Integrating ReSET with glycosyl iodide glycosylation in step-economy syntheses of tumor-associated carbohydrate antigens and immunogenic glycolipids

Article abstract of DOI:10.1021/jo402736g

Carbohydrates mediate a wide range of biological processes, and understanding these events and how they might be influenced is a complex undertaking that requires access to pure glycoconjugates. The isolation of sufficient quantities of carbohydrates and glycolipids from biological samples remains a significant challenge that has redirected efforts toward chemical synthesis. However, progress toward complex glycoconjugate total synthesis has been slowed by the need for multiple protection and deprotection steps owing to the large number of similarly reactive hydroxyls in carbohydrates. Two methodologies, regioselective silyl exchange technology (ReSET) and glycosyl iodide glycosylation have now been integrated to streamline the synthesis of the globo series trisaccharides (globotriaose and isoglobotriaose) and α-lactosylceramide (α-LacCer). These glycoconjugates include tumor-associated carbohydrate antigens (TACAs) and immunostimulatory glycolipids that hold promise as immunotherapeutics. Beyond the utility of the step-economy syntheses afforded by this synthetic platform, the studies also reveal a unique electronic interplay between acetate and silyl ether protecting groups. Incorporation of acetates proximal to silyl ethers attenuates their reactivity while reducing undesirable side reactions. This phenomenon can be used to fine-tune the reactivity of silylated/acetylated sugar building blocks.

Full text of DOI:10.1021/jo402736g

Subscriber access provided by UIC Library
Article
Integrating ReSET with Glycosyl Iodide Glycosylation
in Step-Economy Syntheses of Tumor Associated
Carbohydrate Antigens and Immunogenic Glycolipids
Hsiao-Wu Hsieh, Matthew W. Schombs, and Jacquelyn Gervay-Hague
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/jo402736g • Publication Date (Web): 03 Feb 2014
Downloaded from http://pubs.acs.org on February 4, 2014
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 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Integrating ReSET with Glycosyl Iodide Glycosylation in Step-Economy
Syntheses of Tumor Associated Carbohydrate Antigens and Immunogenic
Glycolipids
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
HsiaoꢀWu Hsieh, Matthew W. Schombs, and Jacquelyn GervayꢀHague*
Tel: 1ꢀ530ꢀ754ꢀ9577. Fax: 1ꢀ530ꢀ754ꢀ6915.
jgervayhague@ucdavis.edu
Department of Chemistry, University of California, Davis, One Shields Avenue, Davis,
California 95616, United States
ꢀ
Abstract
HO
HO
OH
O
OH
O
O
Oligosaccharide Extension
O
HO
HN
C17H35
C13H27
HO
HO
O
Regioselective Silyl Exhange Technology
(ReSET)
OH
α-LacCer
HO
HO
TMSO
TMSO
OH
O
OTMS
O
OTMS
O
OH
OH
OH
O
O
OTMS
O
TMSO
HO
TMSO
O
O
TMSO
OH
HO
HO
Glycosyl Iodide Glycosylation
HO
HO
HO
OH
O
Isoglobotriaose
HO
HO
O
OH
OH
O
O
O
Anomeric Functionalization
OH
HO
HO
HO
Globotriaose
Carbohydrates mediate a wide range of biological processes, and understanding these events
and how they might be influenced is a complex undertaking that requires access to pure
glycoconjugates. The isolation of sufficient quantities of carbohydrates and glycolipids from
biological samples remains a significant challenge that has redirected efforts toward chemical
synthesis. However, progress toward complex glycoconjugate total synthesis has been slowed
by the need for multiple protection and deprotection steps owing to the large number of similarly
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 39
1
2
3
4
5
6
7
8
9
reactive hydroxyls in carbohydrates. Two methodologies, Regioselective Silyl Exchange
Technology (ReSET) and glycosyl iodide glycosylation have now been integrated to streamline
the synthesis of the globo series trisaccharides (globotriaose and isoglobotriaose) and αꢀ
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
lactosylceramide (αꢀLacCer). These glycoconjugates include tumorꢀassociated carbohydrate
antigens (TACAs) and immunostimulatory glycolipids that hold promise as immunotherapeutics.
Beyond the utility of the stepꢀeconomy syntheses afforded by this synthetic platform, the studies
also reveal a unique electronic interplay between acetate and silyl ether protecting groups.
Incorporation of acetates proximal to silyl ethers attenuates their reactivity while reducing
undesirable side reactions. This phenomenon can be used to fineꢀtune the reactivity of
silylated/acetylated sugar building blocks.
ꢀ
Introduction
For the past five decades, cancer has been one of the top ten causes of death in the United
States and the incidence rate is gradually increasing.1,2 Cancer treatments and therapeutics to
eradicate the disease are being intensively investigated in an effort to improve quality of life for
those who suffer. Carbohydrateꢀbased vaccine development is one promising approach to this
end.3ꢀ9 Cell membranes are decorated with carbohydrates in the form of glycolipids and
glycoproteins with unique structures and aberrant glycosylation patterns that are correlated with
tumor progression and metastases.10ꢀ14 Vaccines based on the carbohydrate epitopes of these
glycoconjugates are promising therapeutic targets, as are the more recently discovered bacterial
derived glycolipids that stimulate immune response. Limited access to sufficient quantities of
these biomolecules is a discovery roadblock making practical synthesis of complex
carbohydrates a top priority in a recent National Academy of Sciences publication on the future
2
ACS Paragon Plus Environment

Page 3 of 39
The Journal of Organic Chemistry
1
2
of glycoscience.15
3
4
5
6
7
8
One major class of mammalian glycosphingolipids (GSLs) is the globo series tumorꢀ
associated carbohydrate antigens (TACAs).3,4,10,14,16 Globotriaosyl ceramide (Gb3),
isoglobotriaosyl ceramide (iGb3), Gb4, Gb5 and Globo H are the prominent members of this
class (Figure 1). All of these biomolecules share a lactose core, which is diversified by
galactosylation at either the 3’ or 4’ hydroxyls giving rise to iGb3 or Gb3, respectively. Gb3 in
turn is the core structure shared by Globo H and Gb5. Given the centrality of the Gb3 core, its
synthesis has been the focus of numerous investigations. Nicolaou and coꢀworkers were the first
to report the total synthesis of Gb3, and their approach remains one of the most efficient to
date.17,18 A total of eleven steps from commercially available lactose were required to prepare a
protected globotriaose that served as a donor for the ceramide aglycon. Seven of those steps
were focused on orthogonally protecting lactose for 4’ꢀgalactosylation. Similarly, all syntheses
of globotriaose/isoglobotraose that have engaged a lactosyl acceptor have required between 6ꢀ12
steps to prepare the disaccharide building block.16ꢀ29
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 1. Globo series: A class of common TACAs.
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 39
1
2
3
4
5
6
7
8
While βꢀlinked ceramides are known TACAs, bacterial derived αꢀlinked analogs also
stimulate an immune response and hold promise as vaccine adjuvants.30ꢀ33 Isolating these
glycolipids from their natural environments is a challenging process that can lead to
decomposition or even rearrangement of the molecules of interest.34 The lipid composition of
the ceramide is also usually obtained as an inseparable and nonꢀcharacterizable mixture. For
these reasons, chemical synthesis of αꢀlinked ceramides is the primary source of pure and
homogenous materials for immunological studies. To meet this need, our group developed oneꢀ
pot syntheses of KRN7000, BbGLꢀII and αꢀGalCer (Figure 2) and their glucosyl analogs,35,36
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
while the Wang37 and Savage24 groups have reported syntheses of αꢀlactosylceramide, αꢀGb3
and αꢀiGb3. These novel “sugarꢀcapped” CD1d ligands for natural killer T (NKT) cells have
been tested for their ability to stimulate cytokine release. The results established that the αꢀ
linkage is required for immune response and that the oligosaccharide structures serve as
immunomodulators. These findings and biological evaluations of TACAs clearly point to an
emerging area of immunotherapeutic discovery based upon αꢀ and βꢀlinked ceramides; and
access to sufficient quantities of these glycoconjugates is a critical need.
Figure 2. Examples of immunostimulatory glycolipids
4
ACS Paragon Plus Environment

Page 5 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
Recently, the concept of stepꢀeconomy was introduced and has subsequently been
advocated among synthetic chemists.38ꢀ41 Instead of the traditional linear synthesis or tour de
force total synthesis, considerations of semiꢀsynthesis, diversityꢀoriented synthesis, functionꢀ
oriented synthesis and convergent synthesis can lead to high efficiencies in the production of
drug leads. In applying this concept to glycoconjugate syntheses, eliminating the need for
multiple protection/deprotection steps is an obvious starting point. PerꢀOꢀTMS protected
monosaccharides have proven to be especially useful starting materials for stepꢀeconomy
protecting group manipulations.42ꢀ44 We have exploited these readily available compounds in a
process coined Regioselective Silyl Exchange Technology (ReSET).45,46 In just two steps from
free sugars, a wide range of partially acetylated/silylated carbohydrates can be prepared and
readily transformed to either the corresponding glycosyl donor or acceptor (Figure 3). The
introduction of electronꢀwithdrawing acetate groups to silyl sugars affords biꢀfunctional modules
that are more stable than their perꢀOꢀTMS counterparts, yet highly reactive in glycosylation
reactions. The anomeric acetate can be converted into various leaving groups including halides,
thioethers and acetimidates. Efforts reported herein have concentrated on integrating glycosyl
iodide glycosylation with a tandem ReSET strategy to accomplish stepꢀeconomy syntheses of
globo series trisaccharides and αꢀlactosylceramide. The dual methodology platform expands the
current organic synthesis toolbox and provides new insight into the electronic interplay of acetate
and silyl ether functionalities and their influence upon chemical reactivity.
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
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 39
1
2
3
4
5
6
7
8
9
Figure 3. Application of ReSET to generate biꢀfunctional modules for oligosaccharide and
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
glycoconjugate syntheses.
ꢀ
Results and Discussion
Tandem ReSET to prepare biꢀfunctional lactose modules. ReSET begins with perꢀOꢀ
silylated sugars, which undergo selective exchange of silyl ethers for acetate protecting groups.
The reactions are typically run in pyridine with excess acetic anhydride and exchange is
mediated by the addition of acetic acid. Regiocontrol is correlated with acetic acid equiv. and
microwave reaction time. Typically, the reactions afford a mixture of products, all of which are
useful in making a library of different analogs. For example, treating perꢀOꢀsilylated lactose
with 3.0 equiv. of AcOH for 1.25 h leads to production of the diꢀ and triacetylated compounds (1
and 2, Scheme 1) in 20% and 53% yields, respectively. Whereas, increasing the amount of acetic
acid to 7 equiv. forces the production of compound 4 having only one silyl ether at the 4’ position
and perꢀOꢀAc lactose (5) after 3.75 h.46 Since the lactosyl 3’ and 4’ positions are the major
glycosylation sites for globo series antigens, we attempted to establish conditions for preparing
the disilyl ether analog (3) but that compound could not be directly obtained in ReSET reactions
of perꢀOꢀsilylated lactose under any of the conditions evaluated. Previous studies in our lab
indicated that proximal acetate groups facilitate silyl exchange,46 thus we reasoned that greater
regiocontrol might be achieved using compounds 1 and 2 in a tandem process. To our delight,
the reaction proceeded nicely to afford the desired analogs, 3 and 4, in 31% and 25% yields,
6
ACS Paragon Plus Environment

Page 7 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
respectively. With these partially acetylated building blocks (1ꢀ4) in hand, focus shifted to
employing them as either glycosyl donors or acceptors in oligosaccharide and glycoconjugate
syntheses.
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. Tandem ReSET affords differentially protected lactose modules.
ReSET products as glycosyl donors: αꢀlactosylceramide synthesis. Previous attempts to form
and utilize perꢀOꢀTMSꢀlactosyl iodide led to complex reaction mixtures due to glycosidic bond
cleavage and silyl exchange.47 We later discovered that Cꢀ6 acetates protect glycosidic linkages
from TMSI degradation.48,49 Consistent with these findings, we were able to cleanly and
quantitatively generate the lactosyl iodide from both 1 and 2 (Scheme 2A).46 Synthetic ceramide
(7)36,50 was selected as the acceptor in the glycosylation studies due to its biological relevance.
Iodide 6 was reconstituted in dry benzene and cannulated into a mixture of 7,
tetrabutylammonium iodide (TBAI) and diisopropylethylamine (DIPEA). After heating the
reaction to 65 ˚C in anhydrous benzene overnight, the reaction mixture was concentrated and
subjected to acidic methanolysis prior to perꢀOꢀacetylation for characterization purposes
(Scheme 2B). A mixture of the perꢀOꢀacetylated isomers was isolated in 89% yield and the ratio
of primary adduct (8) to secondary adduct (9) was 1.5 to 1.0, respectively, as determined by
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 39
1
2
anomeric proton integration values obtained by 1H NMR.
3
4
5
6
7
8
The in situ anomerization process promoted by TBAI resulted in the exclusive formation
of αꢀlinked glycosides, however multiple flash column chromatography purifications were
required to separate 8 and 9. To alleviate this complication, compound 10 was prepared by
selective desilylation of diꢀOꢀTMS ceramide42 according to a method recently published by Cui
and coꢀworkers.51 Ceramide 10 was then coupled with 6 under the TBAIꢀpromoted conditions
(Scheme 2B). After acidic methanolysis and peracetylation, compound 8 was the only
glycosylation product isolated. The glycosylation yield (61%) was slightly decreased compared
to the glycosylation result of ceramide 7 (89%), possibly due to the result of increased steric
hindrance from the secondary TMS ether. Nevertheless, deacetylation of 8 using NaOMe/MeOH
led to fully deprotected αꢀlactosylceramide (11) in quantitative yield.
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. Synthesis of αꢀlactosyl ceramide.
8
ACS Paragon Plus Environment

Page 9 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
The synthesis of αꢀlactosyl ceramide has several salient features: 1) the glycosyl donor (6)
can be prepared in three steps and 72% overall from free lactose; 2) the lactosyl iodide is reactive
enough to couple with unprotected or partially protected ceramides, yet the interꢀresidue
glycosidic bond is stable enough to survive glycosylation; 3) only the desired αꢀlinkage is
obtained in good yield; 4) only TMS ether and acetate protecting groups are utilized allowing
mild deprotection steps that are compatible with alkene and amide functionality in the ceramide
component. This methodology nicely complements reactions of perꢀOꢀacetylated lactose, which
afford the βꢀanomer due to neighboring group participation of the Cꢀ2 acetate.52
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
ReSET products as acceptors: Globo series trisaccharide syntheses. Having demonstrated
glycosylation efficiency with partially acetylated silyl donors, we set our sights on exploiting
these substrates as glycosyl acceptors. Chemical syntheses of the Globo series TACAs involve
the construction of the crucial αꢀ1,3 and αꢀ1,4 glycosidic linkages. Yet a simple first order
disconnection of globotriaose and isoglobotriaose at these linkages leads to 4’ꢀOH acceptor 12,
3’,4’ꢀdiꢀOH acceptor 13 or 3’ꢀOH acceptor 14 (Figure 4), which are all readily available from
the tandem ReSET products 2, 3 and 4.
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 39
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 4. Retrosynthetic analysis of globo series trisaccharides.
To reduce the glycosylation strategy to practice, compound 4 was treated with Dowex
acidic resin in MeOH to give the 4’ꢀOH acceptor 12 in 90% yield (Scheme 3). The long reaction
time (10 h) was consistent with earlier results in our lab indicating that the rate of
protodesilylation is attenuated with increasing numbers of acetate protecting groups.46 In an
attempt to shorten the reaction time, we initially subjected 4 to TBAF/AcOH, but the reaction led
to unresolved transꢀacetylation products. However, a report from Ikawa and coꢀworkers,53,54
encouraged us to explore the deprotection of silyl ethers using Pdꢀcatalyzed hydrogenolysis.
Much to our delight, after 30 min under 1 atm H2, 4 was transformed to 4’ꢀOH acceptor 12 in
91% yield. In a similar manner, either acidic methanolysis or Pdꢀcatalyzed hydrogenolysis was
applied to 3, leading to 3’,4’ꢀdiꢀOH compound 13 in 92% and 91% yields, respectively (Scheme
3B). To confirm the deprotection was the result of hydrogenolysis and not acidic catalysis, a
control experiment without the introduction of H2 gas was conducted. No reaction was observed
after 2 h indicating that H2 gas is required for the Pdꢀcatalyzed deprotection.
10
ACS Paragon Plus Environment

Page 11 of 39
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. Synthesis of acceptors 12 and 13.
In order to prepare 3’ꢀOH acceptor 14, we looked to the work of Lin and coꢀworkers who
published a oneꢀpot procedure to selectively acetylate at the 4ꢀposition of galactose via 3,4ꢀ
orthoester formation followed by selective acidic hydrolysis.55 The procedure showed excellent
results when applied to monosaccharides but not oligosaccharides, presumably due to solubility
issues.56 An adapted version of the methodology was applied to ReSET products 2 and 3
(Scheme 4). Compound 2 was first subjected to Pdꢀcatalyzed hydrogenolysis and then
concentrated to dryness. The resulting residue was reacted with trimethyl orthoacetate in the
presence of catalytic camphor sulfonic acid (CSA) to form the cyclic orthoester at the 3’ and 4’
positions. After peracetylation, the orthoester was selectively hydrolyzed affording the 3’ꢀOH
acceptor (14) in almost quantitative yield. The fourꢀstep procedure was carried out in less than
3h and only one flash column chromatography purification was needed to obtain the target
molecule. The consecutive hydrogenolysisꢀorthoester formationꢀacidic hydrolysis protocol was
also applied to diꢀOꢀTMS compound 3, leading to the same 3’ꢀOH acceptor 14 in almost
quantitative yield (Scheme 4).
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 39
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 4. Synthesis of 3’ꢀOH acceptor 14.
With all the acceptors (12ꢀ14) in hand, we next examined the glycosylation reactions to
form the globo series trisaccharides (Table 1). Wishing to achieve efficient αꢀgalactosidation of
the 4’ꢀOH acceptor (12), we first tried using perꢀOꢀTMS galactosyl iodide promoted by TBAI in
situ anomerization;35,36 but the major product obtained was silylated acceptor. We previously
observed similar transꢀsilylation complications, especially in cases where the acceptor is
hindered or unreactive.57,58 However, the perꢀOꢀbenzyl galactosyl iodide derived from anomeric
acetate 1559 was reactive when activated with AgOTf, and the presence of benzyl protecting
groups conveniently allowed UV monitoring of the reaction progress. Importantly, only the
desired αꢀ1,4 linkage was obtained, which simplified the purification (Table 1, entry 1). The
same reaction conditions with 1.5 and 2.5 equiv. of iodide were applied to the glycosylation
reaction with acceptor 13 to examine the relative reactivity of the two hydroxyl groups. When
1.5 equiv. of iodide was used, the desired protected isoglobotraose (17) was obtained, albeit in a
lower yield of 44% (Table 1, entry 2). When 2.5 equiv. of iodide was added, not only was 17
obtained in 26% yield but also the diꢀgalactosylated product (18) in 31% yield (Table 1, entry 3).
To avoid over galactosylation, the 3’ꢀOH acceptor 14 was introduced. In this case, the
desired product (19) was obtained in 58% yield (Table 1, entry 4). Side reactions, which we
attributed to acyl migration were also observed on the TLC plate. To minimize this possibility,
12
ACS Paragon Plus Environment

Page 13 of 39
The Journal of Organic Chemistry
1
2
an acid scavenger (tetramethylurea, TMU)60 was added to the AgOTfꢀpromoted glycosylation
(Table 1, entry 5). The neutralized glycosylation procedure required higher temperature and
longer time to complete, however the target molecule (19) was obtained in 75% yield, indicating
that the side reactions could be suppressed by the addition of weak base.
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
Table 1. Glycosyl iodide glycosylation gave Gb3 and iGb3 trisaccharide scaffolds.
Debenzylation of compound 16 followed by reacetylation led to perꢀOꢀAc globotriaose
20 in 95% yield (Scheme 5). Storing the trisaccharide in the peracetylated form was desired to
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 39
1
2
increase stability. Moreover, the 1H NMR signals of compound 20 were better resolved than the
benzylated counterpart (16), making compound characterization and quality control more
reliable. Acetate protecting groups are also preferred when generating the glycosyl iodide of the
trisaccharides. When compound 16 was treated with TMSI, the reaction became messy and
glycosidic bond cleavage products were observed in crude MS analyses; another example of
electron donating protecting groups rendering the glycosidic linkage susceptible to cleavage. In
contrast, peracetylated 20 readily transformed to the corresponding iodide (22) in situ and
subsequent reaction leads to βꢀlinked glycoconjugates.
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
Global deprotection of 16 continued with Pdꢀcatalyzed hydrogenolysis followed by
deacetylation yielding globotriaose (21) in nearly quantitative yield (Scheme 5). The total
synthesis of globotriaose required only four steps from free lactose to form the globotriaose
scaffold, and two more consecutive protecting group manipulations (total six steps) to the natural
product globotriaose (21).
Scheme 5. Sixꢀstep synthesis of globotriaose from free lactose.
14
ACS Paragon Plus Environment

Page 15 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
In 1988, Nicolaou and coꢀworkers published the first synthesis of Gb3, which is arguably
the most efficient synthesis until now.17 The approach required seven steps to prepare a
selectively protected 4’ꢀOH acceptor, which was glycosylated with perꢀOꢀBn galactosyl fluoride
under the activation of AgCl4 and SnCl2. During the process, seven protecting groups (OAc,
OBn, OBz, OPiv, SPh, F, and benzylidene) were utilized. In contrast, the simplified
RESET/glycosyl iodide synthetic platform involves only three steps to prepare the 4’ꢀOH
glycosyl acceptor, and arrives at the globotriaose scaffold in four steps. Moreover, the glycosyl
iodides react under neutral conditions avoiding toxic tin reagents as the promoter for αꢀ1,4
glycosylation.
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
Compounds 17 and 19 could be transformed to perꢀOꢀAc isoglobotriaose (23) in 85%
yield by hydrogenolysis and perꢀacetylation (Scheme 6). In 2009, Castillon and coꢀworkers
published an efficient TBAI/AW300ꢀpromoted synthesis of iGb3 using perꢀOꢀAc isoglobotriaose
iodide and stannylceramide.61 The group purchased isoglobotriaose and perꢀacetylated it prior to
generating the iodide. In the total synthesis approach to iGb3 reported here, both 23 and the in
situ generated iodide intercept the Castillon synthesis, providing a formal synthesis of iGb3.
Scheme 6. Formal synthesis of iGb3 from compound 17.
15
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 39
1
2
3
4
5
6
7
8
9
Among the published syntheses of isoglobotriaose scaffold using 3’, 4’ꢀdiꢀOH acceptors,
galactosylation typically suffers from low yields and stereoselectivity. Moreover, the 3’, 4’ꢀdiꢀ
OH acceptors require at least six or seven steps to prepare from free lactose. When using
acetylated galactosyl bromide24 and trichloroacetimidate26 donors the glycosylation yields ranged
from 28ꢀ31% with inseparable αꢀ1,3 and βꢀ1,3 isomers. In order to eliminate side reactions and
increase donor reactivity, both Ogawa25 and Schmidt27 have introduced Oꢀbenzylꢀprotected
galactosyl thiomethyl and trichloroacetimidate donors. These reactive donors provided increased
yields, but did not prevent the formation of βꢀlinked isomers. In contrast, Oꢀbenzyl protected
galactosyl iodide (Table 1, entries 2 & 3) readily reacts with acceptors 12, 13 and 14 to afford the
αꢀlinked products exclusively and in yields ranging from 44ꢀ75%.
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
Selectively protected 3’ꢀOH acceptors generally afford better yields than 3’,4’ꢀdiꢀOH
acceptors, since there is no competition between the two hydroxyl groups. However, six to eight
steps were required in order to prepare the 3’ꢀOH acceptors. The protecting group manipulations
involved OAc, OBn, OPiv, OPMB and SPh groups, and the preparation took days to
complete.28,29 In comparison, the ReSET approach (Table 1, entry 4 & 5), required only three
protecting groups (OBn, OAc and OTMS) and 21 h to synthesize acceptor 14. The optimized
glycosylation of compound 14 reached 75% yield when coupled with perꢀOꢀBn galactosyl iodide.
ꢀ
Conclusion
The combined ReSET/glycosyl iodide glycosylation strategy provides a stepꢀeconomy
platform for glycoconjugate synthesis that centers on the conversion of perꢀOꢀTMSꢀlactose into
selectively protected modules with differential reactivites.62 The orthogonally protected
intermediates can be transformed into reactive glycosyl iodides in situ and coupled with
16
ACS Paragon Plus Environment

Page 17 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
unprotected or partially protected ceramides with high stereoselectivity in good yields. The
marriage of these powerful platforms results in the exclusive formation of alphaꢀlinked
glycosides, which leads to increased efficiencies during the purification process. The
bifunctional nature of these modules can be further exploited upon removal of the TMS groups
yielding glycosyl acceptors ready for sugar chain elongation. When introduced into
glycosylation reactions, the corresponding globo series tumorꢀassociated carbohydrate antigens
can be obtained in respectable yields. The TACA syntheses require only three protecting groups
(OAc, OBn and OTMS), which significantly reduces the number of protection/deprotection steps;
not only in preparing the glycosyl acceptors, but also by direct activation of the anomeric acetate
or silyl ether to the glycosyl iodide. These combined features characterize a versatile synthetic
platform for the rapid assembly of biologically relevant glycolipids.
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
Beyond the time and step efficiencies of these methodologies, a unique interplay between
acetate and silyl ether protecting groups is revealed. The acetate protecting groups help suppress
side reactions such as silyl migration and interꢀresidue glycosidic bond cleavage during TMSIꢀ
promoted iodide formation. Similarly, we find that proximal acetates significantly alter the
reactivity of silyl ethers toward protodesilylation. This phenomenon is clearly evidenced in the
acidic methanolysis of 4’ꢀOTMS acetylated lactose (4), which took twice as long as analog 3
having one less acetate i.e. 3’,4’ꢀdiꢀOTMS acetylated lactose. We attribute the reactivity
attenuation to acetate electron withdrawing effects, which inductively reduce the basicity of the
ether and acetal oxygen atoms. Exquisite control is afforded by acetate incorporation, as shown
in the contrasting behavior of perꢀOꢀsilylated lactose and 6,6’ꢀdiꢀOꢀAcꢀperꢀOꢀsilyllactose (1)
under the action of TMSI. Diꢀ and triꢀsaccharide substrates having no acetates undergo interꢀ
residue glycosidic bond cleavage, whereas incorporation of only two acetate groups at the Cꢀ6
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 39
1
2
3
4
5
6
7
8
9
positions directs reactivity toward the reducing end acetal resulting in the quantitative generation
of the corresponding glycosyl iodides. These findings offer opportunities in systems removed
from carbohydrate substrates where one could capitalize on the concept of attenuating ether
reactivity by the inductive effects of proximal protecting groups.
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
ꢀ
Experimental Section
Procedure for Tandem Regioselective Silyl Exchange (ReSET) of PerꢀOꢀTMS Lactose
(Scheme 1): In a 10 mL microwave reactor vessel with a dry stir bar, perꢀOꢀTMS lactose (500
mg, 0.54 mmol) was dissolved in anhydrous benzene (3 mL). The solvent was removed under
rotary evaporation with argon backfilling. The azeotropic distillation was repeated two
additional times to dryness affording viscous syrup. To the reaction flask was added anhydrous
pyridine (2.0 mL/per mmol TMS sugar: 1.1 mL), Ac2O (1.5 mL/per mmol TMS sugar: 8.1 mL),
and AcOH (93 ꢀL, 1.62 mmol, 3.0 equiv.). The reaction vessel was subjected to microwave
irradiation (standard mode, 100 W, 125 °C, ~40 psi) for 1.25 h. Once TLC showed the reaction
was complete, the reaction mixture was transferred into a 50 mL round bottom flask where it was
azeotroped with copious amounts of anhydrous benzene to remove excess reagents and solvent.
The crude mixture was immediately purified by gradient flash column chromatography
(EA/Hex/NEt3 = 5:85:10 to 14:85:1 to 29:70:1) to afford compound 1 (93 mg, 20%) and 2 (239
mg, 53%). Later on, compound 1 and 2 were combined. A mixture of compound 1 and 2 (194
mg, ~ 0.23 mmol: calculated based on compound 2’s molecular weight; same as the following)
was then treated with dry pyridine (0.35 mL), Ac2O (0.46 mL) and AcOH (40 ꢀL, 0.69 mmol,
~3.0 equiv.), followed by microwave irradiation (standard mode, 100 W, 125 ˚C, ~40 psi) for 1.5
h. After the starting material was completely consumed, the reaction was azeotroped with dry
18
ACS Paragon Plus Environment

Page 19 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
benzene to remove excess reagents and solvent. The resulting residue was immediately purified
by gradient flash column chromatography (EA/Hex = 40:60 to 60:40) to afford compound 3 (54
mg, ~31%), compound 4 (41 mg, ~25%) and compound 5 (42 mg, ~26%) as white foams.
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
(2,6ꢀDiꢀOꢀacetylꢀ3,4ꢀOꢀdiꢀtrimethylsilylꢀβꢀ
Dꢀgalactopyranosyl)ꢀ(1ꢀ4)ꢀOꢀ(1,2,3,6ꢀtetraꢀOꢀ
acetyl)ꢀ ꢀglucopyranoside (3). Compound 3 (54 mg, ~31%) was obtained from consecutive
D
ReSET of perꢀOꢀTMS lactose (see ReSET procedure for reaction and purification conditions).
The product consisted of inseparable α/β anomeric acetates and the major βꢀanomer is reported:
1
TLC (EA/Hex = 40:60) Rf 0.38. H NMR (800 MHz, CDCl3) δ 5.66 (d, J = 8.2 Hz, 1H, Hꢀ1),
5.21 (appt. t, J = 9.3 Hz, 1H, Hꢀ3), 5.06 – 5.00 (m, 2H, Hꢀ2, Hꢀ2’), 4.40 (dd, J = 12.0, 1.7 Hz, 1H,
Hꢀ6a), 4.28 (d, J = 7.9 Hz, 1H, Hꢀ1’), 4.19ꢀ4.14 (m, 2H, Hꢀ6b, Hꢀ6’a), 4.10 (dd, J = 11.0, 6.6 Hz,
1H, Hꢀ6’b), 3.79 (appt. t, J = 9.3 Hz, 1H, Hꢀ4), 3.77 – 3.72 (m, 2H, Hꢀ4’, Hꢀ5), 3.57 (appt. t, J =
6.6 Hz, 1H, Hꢀ5’), 3.53 (dd, J = 9.8, 2.7 Hz, 1H, H3’), 2.11 (s, 3H), 2.10 (s, 3H), 2.09 (s, 3H),
2.05 (s, 3H), 2.04 (s, 3H), 2.02 (s, 3H), 0.12 (s, 9H), 0.11 (s, 9H).. 13C NMR (200 MHz, CDCl3)
δ 170.6, 170.5, 169.8, 169.7, 168.9, 168.89, 101.1, 91.6, 74.9, 73.6, 72.9, 72.6, 72.4, 71.6, 71.1,
70.5, 62.5, 62.1, 21.1, 20.84, 20.82, 20.7, 20.6, 0.4, 0.2. HRMS (ESIꢀion trap): m/z calcd for
+
[C30H50O17Si2 +NH4 ] 756.2925; found 756.2947.
General Procedure for αꢀLactosylceramide Synthesis (Scheme 2). In an ovenꢀdried NMR
tube, compound 2 (59 mg, 0.07 mmol) was added and dissolved in dry benzeneꢀd6 (0.7 mL).
After TMSI (12 ꢀL, 0.08 mmol) was introduced to the reaction vessel, the reaction was kept and
0 ˚C and gradually warmed to rt over 4ꢀ5 h. The reaction was monitored by proton NMR until
the corresponding iodide (6) formed in situ. Next, compound 6 was transferred to a 25 mL pearꢀ
shape bottle, azeotroped with dry benzene (3 mL × 3) and dried under highꢀvacuum for 1 h to
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 39
1
2
3
4
5
6
7
8
9
afford the iodide as a light yellow foam. Note: The iodide is highly reactive and moisture
sensitive. Column chromatography or aqueous workup should be avoided. The compound 6
was next dissolved in dry benzene (2 mL) and kept under Ar. In a separate round bottom flask,
ceramide 7 (20 mg, 0.035 mmol), TBAI (52 mg, 0.14 mmol), DIPEA (25 mL, 0.14 mmol) and
activated 4 Å molecular sieves (80 mg) were dissolved in dry bezene (2 mL) and stirred at rt
under Ar. The iodide solution was transferred to the glycosyl acceptor solution via cannula
dropwisely over 3 min at rt. Once transferred, the reaction mixture was gently heated to 65 ˚C
overnight. Next, the reaction mixture was cooled to rt, filtered through a short celite pad and
concentrated under reduced pressure. The resulting residue was dissolved in MeOH (5 mL) and
stirred with the Dowex acidic resin (pH =2ꢀ3) at rt. After 2 h, the resin was filtered, and the
solvent was evaporated under reduced pressure to afford a viscous orange oil. The resulting oil
was then dissolved in CH2Cl2 (1 mL) under Ar. To the solution was then added Ac2O (0.8 mL),
NEt3 (1.2 mL) and catalytic DMAP (~2 mg), and stirred at rt overnight. Next, the reaction
mixture was concentrated under reduced pressure and was purified using flash column
chromatography (EA/Hex = 50:50) to afford a mixture of isomers (compounds 8 and 9) (39 mg,
89%) as a white foam. The mixture could be further purified by gradient flash column
chromatography (EA/Hex = 40:60 to 60:40) to separate both isomers for characterization
purpose. MonoꢀOꢀTMS protected ceramide 10 (19 mg, 0.03 mmol) could also be applied to the
above reaction conditions but using 50 ˚C in the glycosylation step. Followed by desilylation
and reꢀacetylation, compound 8 (22 mg, 61%) was obtained as the only isomer.
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
(2S,3R,4E)ꢀ3ꢀOꢀAcetylꢀ1ꢀOꢀ(αꢀperꢀOꢀacetylꢀDꢀlactopyranosyl)ꢀ2ꢀ(Nꢀ
octadecanosylamino)octadecꢀ4ꢀeneꢀ1,3ꢀdiol (8). Compound 8 was obtained either from
20
ACS Paragon Plus Environment

Page 21 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
glycosyl iodide glycosylation using iodide 6, and ceramide 7 or ceramide 10 (see general
procedure for αꢀlactosylceramide synthesis for reaction and purification conditions). TLC
7
1
(EA/Hex = 1/1) Rf 0.43. [α]24D +16.5 (c 0.26, CHCl3). H NMR (800 MHz, C6D6) δ 5.88 (appt.
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
t, J = 9.5 Hz, 1H, Hꢀ3), 5.86 – 5.82 (m, 1H, alkene Hꢀe), 5.55 (appt. t, J = 7.9 Hz, 1H, Hꢀc), 5.52
– 5.47 (m, 3H, alkene Hꢀd, Hꢀ2’, Hꢀ4’), 5.44 (d, J = 9.2 Hz, 1H, NH), 5.14 (dd, J = 10.4, 3.4 Hz,
1H, Hꢀ3’), 5.10 (dd, J = 9.5, 3.7 Hz, 1H, Hꢀ2), 5.05 (d, J = 3.7 Hz, 1H, Hꢀ1), 4.73 (d, J = 10.4 Hz,
1H, Hꢀ6a), 4.63 – 4.58 (m, 1H, Hꢀb), 4.39 (d, J = 7.8 Hz, 1H, Hꢀ1’), 4.22 (dd, J = 11.7, 6.4 Hz,
1H, Hꢀ6b), 4.14 (dd, J = 11.1, 6.6 Hz, 1H, Hꢀ6’a), 4.11 – 4.05 (m, 2H, Hꢀ5, Hꢀ6’b), 3.77 (dd, J =
10.7, 3.2 Hz, 1H, Hꢀa), 3.67 (appt. t, J = 9.5 Hz, 1H, Hꢀ4), 3.50 (appt. t, J = 6.6 Hz, 1H, Hꢀ5’),
3.44 (dd, J = 10.7, 4.0 Hz, 1H, Hꢀa’), 2.02 (s, 3H), 2.01 (s, 3H), 2.00 (s, 3H), 1.95 (s, 3H), 1.87 (s,
3H), 1.74 (s, 3H), 1.69 (s, 3H), 1.66 (s, 3H), 1.55 (s, 3H), 1.35ꢀ1.26 (m, 52H, alkyl chain), 0.93ꢀ
13
0.91 (m, 6H, terminal Me ×2). C NMR (200 MHz, C6D6) δ 172.3, 170.3, 170.2, 170.1, 169.9,
169.8, 169.4, 169.2, 169.1, 137.3, 128.4, 128.3, 128.2, 128.1, 128.0, 127.9, 126.1, 101.6, 96.8,
77.2, 73.1, 71.6, 71.5, 70.9, 70.7, 69.7, 69.2, 67.4, 67.0, 62.6, 61.1, 51.3, 36.5, 32.8, 32.38, 32.37,
30.3, 30.28, 30.26, 30.25, 30.24, 30.19, 30.18, 30.0, 29.88, 29.87, 29.77, 29.72, 29.4, 26.1, 23.2,
20.9, 20.7, 20.6, 20.5, 20.49, 20.2, 20.16, 19.8, 14.4. HRMS (ESIꢀion trap): m/z calcd for
[C64H107NO21+Na+] 1248.7228; found 1248.7227.
(2S,3R,4E)ꢀ1ꢀOꢀAcetylꢀ3ꢀOꢀ(αꢀperꢀOꢀacetylꢀDꢀlactopyranosyl)ꢀ2ꢀ(Nꢀ
octadecanosylamino)octadecꢀ4ꢀeneꢀ1,3ꢀdiol (9). Compound 9 was obtained from the glycosyl
iodide glycosylation using iodide 6 and ceramide 7 (see general procedure for αꢀ
lactosylceramide synthesis for reaction and purification conditions). TLC (EA/Hex = 1/1) Rf
1
0.40. [α]24D +21.6 (c 0.47, CHCl3). H NMR (800 MHz, C6D6) δ 5.88 (appt. t , J = 9.5 Hz, 1H,
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 39
1
2
3
4
5
6
7
8
9
Hꢀ3), 5.69 (d, J = 8.6 Hz, 1H, NH), 5.60 (ddd, J = 18.6, 9.1, 3.4 Hz, 1H, alkene Hꢀe), 5.55 – 5.52
(m, 1H, Hꢀ2’), 5.49 (d, J = 3.5 Hz, 1H, Hꢀ4’), 5.38 (dd, J = 15.5, 8.0 Hz, 1H, alkene Hꢀd), 5.21
(d, J = 3.7 Hz, 1H, Hꢀ1), 5.15 (dd, J = 10.5, 3.5 Hz, 1H, Hꢀ3’), 5.06 (dd, J = 10.2, 3.7 Hz, 1H, Hꢀ
2), 4.67 (dd, J = 11.8, 2.0 Hz, 1H, Hꢀ6a), 4.51 (d, J = 7.9 Hz, 1H, Hꢀ1’), 4.42 (ddd, J = 8.9, 7.4,
4.6 Hz, 1H, Hꢀb), 4.36 (dd, J = 11.6, 6.0 Hz, 1H, Hꢀa), 4.31 (dd, J = 11.9, 4.6 Hz, 1H, Hꢀ6b),
4.22 – 4.06 (m, 5H, Hꢀ5, Hꢀ6’ab, Hꢀc, Hꢀa’), 3.87 – 3.83 (m, 1H, Hꢀ4), 3.48 (appt. t, J = 7.0 Hz,
1H, Hꢀ5’), 2.01 (s, 3H), 1.98 (s, 3H), 1.91 (s, 3H), 1.83 (s, 3H), 1.78 (s, 1H), 1.74 (s, 3H), 1.64 (s,
3H), 1.53 (s, 3H), 1.43 – 1.22 (m, 52H, alkyl chain), 0.95 – 0.91 (m, 6H, terminal Me×2). 13C
NMR (200 MHz, C6D6) δ 171.8, 171.1, 170.4, 170.1, 169.9, 169.8, 169.7, 169.2, 168.9, 136.1,
127.3, 101.6, 97.2, 81.8, 76.9, 72.2, 71.6, 70.9, 70.8, 69.9, 69.3, 66.9, 63.0, 62.2, 60.9, 52.6, 36.8,
32.6, 32.38, 32.37, 30.3, 30.27, 30.25, 30.23, 30.20, 30.18, 30.1, 30.0, 29.97, 29.9, 29.87, 29.80,
29.5, 26.0, 23.1, 20.9, 20.7, 20.6, 20.5, 20.4, 20.2, 20.1, 19.8, 14.4. HRMS (ESIꢀion trap): m/z
calcd for [C64H107NO21+Na+] 1248.7228; found 1248.7272.
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
(2S,3R,4E)ꢀ3ꢀOꢀTrimethylsilylꢀ2ꢀ(Nꢀoctadecanosylamino)octadecꢀ4ꢀeneꢀ1,3ꢀdiol (10). To a 0
˚C CH2Cl2 (3 mL) solution of compound 7 (150 mg, 0.27 mmol) was added 1,1,1,3,3,3ꢀ
hexamethyldisilazane (HMDS, 121 ꢀL, 0.58 mmol) and TMSOTf (~5 ꢀL, 0.03 mmol), and
gradually warmed to rt in 2 h (based on Wang’s method).63 The reaction mixture was then
diluted with EA/Hex (v/v =1:1, 20 mL) and washed with water and brine. The organic layer was
dried over Na2SO4 and concentrated under reduce pressure to afford perꢀOꢀTMS ceramide (179
mg, 95%) as a white powder. PerꢀOꢀTMS ceramide (179 mg, 0.25 mmol) was dissolved in a
wellꢀstirred CH2Cl2/MeOH (v/v = 1:1, 3 mL) solution. To the solution was added NH4OAc (38
mg, 0.5 mmol) and the reaction was stirred at rt for 10 h. The reaction mixture was concentrated
22
ACS Paragon Plus Environment

Page 23 of 39
The Journal of Organic Chemistry
1
2
3
4
and purified using flash column chromatography (EA/Hex = 1/4 to 1/2 ) to afford monoꢀOꢀTMSꢀ
5
6
protected ceramide 10 (91 mg, 58%) as a white amorphous solid. TLC (EA/Hex = 1/5) Rf 0.20.
7
8
9
1
[α]24 ꢀ15.0 (c 0.02, MeOH). H NMR (600 MHz, C6D6) δ 5.94 (d, J = 7.4 Hz, 1H, NH), 5.63
D
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
(ddd, J = 7.9, 6.8, 3.4 Hz, 1H, Hꢀ5), 5.42 (dd, J = 15.4, 6.1 Hz, 1H, Hꢀ4), 4.56 – 4.49 (m, 1H, Hꢀ
3), 4.06 (d, J = 11.5 Hz, 1H, Hꢀ1a), 3.97 (dt, J = 7.2, 3.5 Hz, 1H, Hꢀ2), 3.65 – 3.55 (m, 1H, Hꢀ1b),
3.06 (d, J = 7.0 Hz, 1H, OH), 1.96ꢀ1.93 (m, 4H, Hꢀ6ab, Hꢀ2’ab), 1.70 (dd, J = 14.1, 7.0 Hz, 2H,
Hꢀ7ab), 1.46 – 1.17 (m, 52H, alkyl chain), 0.97 – 0.85 (m, 6H, terminal Me×2), 0.14 – 0.03 (s,
9H, TMS). 13C NMR (150 MHz, C6D6) δ 172.3, 133.1, 130.2, 75.7, 62.5, 55.2, 36.7, 32.6, 32.4,
30.24, 30.20, 30.16, 30.1, 29.98, 29.92, 29.9, 29.8, 29.7, 29.6, 26.0, 23.1, 14.4, 0.2. HRMS
(ESIꢀion trap): m/z calcd for [C39H79NO3Si+H+] 638.5902; found 638.5930.
(2S,3R,4E)ꢀ3ꢀOꢀAcetylꢀ1ꢀOꢀ(αꢀDꢀlactopyranosyl)ꢀ2ꢀ(Nꢀoctadecanosylamino)octadecꢀ4ꢀeneꢀ
1,3ꢀdiol (11). Compound 8 (17 mg, 0.012 mmol) was dissolved in dry MeOH (2 mL) and treated
with 25% NaOMe in MeOH solution (~5 ꢀL) at rt. After 1 h, Dowex acidic resin was added to
the reaction mixture until pH =7. The mixture was then filtered through a short celite plug and
concentrated under reduced pressure to afford compound 11 (15 mg, >98%) as a white powder.
TLC (EA/MeOH/H2O = 7:2:1) Rf 0.77. [α]24 +2.7 (c 0.40, MeOH). 1H NMR (800 MHz,
D
pyridineꢀd5) δ 8.48 (d, J = 8.5 Hz, 1H, NH), 6.01ꢀ5.99 (m, 2H, alkene H ×2), 5.37 (d, J = 3.7 Hz,
1H, Hꢀ1), 5.09 (d, J = 7.8 Hz, 1H, Hꢀ1’), 4.88ꢀ4.85 (m, 1H, Hꢀc), 4.82ꢀ4.80 (m, 1H, Hꢀb), 4.57ꢀ
4.39 (m, 8H, Hꢀ3, Hꢀ2’, Hꢀ4’, Hꢀ6ab, Hꢀ6’ab, Hꢀa), 4.37 – 4.34 (m, 1H, Hꢀ5), 4.32ꢀ4.30 (m, 1H,
Hꢀa’), 4.25 (appt. t, J = 9.3 Hz, Hꢀ4), 4.16ꢀ4.14 (m, 2H, Hꢀ3’, Hꢀ5’), 4.13 (dd, J = 3.7, 9.3 Hz,
1H, Hꢀ2), 2.44 (appt. t, J = 7.5, 2H, Hꢀg, Hꢀg’), 2.11ꢀ2.08 (m, 2H, Hꢀf, Hꢀf’), 1.88ꢀ1.81 (m, 2H,
Hꢀh, Hꢀh’), 1.39 – 1.22 (m, 52H, alkyl chain), 0.90ꢀ0.87 (m, 6H, terminal Me×2). 13C NMR
23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 39
1
2
3
4
5
6
(200 MHz, pyridineꢀd5) δ 174.0, 133.0, 132.5, 106.3, 101.7, 83.1, 77.7, 75.7, 74.0, 73.9, 73.2,
73.0, 72.9, 70.6, 69.6, 62.8, 62.6, 55.6, 37.3, 33.3, 32.6, 30.54, 30.51, 30.50, 30.45, 30.43, 30.41,
7
8
9
30.37, 30.29, 30.25, 30.1, 26.9, 23.4, 14.8. HRMS (ESIꢀion trap): m/z calcd for [C48H91NO13+H+]
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
890.6563; found 890.6569.
(2,6ꢀdiꢀOꢀacetylꢀ3,4ꢀdiolꢀβꢀDꢀgalactopyranosyl)ꢀ(1ꢀ4)ꢀOꢀ(1,2,3,6ꢀtetraꢀOꢀacetyl)ꢀDꢀ
glucopyranoside (13). Method A: To a MeOH (3 mL) solution of compound 3 (41 mg, 0.06
mmol) was added Dowex H+ resin (~80 mg) until the pH = 2ꢀ3. The suspension was allowed to
stir at rt for 4.5 h until the starting material was completely consumed. Next, the Dowex acidic
resin was removed via filtration and the filtrate was concentrated to afford a viscous oil. The
resulting residue was immediately purified by flash column chromatography (100% EA) to
obtain compound 13 (30 mg, 92%) as a white foam. Method B: To a MeOH solution of
compound 3 (115 mg, 0.16 mmol) was added Pd(OH)2/C (20% Pd, 100 mg). The reaction
mixture was allowed to stir under H2 gas (1 atm) at rt for 0.5 h until the starting material was
completely consumed. Next, the Pdꢀcatalyst was removed by a short plug of MeOHꢀpacked
celite. The filtrate was concentrated to afford a viscous oil. The resulting residue was purified
by flash column chromatography (100% EA) to obtain compound 13 (85 mg, 91%) as a white
foam. The product consisted of inseparable α/β anomeric acetates and the major βꢀanomer is
1
reported: TLC (100% EA) Rf 0.36. H NMR (600 MHz, CDCl3) δ 5.67 (d, J = 8.8 Hz, 1H, Hꢀ1),
5.23 (appt. t, J = 8.8 Hz, 1H, Hꢀ3), 5.07 (appt. t, J = 8.8 Hz, 1H, Hꢀ2), 4.81 (appt. t, J = 8.4 Hz,
1H, Hꢀ2’), 4.46 (d, J = 12.4 Hz, 1H, Hꢀ6a), 4.38ꢀ.34 (m, 2H, Hꢀ1’, Hꢀ6b), 4.28 – 4.22 (m, 1H, Hꢀ
6’a), 4.22 (m, 1H, Hꢀ6’b), 3.84ꢀ3.77 (m, 3H, Hꢀ4, Hꢀ5, Hꢀ4’), 3.69 – 3.54 (m, 2H, Hꢀ3’, Hꢀ5’),
2.13 (s, 3H), 2.12 (s, 3H), 2.11 (s, 3H) 2.10 (s, 3H), 2.06 (s, 2H), 2.03 (s, 2H). 13C NMR (150
24
ACS Paragon Plus Environment

Page 25 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
MHz, CDCl3) δ 171.9, 171.0, 170.4, 169.9, 169.5, 168.9, 100.5, 91.7, 75.6, 74.1, 73.7, 72.9, 72.4,
72.0, 70.4, 68.1, 62.1, 61.9, 20.9, 20.84, 20.82, 20.8, 20.76, 20.6. HRMS (ESIꢀion trap): m/z
calcd for [C24H34O17+NH+] 612.2134; found 612.2134.
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
(2,6ꢀDiꢀOꢀacetylꢀβꢀDꢀgalactopyranosyl)ꢀ(1ꢀ4)ꢀOꢀ(1,2,3,6ꢀtetraꢀOꢀacetyl)ꢀDꢀglucopyranoside
(14). Method A: To a MeOH (3 mL) solution of compound 2 (70 mg, 0.08 mmol) was added
Pd(OH)2/C (20% Pd, 50 mg). The reaction mixture was allowed to stir under H2 (1 atm) at rt for
0.5 h. Next, the reaction mixture was filtered through a MeOHꢀpacked celite pad to remove the
Pd catalyst, and the filtrate was concentrated under reduced pressure. The resulting residue was
then dissolved in dry acetonitrile (2 mL) and treated with trimethyl orthoacetate (32 ꢀL, 0.25
mmol) and catalytic camphorsulfonic acid (~4 mg, 0.02 mmol). After 1 h, NEt3 (~0.1 mL) was
added to quench the reaction, and the reaction mixture was concentrated under reduced pressure.
The resulting residue was then dissolved in CH2Cl2 (2 mL) and treated with Ac2O (50 ꢀL, 0.50
mmol), NEt3 (105 ꢀL, 0.75 mmol) and catalytic DMAP (~2 mg). After 0.5 h, the solvent was
evaporated and the dry residue was treated with 80% AcOH(aq) at rt with vigorous stirring for
another 0.5 h. The reaction mixture was then azeotroped with benzene to remove excess
reagents, and the resulting residue was purified using flash column chromatography (EA/Hex =
3/1 to 4/1) to afford compound 14 (52 mg, 95%) as a white foam. Method B: To a MeOH (3 mL)
solution of compound 3 (65 mg, 0.088 mmol) was added Pd(OH)2/C (20% Pd, 60 mg). The
reaction mixture was allowed to stir under H2 (1 atm) at rt for 0.5 h. Next, the reaction mixture
was filtered through a MeOHꢀpacked celite pad to remove the Pd catalyst, and the filtrate was
concentrated under reduced pressure. The resulting residue was then dissolved in dry ACN (2
mL) and treated with trimethyl orthoacetate (35 ꢀL, 0.26 mmol) and catalytic CSA (~4 mg, 0.02
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 39
1
2
3
4
5
6
7
8
9
mmol). After 1 h, NEt3 (~0.1 mL) was added to quench the reaction, and the reaction mixture
was concentrated under reduced pressure. The resulting residue was treated with 80% AcOH(aq)
at rt with vigorous stirring for another 0.5 h. The reaction mixture was then azeotroped with
benzene to remove the excess reagents, and the resulting residue was purified using flash column
chromatography (EA/Hex = 3/1 to 4/1) to afford compound 14 (57 mg, 98%) as a white foam.
The product consisted of inseparable α/β anomeric acetates and the major βꢀanomer is reported:
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
TLC (EA/Hex = 2/1) Rf 0.28. H NMR (600 MHz, CDCl3) δ 5.68 (d, J = 8.4 Hz, 1H, Hꢀ1), 5.30
(d, J = 2.9 Hz, 1H, Hꢀ4’), 5.24 (appt. t, J = 9.3 Hz, 1H, Hꢀ3), 5.06 (dd, J = 9.3, 8.4 Hz, 1H, Hꢀ2),
4.85 (dd, J = 10.0, 7.9 Hz, 1H, Hꢀ2’), 4.47 (dd, J = 12.1, 1.8 Hz, 1H, Hꢀ6a), 4.42 (d, J = 7.9 Hz,
1H, Hꢀ1’), 4.20 (dd, J = 12.1, 4.9 Hz, 1H, Hꢀ6b), 4.15 – 4.04 (m, 2H, Hꢀ6’ab), 3.86 – 3.73 (m,
4H, Hꢀ4, Hꢀ5, Hꢀ3’, Hꢀ5’), 2.46 (d, J = 6.3 Hz, 1H, OH), 2.18 (s, 3H), 2.13 (s, 3H), 2.12 (s, 3H),
2.10 (s, 3H), 2.09 (s, 3H), 2.05 (s, 3H), 2.03 (s, 3H). 13C NMR (150 MHz, CDCl3) δ 171.2,
170.7, 170.5, 170.4, 169.7, 169.5, 100.6, 91.6, 75.7, 73.6, 73.1, 72.5, 71.8, 71.0, 70.4, 69.1, 61.9,
61.4, 20.84, 20.81, 20.75, 20.73, 20.67, 20.5. HRMS (ESIꢀion trap): m/z calcd for
+
[C26H36O18+NH4 ] 654.2240; found 654.2289.
General Procedure for Constructing Globotriaose and Isoglobotriaose (Table 1). To a 0 ˚C
CH2Cl2 solution of compound 15 (conc. 0.1ꢀ0.2 M, 2.5 equiv.) was added TMSI (2.8 equiv.)
under argon atmosphere. After stirring for 30 min, the reaction mixture was azeotroped with
anhydrous benzene (5 mL × 3) and dried under highꢀvacuum system for 1 h to afford the
corresponding glycosyl iodide as a light yellow foam. The iodide was then dissolved in CH2Cl2
o
(1 mL) under argon and cooled to ꢀ78 C. In a separate flask, AgOTf (3.2 equiv.), acceptor
(compound 12, 13 or 14; ~0.10 mmol scale; 1.0 equiv.) and activated 4 Å molecular sieves (~200
26
ACS Paragon Plus Environment

Page 27 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
mg) were allowed to stir in a CH2Cl2 (2 mL) solution under argon atmosphere at rt for 30 min.
The acceptor solution was then cooled to ꢀ78 ˚C and the cooled glycosyl iodide solution was
transferred dropwisely to the acceptor flask via cannula. After 4 h, the reaction mixture was
gradually warmed from ꢀ78 to ꢀ30 ˚C, diluted with EA (10 mL), and quenched by NEt3 (~1 mL).
The suspension was filtered through a wellꢀpacked celite pad and washed with EA (~30 mL).
The filtrate was washed with saturated NaHCO3(aq) (20 mL × 1) and brine (20 mL × 2), dried
over anhydrous Na2SO4, and concentrated under reduced pressure. The resulting residue was
purified by gradient flash column chromatography (EA/Hex) to obtain corresponding
oligosaccharides (compound 16, 17, 18 or 19) as white foams.
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
(2,3,4,6ꢀTetraꢀOꢀbenzylꢀαꢀ
D
ꢀgalactopyranosyl)ꢀ(1ꢀ4)ꢀOꢀ(2,3,6ꢀtriꢀOꢀacetylꢀβꢀOꢀ
Dꢀ
galactopyranosyl)ꢀ(1ꢀ4)ꢀOꢀ1,2,3,6ꢀtetraꢀOꢀacetylꢀ
D
ꢀglucopyranoside (16). Following the
general procedure for constructing the globotriaose, compound 15 (510 mg, 0.88 mmol) and
TMSI (142 ꢀL, 0.97 mmol) were used for galactosyl iodide formation. The forming iodide
CH2Cl2 solution (1mL) was then cannulated to the CH2Cl2 solution (1.5 mL) of 4’ꢀOH acceptor
12 (220 mg, 0.35 mmol) under the activation of AgOTf (270 mg, 1.05 mmol) at ꢀ78 ˚C to ꢀ30 for
4 h with the presence of activated 4 Å molecular sieves (800 mg). After workup, the reaction
mixture was then purified by gradient flash column chromatography (EA/Hex = 2/3 to 1/1) to
obtain 16 (300 mg, 75%) as a white foam (Table 1, entry 1). The product consisted of
inseparable α/β anomeric acetates and the major βꢀanomer is reported: TLC (EA/Hex = 1/1) Rf
1
0.43. H NMR (800 MHz, CDCl3) δ 7.43ꢀ7.24 (m, 20H, ArH), 5.65 (d, J = 8.9 Hz, 1H, Hꢀ1),
5.20 (appt. t., J = 8.9 Hz, 1H, Hꢀ3), 5.13 (dd, J = 10.2, 7.9 Hz, 1H, Hꢀ2’), 5.04 (appt. t, J = 8.9
Hz, 1H, Hꢀ2), 4.92 (d, J = 12.0 Hz, 1H, PhCH), 4.86ꢀ4.80 (m, 2H, PhCH2), 4.76ꢀ4.73 (m, 3H, Hꢀ
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 39
1
2
3
4
5
6
7
8
9
1’’, Hꢀ3’, PhCH), 4.67 (d, J = 12 Hz, 1H, PhCH), 4.56 (d, J = 12 Hz, 1H, PhCH), 4.50 – 4.40 (m,
4H, Hꢀ6a, Hꢀ6’ab, PhCH× 2), 4.39 (d, J = 7.9 Hz, 1H, Hꢀ1’), 4.27 (dd, J = 9.0, 4.7 Hz, 1H, Hꢀ
5’’), 4.16 (s, 1H, Hꢀ4’’), 4.10 (dd, J = 12.3, 4.6 Hz, 1H, Hꢀ6b), 4.05 (m, 1H, Hꢀ2’’), 3.97 (d, J =
2.3, 1H), 3.77 (appt. t, J = 8.9 Hz, 1H, Hꢀ4), 3.72 (dd, J = 10.0, 4.7 Hz, 1H, Hꢀ5), 3.64 (appt. t, J
= 6.8 Hz, 1H, Hꢀ5’), 3.61 (appt. t, J = 8.6 Hz, 1H, Hꢀ6’’a), 3.42 (dd, J = 8.6, 4.9 Hz, 1H, Hꢀ6’’b),
2.11 (s, 3H), 2.09 (s, 3H), 2.04 (s, 6H), 2.03 (s, 3H), 2.02 (s, 3H), 1.86 (s, 3H), 1.85 (s, 3H). 13C
NMR (200 MHz, CDCl3) δ 170.7, 170.4, 170.3, 169.8, 169.5, 168.9, 168.7, 138.8, 138.7, 138.1,
137.9, 128.4, 128.33, 128.30, 128.2, 128.15, 128.11, 128.0, 127.7, 127.6, 127.4, 127.36, 127.34,
101.4, 101.0, 91.7, 79.1, 76.0, 75.5, 75.3, 75.0, 74.6, 74.2, 73.6, 73.3, 72.8, 72.5, 72.3, 72.2, 70.3,
69.6, 69.4, 67.6, 61.7, 61.0, 20.84, 20.82, 20.6, 20.5. HRMS (ESIꢀion trap): m/z calcd for
[C60H70O23+Na+] 1181.4200; found 1181.4241.
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
(2,3,4,6ꢀTetraꢀOꢀbenzylꢀαꢀ
D
ꢀgalactopyranosyl)ꢀ(1ꢀ3)ꢀOꢀ(2,6ꢀdiꢀOꢀacetylꢀβꢀOꢀ
Dꢀ
galactopyranosyl)ꢀ(1ꢀ4)ꢀOꢀ1,2,3,6ꢀtetraꢀOꢀacetylꢀ
D
ꢀglucopyranoside (17).
Following the
general procedure for constructing the isoglobotriaose scaffold, compound 15 (183 mg, 0.32
mmol) and TMSI (50 ꢀL, 0.36 mmol) were used for galactosyl iodide formation. The CH2Cl2
solution (1.5 mL) of forming iodide was then cannulated to the CH2Cl2 solution (1.5 mL) of
3’,4’ꢀdiꢀOH acceptor 13 (75 mg, 0.13 mmol) under the activation of AgOTf (103 mg, 0.42 mmol)
at ꢀ78 ˚C to ꢀ30 for 4 h with the presence of activated 4 Å molecular sieves (300 mg). After
workup, the reaction mixture was then purified by flash column chromatography (EA/Hex =
50:50) to obtain 17 (37 mg, 26%) as a white amorphous foam and compound 18 (63 mg, 31%) as
a colorless oil (Table 1, entry 3). The product consisted of inseparable α/β anomeric acetates
1
and the major βꢀanomer is reported: TLC (EA/Hex = 1/1) Rf 0.43. H NMR (800 MHz, CDCl3)
28
ACS Paragon Plus Environment

Page 29 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
δ 7.41 – 7.34 (m, 5H, ArH), 7.34 – 7.20 (m, 15H, ArH), 5.64 (d, J = 8.3 Hz, 1H, Hꢀ1), 5.21 (appt.
t, J = 9.4 Hz, 1H, Hꢀ3), 5.06ꢀ5.03 (m, 2H, Hꢀ2, Hꢀ2’), 4.88 (d, J = 11.6 Hz, 1H, PhCH), 4.86 (d,
J = 11.4, 1H, PhCH), 4.74 (s, 2H, PhCH2), 4.68 (d, J = 3.7 Hz, 1H, Hꢀ1”), 4.63 (d, J = 11.6 Hz,
1H, PhCH), 4.52 (d, J = 11.4 Hz, 1H, PhCH), 4.46 – 4.40 (m, 2H, Hꢀ6a, PhCH), 4.33 (d, J = 11.9
Hz, 1H, PhCH), 4.30 – 4.24 (m, 3H, Hꢀ1’, Hꢀ6’ab), 4.15 – 4.09 (m, 1H, Hꢀ6b), 4.01 (dd, J = 9.6,
3.8 Hz, 1H, Hꢀ2”), 3.95 – 3.89 (m, 2H, Hꢀ4’, Hꢀ3”), 3.81ꢀ3.78 (m, 2H, Hꢀ4, Hꢀ5”), 3.76 – 3.69
(m, 2H, Hꢀ5, Hꢀ4’), 3.60 – 3.55 (m, 2H, Hꢀ5’, Hꢀ3’), 3.43 – 3.36 (m, 2H, Hꢀ6”ab), 2.14 (s, 3H),
2.10 (s, 3H), 2.09 (s, 3H), 2.04 (s, 3H), 2.02 (s, 3H), 1.95 (s, 3H). 13C NMR (200 MHz, CDCl3)
δ 170.8, 170.4, 169.9, 169.6, 169.5, 168.9, 138.3, 137.6, 137.4, 128.6, 128.57, 128.46, 128.4,
128.3, 128.27, 128.24, 127.9, 127.8, 127.7, 127.66, 127.5, 100.4, 96.5, 91.7, 79.6, 78.5, 75.3,
74.9, 74.8, 74.7, 74.3, 73.7, 73.5, 72.9, 72.1, 71.9, 70.7, 70.4, 70.1, 68.8, 65.0, 63.1, 61.8, 20.9,
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
+
20.8, 20.7, 20.67, 20.63. HRMS (ESIꢀion trap): m/z calcd for [C58H68O22+NH4 ] 1134.4540;
found 1134.4445.
(2,3,4,6ꢀTetraꢀOꢀbenzylꢀαꢀ
D
ꢀgalactopyranosyl)ꢀ(1ꢀ3)ꢀOꢀ(2,3,6ꢀtriꢀOꢀacetylꢀ4ꢀ(2,3,4,6ꢀTetraꢀ
ꢀgalactopyranosyl)ꢀ(1ꢀ4)ꢀOꢀ1,2,3,6ꢀtetraꢀOꢀacetylꢀDꢀ
Oꢀbenzylꢀαꢀ ꢀgalactopyranosyl)ꢀβꢀOꢀ
D
D
glucopyranoside (18). Compound 18 (63 mg, 31%) was obtained as a colorless oil (Table 1,
entry 3; see compound 17 and general procedure for constructing isoglobotriaose for reaction
and purification conditions). The product consisted of inseparable α/β anomeric acetates and the
1
major βꢀanomer is reported: TLC (EA/Hex = 1/1) Rf 0.63. H NMR (600 MHz, methanolꢀd4) δ
5.71 (d, J = 8.3 Hz, 1H, Hꢀ1), 5.29 – 5.20 (m, 2H), 5.07 (appt. t , J = 8.9 Hz, 1H), 4.98ꢀ4.92 (m,
3H), 4.80 (d, J = 11.2 Hz, 1H), 4.77 – 4.70 (m, 2H), 4.65 (d, J = 11.3 Hz, 2H), 4.59 – 4.23 (m,
18H), 4.20 (d, J = 8.3 Hz, 1H), 4.13 (dd, J = 12.3, 5.1 Hz, 1H), 4.10 – 3.98 (m, 5H), 3.97 – 3.75
29
ACS Paragon Plus Environment