously.19
Scheme 4. Synthesis of 6ꢀdeoxyꢀ1,2:3,4ꢀdiꢀOꢀisopropylideneꢀαꢀDꢀheptopyranoses 20 and 24.
An alternative Wittig olefination with Ph3P=CHOBn (22) was then carried out using again
aldehyde 17. Ylide 22 was obtained via deprotonation of the benzyloxytriphenylphosphonium
salt 21 with nꢀbutyllithium in THF. After stirring overnight, the desired benzylic enol ether 23
was formed and isolated in ~60% yield as a mixture of ~1:1 E/Z isomers (Scheme 4).
Previously, Streicher et al25 reported an analogous reaction using Ph3P=CHOMe as an ylide to
afford the corresponding alkenes in a similar yield. The two geometric isomers (23E and 23Z)
had very similar polarity. Alternatively, the E/Z mixture of benzylic enol ether 23 could be
directly subjected to catalytic hydrogenation in methanol using Pearlman’s catalyst; after 24
ACS Paragon Plus Environment

Page 13 of 45
The Journal of Organic Chemistry
1
2
hours, the desired 6ꢀdeoxyꢀ1,2:3,4ꢀdiꢀOꢀisopropylideneꢀαꢀD
ꢀgalactoꢀheptopyranose 2025 was
3
4
5
6
7
8
1
obtained and isolated in 75% yield. The H NMR spectrum of compound 20 showed two sets of
multiplets at 2.00 and 1.80 ppm which are characteristic for the shielded Hꢀ6a and Hꢀ6b protons,
as well as a second set of diastereotopic protons in a narrow region (3.74ꢀ3.88 ppm) which
correspond to Hꢀ7a and Hꢀ7b. In the 1Hꢀ13C HSQC correlation spectrum, compound 20 revealed
a carbon signal (Cꢀ6) at 32.8 ppm which correlated to both Hꢀ6a and Hꢀ6b, confirming the 6ꢀ
deoxy functionality. In addition, a more deshielded carbon signal (Cꢀ7) at 59.9 ppm correlated to
both Hꢀ7a and Hꢀ7b, confirming the presence of a hydroxyl group at Cꢀ7.
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
During the catalytic hydrogenation step, since two different functionalities (alkene and Oꢀ
benzyl) were reduced concomitantly, we decided to monitor the reaction progress in order to
determine whether a regioselective manipulation of the functional groups would be possible.
Indeed, after subjecting the mixture to the same reaction conditions as above for only one hour,
we observed the formation of a major product with an Rf of 0.64 (20% EtOAc−hexanes). The
new compound was only slightly less polar than the starting materials (Rf = 0.62, 20%
EtOAc−hexanes), and after stopping the reaction by a simple filtration, the newly formed
1
product was isolated in 96% yield. To our delight, the H NMR spectrum of the new product
proved to be compound 24 (Scheme 4), which still had the characteristic benzylic protons at 4.55
and 4.49 ppm as two doublets (J = 11.8 Hz). In addition, two sets of proton signals were
observed in a highly shielded region, 1.97 ppm (dddd, 1H, J = 13.9, 9.5, 5.3, 4.9 Hz, Hꢀ6a) and
1.90 ppm (dddd, 1H, J = 14.2, 9.8, 5.3, 4.9 Hz, Hꢀ6b), which corresponded to the diastereotopic
protons Hꢀ6a and Hꢀ6b of compound 24. As would be expected, the E/Z double bond of the
benzylic enol ethers 23 preferentially reacted under hydrogenolysis conditions before cleavage of
the Oꢀbenzyl group could occur. However, when this reaction was performed on a larger scale (>
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 45
1
2
3
4
2 g), about 20% of product 20 was still formed but could be easily Oꢀbenzylated again using
5
6
standard alkylation conditions.
7
8
9
These encouraging results obtained during the 6ꢀhomologation of
D
ꢀgalactose prompted us 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
design a synthetic route to the desired 6ꢀdeoxyꢀβꢀ
Dꢀgalactoꢀheptopyranoside 29 from compound
24 (Scheme 5).
Scheme 5. Synthesis of 6ꢀdeoxyꢀDꢀgalactoꢀheptopyranosyl imidate donor 27 and subsequent
conversion to the corresponding heptopyranosides 29 and 30.
The synthesis began with removal of the two isopropylidene groups of compound 24 in 50%
acetic acid ꢀ water at 80 °C for 8 hours; the solution was then evaporated and treated with acetic
anhydrideꢀpyridine to afford fully acetylated intermediate 25, isolated as an α/βꢀmixture in 93%
yield over two steps. In order to activate the anomeric position for glycosylation, the anomeric
Oꢀacetate of compound 25 was regioselectively removed using hydrazinium acetate30 in DMF
(85% yield), and the obtained hemiacetal 26 was then converted to the desired Oꢀ
trichloroacetimidate 27 by reaction with trichloroacetonitrile in dichloromethane using
ACS Paragon Plus Environment

Page 15 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
anhydrous potassium carbonate as base. Since the imidate functionality is often sensitive to the
mild acidity of silica gel, the crude reaction mixture was instead subjected to a quick workꢀup
with extraction using ethyl acetate and brine, and then coꢀevaporated with toluene. This afforded
the crude imidate 27 which was used directly for subsequent glycosylation with the previously
reported 6ꢀazidohexanol (28). 31 The alcohol 28 was used in excess (5 equiv.) and the
glycosylation carried out at ꢀ40 °C in anhydrous dichloromethane using trimethylsilyl
trifluoromethanesulfonate (TMSOTf) as a promoter in the presence of 4 Å molecular sieves;32,33
after 4 hours, the reaction solution was warmed to ambient temperature. Surprisingly, 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
expected 6ꢀazidohexyl 2,3,4ꢀtriꢀOꢀacetylꢀ7ꢀOꢀbenzylꢀ6ꢀdeoxyꢀβꢀDꢀgalactoꢀheptopyranoside 29
was not isolated, but instead an analog (30) which had lost the 2ꢀOꢀacetyl group was obtained. A
1Hꢀ1H COSY NMR experiment (Figure 2) confirmed the structure of the isolated product: for
example, the anomeric proton was observed as a doublet at 4.29 ppm with a large coupling
constant (J1,2 = 7.7 Hz), confirming the βꢀanomeric configuration. In addition, the Hꢀ2 proton
was observed at 3.78 ppm as a doublet of doublets, with the lower chemical shift supporting the
absence of the expected acetyl group at Oꢀ2. On the other hand, the Hꢀ3 and Hꢀ4 protons were
observed at 4.96 and 5.30 ppm, respectively; both protons are significantly more deshielded than
Hꢀ2, confirming the presence of an acetyl group at both Oꢀ3 and Oꢀ4. Furthermore, a very weak
correlation peak was observed between Hꢀ2 and the broad peak at 2.30 ppm, which was assigned
to be the hydroxyl attached to Cꢀ2. This unexpected result was indeed very beneficial to our
synthesis of the final 6ꢀdeoxyꢀβꢀDꢀidoꢀheptopyranoside target, as the free alcohol at Cꢀ2 could be
selectively activated to generate a good leaving group, which could facilitate formation of the
key 2,3ꢀanhydro intermediate (see later).
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 45
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
Figure 2. 1Hꢀ1H gCOSY spectrum of 6ꢀazidohexyl 3,4ꢀdiꢀOꢀacetylꢀ7ꢀOꢀbenzylꢀ6ꢀdeoxyꢀβꢀ
galactoꢀheptopyranoside 30 (CDCl3, 400 MHz, 298 K).
Dꢀ
The reason for the regioselective loss of the 2ꢀOꢀacetyl group could be explained by Scheme
6. As illustrated, after activation by TMSOTf, imidate 27 should be converted to the reactive
oxocarbenium ion, which could be stabilized by participation of the Oꢀ2ꢀacetyl group to form a
bicyclic oxonium intermediate. Alcohol 28 could attack either the anomeric carbon from the top
face (Path A) to form expected compound 29, or the electrophilic C atom of the Oꢀ2ꢀacetate
(Path B) which would lead to the formation of an orthoester. After a proton transfer to the
anomeric oxygen, a second molecule of alcohol 28 could attack from the top face, resulting in
ACS Paragon Plus Environment

Page 17 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
the formation of the βꢀglycoside with an orthoacetic acid functionality which would subsequently
be hydrolyzed to release the free OHꢀ2 group and afford observed compound 30. Previously,
Nitz et al34 also reported a similar 2ꢀOꢀdeacetylation when an orthoester was used as a glycosyl
donor in the presence of excess allyl alcohol acceptor.
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 6. A proposed pathway for the formation of 6ꢀdeoxyꢀβꢀ
and 30.
Dꢀgalactoꢀheptopyranosides 29
It is worth noting that the outcome of the glycosylation reaction varied when different
amounts of alcohol (28) were used. For example, when 1.5 equivalents of alcohol (28) were used
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 45
1
2
3
4
5
6
during glycosylation with imidate 27, both the expected 2,3,4ꢀtriꢀOꢀacetylated product 29 and the
2ꢀOꢀdeacetylated product 30 were obtained in a 1.7:1 ratio.
7
8
9
With the unexpected availability of glycoside 30, we proceeded to activate the free OHꢀ2 via
mesylation using mesyl chloride in anhydrous pyridine, and the desired 2ꢀOꢀmesylate 31 was
isolated in 91% yield (Scheme 7). Compound 31 could then be directly converted into the 6ꢀ
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
deoxyꢀβꢀDꢀidoꢀheptopyranoside 34 in a single step, by dissolving the mesylate 31 into benzyl
alcohol and treating with potassium tertꢀbutoxide: this resulted in a proton exchange between
benzyl alcohol and tertꢀbutoxide, affording the reactive benzyloxide which catalyzed an Oꢀ
transesterification to remove the Oꢀ3ꢀ and Oꢀ4ꢀacetates (ꢀ32). The free Cꢀ3ꢀoxide subsequently
carried out an intramolecular attack of the Oꢀ2ꢀmesylate, to form the desired 2,3ꢀanhydroꢀ6ꢀ
deoxyꢀβꢀ
subjected to nucleophilic attack by the benzyloxide, to form the desired 3,7ꢀdiꢀOꢀbenzylated 6ꢀ
deoxyꢀβꢀ ꢀidoꢀheptopyranoside 34. Using 5 equivalents of potassium tertꢀbutoxide and stirring
Dꢀtalopyranoside intermediate (33). Finally, the formed 2,3ꢀanhydro functionality was
D
at 60 °C for 10 hours successfully afforded the desired compound 34 in ~60% isolated yield. If
the reaction was carried out at a lower temperature and for a shorter time, the key intermediates
could instead be isolated. For example, when the reaction was carried out at ambient temperature
for 1 hour, the diol 32 was isolated in 75% yield; if the reaction was quenched after 12 hours at
ambient temperature, the 2,3ꢀanhydro intermediate 33 was isolated in 86% yield. Apparently,
nucleophilic opening of the anhydro functionality of compound 33 by benzyloxide proceeded
very slowly at ambient temperature but could be significantly accelerated at higher temperatures
(60 °C), and indeed the reaction was very regioselective to afford desired compound 34 as the
only regioisomer isolated.35
ACS Paragon Plus Environment

Page 19 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
In a similar manner, we also carried out the conversion of Oꢀ2ꢀmesylate 31 to the related 3ꢀ
Oꢀallylated analog 35 in 73% yield using allyl alcohol as the solvent and nucleophile under
similar conditions (Scheme 7).
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 7. Formation of 6ꢀdeoxyꢀβꢀDꢀidoꢀheptopyranoside 2.
Success in constructing the 6ꢀdeoxyꢀβꢀ
Dꢀidoꢀheptopyranosyl configurations of compounds 34
and 35 was confirmed by NMR experiments based on their correlation to the unique 3J coupling
constants associated with the stereochemistry of βꢀDꢀidoꢀheptopyranosides. For example, for
compound 34, the anomeric proton was observed at 4.68 ppm as a doublet with a very small
coupling constant (J1,2 = 0.7 Hz); Hꢀ3 was observed at 3.91 ppm as a doublet of doublets again
with small coupling constants (J2,3 ≈ J3,4 ≈ 2.6 Hz), and Hꢀ5 was observed at 4.03 ppm as a
doublet of doublets of doublets (J4,5 < 1 Hz, J5,6a = 5.1 Hz, J5,6b = 8.2 Hz). The observation of
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 45
1
2
3
4
5
6
7
8
9
small coupling constants between all adjacent protons (J1,2, J2,3, J3,4, J4,5) on the pyranose ring is
consistent with the small dihedral angles (~60°) between adjacent protons, as expected in the 4C1
idoꢀconfiguration of the pyranoside. On the other hand, the Hꢀ2 and Hꢀ4 protons were observed
at 3.81 ppm and 3.53 ppm, respectively; analysis of their coupling patterns again afforded small
coupling constants (<4.2 Hz) to neighbouring ring protons, as well as coupling to an attached
hydroxyl group (J2, OHꢀ2 = 2.0 Hz; J4, OHꢀ4 could not be accurately measured due to overlapping
with OCHa of the aglycone chain).
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
Τhe 3,7ꢀdiꢀOꢀbenzylated 6ꢀdeoxyꢀβꢀDꢀidoꢀheptopyranoside 34 was finally subjected to
hydrogenolysis to remove the benzyl protecting groups and concomitantly reduce the azide
functionality on the aglycone. The reaction was initially conducted in methanol using Pearlman’s
catalyst. However, partial Nꢀmethylations were observed to afford the desired compound 2 along
with some Nꢀmethylated byꢀproducts. The Nꢀmethyl groups could be observed in the NMR
spectrum of the crude products: for example, in the 1H NMR spectrum, a singlet at 2.70 ppm and
another singlet at 2.88 ppm were observed, which would correspond to Nꢀmethyl groups in the
byꢀproducts. Electrosprayꢀquadrupole time of flight high resolution mass spectrometry (ESI/Qꢀ
TOF HRMS) further confirmed the presence of Nꢀmethylated products: we observed peaks at
m/z 294.1908, 308.2065, and 322.2218, which correspond respectively to the calculated m/z of
desired product 2 (C13H28NO6 [M + H]+: 294.1911), and the monoꢀ and diꢀNꢀmethylated
products (C14H30NO6 [M + H]+: 308.1995; C15H32NO6 [M + H]+: 322.2155). These partial Nꢀ
methylations have also been reported recently during the catalytic hydrogenation of another
family of oligosaccharideꢀazides.36 Unfortunately, it was difficult to separate the partially Nꢀ
methylated byꢀproducts from desired compound 2 which prevented them from being
characterized fully.
ACS Paragon Plus Environment

Page 21 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
As an alternative, a Birch reduction was adopted to concomitantly remove the benzyl groups
and reduce the azide37,38 (Scheme 7). The Birch reduction was carried out at ꢀ78 °C in liquid
ammoniaꢀTHF using sodium metal as the reducing reagent. After one hour, the desired
compound 2 was isolated as an acetic acid salt using gel filtration followed by chromatography
on reverseꢀphase C18 silica gel using a gradient of 0ꢀ20% of methanol in water as the eluent
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
1
(containing 0.2% acetic acid). Figure 3 shows the Hꢀ13C HSQC NMR spectrum of the isolated
compound 2 in D2O. The anomeric proton (Hꢀ1) was observed at 4.72 ppm and is correlated to
an anomeric carbon at 99.0 ppm. The other protons of the pyranoside ring, Hꢀ2, Hꢀ3, Hꢀ4 and Hꢀ
5 protons are observed at 3.61, 3.96, 3.43 and 3.95 ppm, respectively, and they correlate with
their attached carbons at 69.4 (Cꢀ2), 69.8 (Cꢀ3), 69.8 (Cꢀ4) and 71.1 (Cꢀ5) ppm, respectively.
The observed vicinal coupling constants were consistently small (J1,2 = ~1 Hz, J2,3 = 3.5 Hz, J3,4
= 3.5 Hz, J4,5 = 1.9 Hz), confirming the βꢀ(1→2)ꢀcisꢀanomeric configuration as well as the idoꢀ
configuration. The two diastereotopic protons Hꢀ6a and Hꢀ6b were found in a highly shielded
region (1.91 and 1.70 ppm, respectively), confirming the 6ꢀdeoxy functionality. Finally, HRMS
(ESI/QꢀTOF) also revealed a peak at m/z 294.1908 which corresponds to the calculated m/z
294.1911 for the expected formula of compound 2 C13H28NO6 [M + H]+.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 45
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
Figure 3. 1Hꢀ13C gHSQC spectrum of 6ꢀdeoxyꢀβꢀ
Dꢀidoꢀheptopyranoside 2 (D2O, 400 MHz, 298
K).
3. Conclusions
In conclusion, two distinct strategies have been studied to progress toward total synthesis of
the challenging 6ꢀdeoxyꢀβꢀ ꢀidoꢀheptopyranoside related to C. jejuni HS:4 CPS. The first
strategy involved establishment of the βꢀ ꢀidopyranosyl configuration from βꢀ
galactopyranosides prior to extending the ꢀhexose to the desired 6ꢀdeoxyꢀ ꢀheptose via Cꢀ6ꢀ
homologation. However, this approach was found to be impractical, as the Cꢀ6ꢀposition of the βꢀ
D
D
Dꢀ
D
D
D
ꢀidopyranosides was observed to exhibit significantly reduced reactivity, probably caused by an
ACS Paragon Plus Environment

Page 23 of 45
The Journal of Organic Chemistry
1
2
3
4
increased steric hindrance of the 6ꢀposition which is exacerbated by enhanced conformational
5
6
7
8
flexibility of the βꢀ
homologation on the less flexible
heptopyranoside, followed by establishment of the βꢀ
heptopyranoside 2. This strategy was found to be very productive, allowing for the successful
synthesis of a 6ꢀdeoxyꢀβꢀ ꢀidoꢀheptopyranoside (2) for the first time. In both strategies, the
conversion of βꢀ ꢀgalactoꢀconfiguration to βꢀ ꢀidoꢀconfiguration was achieved by a double
inversion at the Cꢀ2 and Cꢀ3 positions via a 2,3ꢀanhydroꢀβꢀ ꢀtaloꢀpyranosyl intermediate;
D
ꢀidopyranosyl ring. The second strategy involved carrying out a 6ꢀ
ꢀgalactopyranose to first obtain a 6ꢀdeoxyꢀβꢀ ꢀgalactoꢀ
ꢀidoꢀconfiguration in target
D
D
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
D
D
D
D
D
opening of the 2,3ꢀanhydroꢀfunctionality by an alkoxide was found to be very regioꢀ and
stereoselective, with nucleophilic attack occurring exclusively at the Cꢀ3 position. The overall
transformation of stereochemistry was proven to be very efficient, and required minimal
purification. Additionally, our strategy to synthesize βꢀDꢀidopyranosides from βꢀDꢀ
galactopyranosides is very versatile, as it can be successfully carried out on both hexoses and
heptoses. This success in synthesis of the 6ꢀdeoxyꢀβꢀDꢀidoꢀheptopyranoside (2) should permit us
to study its immunological properties in the future.
4. Experimental section
General Methods. All commercial reagents were used as supplied unless otherwise stated.
Thin layer chromatography was performed on Silica Gel 60ꢀF254 (E. Merck, Darmstadt) with
detection by fluorescence, charring with 5% H2SO4 (aq), or a ceric ammonium molybdate
solution. Column chromatography was performed on Silica Gel 60 (Silicycle, Ontario) and
solvent gradients given refer to stepped gradients and concentrations are reported as % v/v.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 45
1
2
3
4
5
6
7
8
9
Organic solutions were concentrated and/or evaporated to dry under vacuum in a water bath (<60
°C). Molecular sieves were stored in an oven at 100° C and flameꢀdried under vacuum before
use. Amberlite IRꢀ120 (H+) ion exchange resin was washed multiple times with MeOH prior to
use. All pH values were determined using universal pH paper. Optical rotations were determined
in a 5 cm cell at 20 ± 2 °C; [α]D20 values are given in units of 10ꢀ1 deg•cm2/g. NMR spectra were
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
1
13
recorded on Bruker spectrometers at 400 MHz, and the firstꢀorder chemical shifts of H and C
(DEPTꢀQ) are reported in δ (ppm) and referenced to residual CHCl3 (δH 7.24, δC 77.23, CDCl3),
1
or residual HDO (δH 4.79, external acetone δC 29.9, D2O). H and 13C NMR spectra were
assigned with the assistance of 2D gCOSY and 2D gHSQC experiments. Highꢀresolution ESIꢀ
QTOF mass spectra were recorded on an Agilent 6520 Accurate Mass Quadrupole Timeꢀofꢀ
Flight LC/MS spectrometer. Elemental analyses were obtained with a Perkin Elmer Series II
2400 CHNS/O Analyzer. All the data were obtained with the assistance of the analytical services
of the Department of Chemistry, University of Calgary.
Methyl 3ꢀOꢀallylꢀ2ꢀOꢀbenzylꢀ6ꢀOꢀtosylꢀβꢀDꢀidopyranoside (7): Compound 6 (481 mg, 1.48
mmol) and pꢀTsCl (318 mg, 1.67 mmol) were dissolved into dry pyridine (5.0 mL) and left
mixing at rt. After 6 h additional pꢀTsCl (43 mg, 0.23 mmol) was added. After 20 h, the reaction
was quenched with MeOH (5 mL), then evaporated to dry via coꢀevaporation with toluene (3 × 5
mL). The crude product was reꢀdissolved into EtOAc (100 mL), washed with saturated NaCl (aq)
solution (3 × 100 mL), dried with Na2SO4, filtered, and evaporated to dry. The crude mixture
was purified by column chromatography on silica gel using 25% acetone – hexanes to afford the
desired product 7 as a colourless syrup (423 mg, 0.884 mmol, 60% yield), as well as a small
amount of diꢀtosylated byꢀproduct. Rf 0.66 (EtOAc : hexanes 2:3). [α]D20: ꢀ72.4° (c 1.02, CHCl3).
ACS Paragon Plus Environment

Page 25 of 45
The Journal of Organic Chemistry
1
2
1H NMR (CDCl3, 400 MHz): δH 7.79 – 7.74 (m, 2H, Ar), 7.31 – 7.24 (m, 7H, Ar), 5.72 (dddd,
1H, J = 17.2, 10.4, 5.6, 5.6 Hz, OCH2CH=CH2), 5.13 (dddd, 1H, J = 17.2, 1.6, 1.6, 1.6 Hz,
OCH2CH=CHaHb), 5.11 (dddd, 1H, J = 10.4, 1.5, 1.3, 1.3 Hz, OCH2CH=CHaHb), 4.83 (d, 1H, J
= 12.1 Hz, PhCHaHb), 4.58 (d, 1H, J = 12.1 Hz, PhCHaHb), 4.57 (d, 1H, J = 1.0 Hz, Hꢀ1), 4.24
(dd, 1H, J = 10.3, 5.1 Hz, Hꢀ6a), 4.20 (dd, 1H, J = 10.3, 6.9 Hz, Hꢀ6b), 4.02 (ddd, 1H, J = 6.7,
5.2, <1 Hz, Hꢀ5), 3.90 (dddd, 1H, J = 12.9, 5.6, 1.4, 1.4 Hz, OCHaHbCH=CH2), 3.86 (dddd, 1H,
J = 12.9, 5.7, 1.4, 1.4 Hz, OCHaHbCH=CH2), 3.64 – 3.62 (m, 1H, Hꢀ3), 3.55 (dd, 1H, J = 3.4, 0.8
Hz, Hꢀ2), 3.50 – 3.48 (m, 4H, OMe and Hꢀ4), 3.42 (broad, 1H, 4ꢀOH), 2.40 (s, 3H, Ts). 13C
NMR (CDCl3, 100 MHz): δC 144.9 (Ar), 137.4 (Ar), 133.8 (OCH2CH=CH2), 132.9 (Ar), 129.9
(Ar), 128.5 (Ar), 128.2 (Ar), 128.14 (Ar), 128.06 (Ar), 117.9 (OCH2CH=CH2), 100.8 (Cꢀ1), 75.1
(Cꢀ3), 74.4 (PhCH2), 74.3 (Cꢀ2), 73.0 (Cꢀ5), 71.2 (OCH2CH=CH2), 69.7 (Cꢀ6), 66.6 (Cꢀ4), 57.2
(OMe), 21.7 (Ts). HRMS (ESIꢀQTOF) m/z: [M + Na]+ calcd for C24H30O8SNa 501.1554, found
501.1551.
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
Methyl 3ꢀOꢀallylꢀ2ꢀOꢀbenzylꢀ6ꢀdeoxyꢀ6ꢀiodoꢀβꢀDꢀidopyranoside (8): A solution of compound
6 (2.078 g, 6.405 mmol), PPh3 (2.196 g, 8.374 mmol), and imidazole (0.800 g, 11.8 mmol) in dry
CH2Cl2 (20 mL) was flushed with Ar, and then cooled to 0 °C in an iceꢀH2O bath. I2 (2.649 g,
10.44 mmol) was added in one portion, and the reaction mixture was slowly warmed back to
ambient temperature. After 24 h, a fresh solution of PPh3 (1.306 g, 4.978 mmol) and I2 (1.350 g,
5.321 mmol) in dry CH2Cl2 (10 mL) was added to the reaction mixture. After an additional 24 h,
the solution was evaporated to dry and then redissolved into EtOAc (120 mL). The organic phase
was washed with saturated Na2S2O3 (aq) solution (4 × 120 mL), dried with Na2SO4, filtered, and
evaporated to dry. The crude material was then purified via column chromatography on silica
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 45
1
2
3
4
5
6
7
8
9
using 8→15→20% acetone – hexanes to afford the pure product 8 as a yellow syrup (721 mg,
1.66 mmol, 26% yield), unreacted starting material (555 mg, 1.71 mmol, 27% recovered). Rf
0.85 (EtOAc : toluene 6:4). [α]D20: ꢀ44° (c 0.86, CHCl3). 1H NMR (CDCl3, 400 MHz): δH 7.34 –
7.28 (m, 5H, Ar), 5.76 (dddd, 1H, J = 17.2, 10.4, 5.7, 5.7 Hz, OCH2CH=CH2), 5.16 (dddd, 1H, J
= 17.2, 1.6, 1.6, 1.6 Hz, OCH2CH=CHaHb), 5.13 (dddd, 1H, J = 10.4, 1.4, 1.4, 1.4 Hz,
OCH2CH=CHaHb), 4.88 (d, 1H, J = 12.2 Hz, PhCHaHb), 4.62 (d, 1H, J = 12.2 Hz, PhCHaHb),
4.61 (d, 1H, J = 1.0 Hz, Hꢀ1), 3.94 – 3.90 (m, 3H, OCH2CH=CH2 and Hꢀ5), 3.70 – 3.66 (m, 2H,
Hꢀ3 and Hꢀ4), 3.60 (s, 3H, OMe), 3.53 (ddd, 1H, J = 3.1, 1.2, 1.2 Hz, Hꢀ2), 3.48 (dd, 1H, J =
12.6, 1.0 Hz, 4ꢀOH), 3.40 (dd, 1H, J = 10.2, 7.7 Hz, Hꢀ6a), 3.37 (dd, 1H, J = 10.2, 6.2 Hz, Hꢀ6b).
13C NMR (CDCl3, 100 MHz): δC 137.6 (Ar), 134.0 (OCH2CH=CH2), 128.7 (Ar), 128.34 (Ar),
128.26 (Ar), 118.1 (OCH2CH=CH2), 101.5 (Cꢀ1), 76.0 (Cꢀ5), 75.9 (Cꢀ3), 74.6 (PhCH2), 74.3 (Cꢀ
2), 71.4 (OCH2CH=CH2), 67.8 (Cꢀ4), 57.4 (OMe), 4.0 (Cꢀ6). HRMS (ESIꢀQTOF) m/z: [M +
Na]+ calcd for C17H23IO5Na 457.0482, found 457.0479.
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
Methyl 3ꢀOꢀallylꢀ2ꢀOꢀbenzylꢀ6ꢀcyanoꢀ6ꢀdeoxyꢀβꢀDꢀidopyranoside (9): Method A: A solution
of the 6ꢀOꢀtosyl starting material 7 (121 mg, 0.252 mmol), KCN (76 mg, 1.2 mmol), and 18ꢀ
crownꢀ6 (299 mg, 1.13 mmol) in dry DMF (1.3 mL) were left mixing at rt. After 48 h, the
mixture was temperature was increased to 60 °C. After 24 h of heating, the reaction mixture was
evaporated to dry, redissolved into EtOAc (50 mL), washed with saturated NaCl (aq) solution (3
× 50 mL), dried with Na2SO4, filtered, and evaporated to dry. The crude material was purified
via column chromatography on silica using 15→25→40% EtOAc – hexanes to afford the desired
product 9 as a colourless syrup (11 mg, 0.032 mmol, 13% yield), the hydrolyzed byꢀproduct 6
ACS Paragon Plus Environment

Page 27 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
(16 mg, 0.048 mmol, 19% yield). Attempts at using Bu4NI instead of 18ꢀcrownꢀ6 were also
unsuccessful, and resulted in significant formation of the hydrolyzed byꢀproduct 6.
7
8
9
Method B: Compound 8 (113 mg, 0.260 mmol) and KCN (59 mg, 0.91 mmol) were
dissolved into dry DMF (1.3 mL) and left mixing at 80 °C. After 48 h, the reaction mixture was
diluted with EtOAc (60 mL), washed with saturated NaCl (aq) solution (3 × 60 mL), dried with
Na2SO4, filtered, and evaporated to dry. The crude material was purified via column
chromatography on silica using 10→15→20% EtOAc – hexanes to afford the pure product as a
pale yellow syrup (34 mg, 0.10 mmol, 39% yield). Attempts to leave the reaction for longer or at
increased temperatures resulted in the formation of a number of uncharacterized byꢀproducts. Rf
0.56 (EtOAc : hexanes 2:3). [α]D20: ꢀ50° (c 0.94, CHCl3). 1H NMR (CDCl3, 400 MHz): δH 7.35 –
7.28 (m, 5H, Ar), 5.75 (dddd, 1H, J = 17.0, 10.5, 5.7, 5.7 Hz, OCH2CH=CH2), 5.15 (dddd, 1H, J
= 17.1, 1.5, 1.5, 1.5 Hz, OCH2CH=CHaHb), 5.14 (dddd, 1H, J = 10.5, 1.3, 1.3, 1.3 Hz,
OCH2CH=CHaHb), 4.88 (d, 1H, J = 12.1 Hz, PhCHaHb), 4.65 (d, 1H, J = 0.9 Hz, Hꢀ1), 4.62 (d,
1H, J = 12.1 Hz, PhCHaHb), 4.08 (ddd, 1H, J = 8.6, 5.5, 1.3 Hz, Hꢀ5), 3.92 (dddd, 2H, J = 5.7,
1.4, 1.4, <1 Hz, OCH2CH=CH2), 3.71 (dd, 1H, J = 3.3, 3.3 Hz, Hꢀ3), 3.59 (ddd, 1H, J = 3.4, 1.0,
1.0 Hz, Hꢀ2), 3.58 (s, 3H, OMe), 3.58 (d, 1H, J = 11.7 Hz, 4ꢀOH), 3.46 (dddd, 1H, J = 11.6, 3.1,
1.2, 1.2 Hz, Hꢀ4), 2.86 (dd, 1H, J = 16.8, 8.6 Hz, Hꢀ6a), 2.64 (dd, 1H, J = 16.8, 5.5 Hz, Hꢀ6b).
13C NMR (CDCl3, 100 MHz): δC 137.4 (Ar), 133.8 (OCH2CH=CH2), 128.7 (Ar), 128.4 (Ar),
118.3 (OCH2CH=CH2), 117.8 (CN), 101.2 (Cꢀ1), 75.3 (Cꢀ3), 74.7 (PhCH2), 74.0 (Cꢀ2), 71.6
(OCH2CH=CH2), 71.3 (Cꢀ5), 67.6 (Cꢀ4), 57.4 (OMe), 20.0 (Cꢀ6). HRMS (ESIꢀQTOF) m/z: [M +
NH4]+ calcd for C18H27N2O5 351.1914, found 351.1917.
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
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 45
1
2
3
4
Methyl 2,3ꢀdiꢀOꢀbenzylꢀ6ꢀdeoxyꢀ6ꢀiodoꢀβꢀ
D
ꢀidopyranoside (11) and 1,6ꢀanhydroꢀ2,3ꢀdiꢀOꢀ
5
6
benzylꢀβꢀ
D
ꢀidopyranose (12): A solution of the compound 10 (152 mg, 0.407 mmol), PPh3 (129
7
8
9
mg, 0.491 mmol), and imidazole (62 mg, 0.91 mmol) in dry CH2Cl2 (1.5 mL) was flushed with
Ar, and then cooled to 0 °C. I2 (170 mg, 0.671 mmol) was added in one portion, and the reaction
mixture slowly warmed back to ambient temperature. After 24 h, a solution of PPh3 (218 mg,
0.832 mmol) and I2 (214 mg, 0.844 mmol) in CH2Cl2 (0.5 mL) was added at 0 °C and then the
solution warmed back to ambient temperature. After another 24 h, the solution was evaporated to
dry and then redissolved into EtOAc (60 mL). The organic phase was washed with saturated
Na2S2O3 (aq) solution (2 × 60 mL), saturated NaCl (aq) solution (2 × 60 mL), dried with Na2SO4,
filtered, and evaporated to dry. The crude material was then purified via column chromatography
on silica using 10→15→20% acetone – hexanes to afford the pure product 11 as a yellow syrup
(124 mg, 0.257 mmol, 63% yield), the 1,6ꢀanhydro byꢀproduct 12 (36 mg, 0.11 mmol, 26%
yield). Data for 11: Rf 0.70 (EtOAc : toluene 3:2). [α]D20: ꢀ64° (c 0.96, CHCl3). 1H NMR (CDCl3,
400 MHz): δH 7.36 – 7.27 (m, 8H, Ar), 7.23 – 7.19 (m, 2H, Ar), 4.85 (d, 1H, J = 12.2 Hz,
PhCHaHb), 4.67 (d, 1H, J = 0.9 Hz, Hꢀ1), 4.61 (d, 1H, J = 12.2 Hz, PhCHaHb), 4.49 (d, 1H, J =
11.9 Hz, PhCHaHb), 4.45 (d, 1H, J = 11.9 Hz, PhCHaHb), 3.98 (ddd, 1H, J = 7.6, 6.3, 0.9 Hz, Hꢀ
5), 3.78 (dd, 1H, J = 3.2, 3.2 Hz, Hꢀ3), 3.74 (dddd, 1H, J = 11.5, 3.1, 1.1, 1.1 Hz, Hꢀ4), 3.61 (s,
3H, OMe), 3.57 (ddd, 1H, J = 3.2, 1.1, 1.0 Hz, Hꢀ2), 3.50 (d, 1H, J = 11.6 Hz, 4ꢀOH), 3.42 (dd,
1H, J = 10.1, 7.7 Hz, Hꢀ6a), 3.39 (dd, 1H, J = 10.2, 6.2 Hz, Hꢀ6b). 13C NMR (CDCl3, 100 MHz):
δC 137.5 (Ar), 137.4 (Ar), 128.7 (Ar), 128.6 (Ar), 128.22 (Ar), 128.16 (Ar), 127.8 (Ar), 101.4
(Cꢀ1), 75.92 (Cꢀ3), 75.88 (Cꢀ5), 74.4 (PhCH2), 74.1 (Cꢀ2), 72.4 (PhCH2), 67.6 (Cꢀ4), 57.2
(OMe), 3.9 (Cꢀ6). HRMS (ESIꢀQTOF) m/z: [M + Na]+ calcd for C21H25IO5Na 507.0639, found
507.0642. Data for 12: 1H NMR (CDCl3, 400 MHz): δH 7.37 – 7.26 (m, 10H, Ar), 5.32 (d, 1H, J
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
ACS Paragon Plus Environment

Page 29 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
= 1.5 Hz, Hꢀ1), 4.91 (d, 1H, J = 11.6 Hz, PhCHaHb), 4.69 (d, 1H, J = 11.6 Hz, PhCHaHb), 4.68
(s, 2H, PhCH2), 4.36 (ddd, 1H, J = 4.6, 4.6, <1 Hz, Hꢀ5), 4.04 (dd, 1H, J = 7.7, <1 Hz, Hꢀ6a),
3.82 – 3.78 (m, 1H, Hꢀ4), 3.68 (dd, 1H, J = 7.7, 5.0 Hz, Hꢀ6b), 3.60 (dd, 1H, J = 8.1, 8.1 Hz, Hꢀ
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
13
3), 3.49 (dd, 1H, J = 8.0, 1.6 Hz, Hꢀ2), 2.08 (d, 1H, J = 2.6 Hz, 4ꢀOH). C NMR (CDCl3, 100
MHz): δC 138.8 (Ar), 138.1 (Ar), 128.8 (Ar), 128.7 (Ar), 128.21 (Ar), 128.16 (Ar), 128.15 (Ar),
128.1 (Ar), 99.8 (Cꢀ1), 82.8 (Cꢀ2), 82.5 (Cꢀ3), 75.24 (PhCH2), 75.20 (Cꢀ5), 73.0 (PhCH2), 71.5
(Cꢀ4), 65.5 (Cꢀ6). HRMS (ESIꢀQTOF) m/z: [M + NH4]+ calcd for C20H26NO5 360.1805, found
360.1802.
Methyl 2,3ꢀdiꢀOꢀbenzylꢀ6ꢀcyanoꢀ6ꢀdeoxyꢀβꢀDꢀidopyranoside (13): The starting material (11,
630 mg, 1.30 mmol) and KCN (674 mg, 10.4 mmol) in dry DMF (10 mL) were heated at 80 °C.
After 24 h the reaction was concentrated to half volume via coꢀevaporation with toluene (2 × 10
mL), then diluted with EtOAc (100 mL). The organic phase was washed with saturated NaCl
(aq) solution (3 × 100 mL), H2O (100 mL), dried with Na2SO4, filtered, and evaporated to dry.
The crude material was used directly for the following step.
Methyl 2,3ꢀdiꢀOꢀbenzylꢀβꢀDꢀidoꢀheptopyranoside (15): The crude starting material 13 (1.30
mmol from previous step) was dissolved into dry toluene (20 mL), and cooled to ꢀ72 °C.
DIBALꢀH solution (1 M in toluene, 19 mL, 19 mmol) was added dropwise over 15 min, and
after 2 h the reaction mixture slowly warmed up to ambient temperature. After 24 h, the reaction
mixture was diluted with CH2Cl2 (20 mL) and then quenched with MeOH (30 mL). The crude
mixture was evaporated to dry, then redissolved into EtOAc (100 mL), washed with 1.2 M HCl
ACS Paragon Plus Environment