Semisynthesis of Intact Complex-Type Triantennary Oligosaccharides

Feb 29, 2016 - ... load: https://cdn.mathjax.org/mathjax/contrib/a11y/accessibility-menu.js .... Complex-Type Triantennary Oligosaccharides from a Bia...
0 downloads 0 Views 3MB Size
Subscriber access provided by UNIV OF NEWCASTLE

Article

Semi-synthesis of intact complex-type triantennary oligosaccharides from a biantennary oligosaccha-ride isolated from a natural source by selective chemical and enzymatic glycosylation Yuta Maki, Ryo Okamoto, Masayuki Izumi, Takefumi Murase, and Yasuhiro Kajihara J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b13098 • Publication Date (Web): 29 Feb 2016 Downloaded from http://pubs.acs.org on March 1, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of the American Chemical Society 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 10

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

Journal of the American Chemical Society

Semi-synthesis of intact complex-type triantennary oligosaccharides from a biantennary oligosaccharide isolated from a natural source by selective chemical and enzymatic glycosylation Yuta Maki,1 Ryo Yasuhiro Kajihara*,1

Okamoto,1

Masayuki

Izumi,1

Takefumi

Murase,2,3

and

1Department

of Chemistry, Graduate School of Science, Osaka University, 1-1 Machikaneyama-cho, Toyonaka-city, Osaka 560-0043, Japan. 2International

Graduate School of Arts and Science, Yokohama City University, 22-2 Seto, Kanazawa-ku, Yokohama, Kanagawa 236-0027, Japan ABSTRACT: Attachment of oligosaccharides to proteins is a major posttranslation al modification. Chemical syntheses of oligosaccharides have contributed to clarifying the functions of these oligosaccharides. However, syntheses of oligosaccharides-linked proteins are still challenging because of their inherent complicated structures including diverse di- to tetra-antennary forms. We report a highly efficient strategy to access the representative two types of triantennary oligosaccharides through only 9- or 10-step chemical conversions from a biantennary oligosaccharide, which can be isolated as an exceptional homogeneous form from egg yolk. Four benzylidene acetals were successfully introduced to the terminal two galactosides and two core -mannosides of the biantennary asialo-nonasaccharide bearing 24 hydroxy groups, followed by protectio n of the remaining hydroxy groups by acetyl group. Selective removal of one of the benzylidene acetals gave two types of suitably protected glycosyl acceptors. Glycosylation toward the individual acceptors with protected Gal-β-1,4-GlcN thioglycoside, and subsequent deprotection steps successfully yielded two types of complex-type triantennary oligosaccharides.

Modification of proteins on cell surfaces and in body fluids with oligosaccharides is one of the major posttranslational modifications.1,2 Oligosaccharides not only regulate functions of glycoproteins, but also act as ligands to receptors, and contribute in several biological events such as immune responses, activation of cell-cell interactions, and endocytosis.3,4 However, as functional studies have been lacking due to the scarcity of oligosaccharides, as well as due to considerable heterogeneities in the structures of oligosaccharides isolated from natural sources,5 chemical syntheses have contributed greatly to provide homogeneous oligosaccharides.6 Previously established regioselective protection/deprotection methods for multiple hydroxy groups of sugars and highly efficient glycosylation reactions have enabled the synthesis of numerous oligosaccharides and glycoconjugates, such as glycolipids and glycoproteins.7-10 These wellestablished chemical methods, including chemoenzymatic syntheses, afford homogeneous oligosaccha-

rides to reveal their roles in biological events at molecular level;11 e.g. oligosaccharide-arrays prepared by using chemically synthesized oligosaccharides are used to elucidate interactions between an oligosaccharide and a protein.12,13 The chemical synthesis of an oligosaccharide requires an efficient synthetic strategy,6 as it is unfortunately a time-consuming protocol due to the need for repetitive protection/deprotection of hydroxy groups, as well as because of the stepwise construction and purification of stereoselective glycosidic linkages. Frequently found on mature glycoproteins, a complex-type oligosaccharide is a representative oligosaccharide form on glycoproteins and exhibits unique antennary structures. The reducing terminal oligosaccharide links with the nitrogen atom of an asparagine side chain. It is known that changing the number of antennae alters physiological properties and biological activities of glycoproteins, as seen in the case of

1 ACS Paragon Plus Environment



Journal of the American Chemical Society

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

f HO

HO O HO

HO HO

NHAc O

O

AcHN OH O HO O HOO O HO HO HO

HO

d O

j

k

HO

OH O

HO

HO HO HO OH O

HO O HO

HO HO HO

H N

O

AcHN

O OH

g

FmocHN

O

1

NHAc O

k’ HO

OH O

O

O

O

HO HO HO

OH O

NHAc O HO

HO O HO

e

OH O HO

HO

OH O

h

f HO

a

b

NHAc

i HO

c

O HO O

O HO

HO

erythropoietin (EPO).14 In the case of cancer cells, it is known that the number of branches increases depending on the tumor stage.15 However, specific details of functions of antennary structures still remain unclear.

e

OH O

HO

c

O

HO O OH O HO O O g AcHN HO OH OH O O O O NHAc HO HO

a

b

OH O

NHAc O O HO

HO O HO

O HO

OH O

H N

O

AcHN

O OH

FmocHN

2

h

Selective protections/deprotections Chemical glycosylation

O

e

O

O AcO O AcO AcO AcO

NHAc O

O

f

d

O

O

O AcO

c

AcO O AcO O O O

O O

AcO

AcO

O AcO

O

b

OAc O AcO O AcO

O

a NHAc O

O AcO

OAc O H N

O

AcHN

O O

g FmocHN

OAc O O

O

5

NHAc

h

i

Sequential protections f HO HO

OH O HO

e HO O HO

Isolation from a natural source

NHAc O

O

d

HO HO HO

O O HO O

HO HO HO HO OH O HO HO

i

O HO

OH O AcHN

h

O O

g

Syntheses of complex-type tri- and tetra-antennary oligosaccharides have been reported by the Danishefsky,16,17 Wong,18 Boons,19 and Unverzagt20 groups; however, these syntheses needed many conversion steps and largely relied on the sophisticated synthetic abilities of the researchers. The preparation of complex-type tri- and tetra-antennary oligosaccharides is a very difficult task, as to the best of our knowledge, even biantennary complex-type oligosaccharides have never been synthesized in less than 20 synthetic steps.21,22

d

O

j’

i

Page 2 of 10

c OH O

b HO O HO

NHAc O

O HO

a OH O

H N

O

AcHN

O OH

FmocHN

3. Asn-linked biantennary complex-type nonasaccharide

Figure 1. A synthetic strategy of typical human-type triantennary oligosaccharides.

Under these circumstances, we have been studying an alternative concept for the semi-syntheses of naturally occurring complex-type multi-antennary oligosaccharides starting from biantennary asialononasaccharide 3, which can be isolated in sufficient amount from egg yolk.23,24 The biantennary oligosaccharide 3 contains two terminal Galf,i-β-1,4-GlcNAce,h linked to two mannoside ends of core pentasaccharide: Mang-α-1,3-(Mand-α-1,6)-Manc-β-1,4-GlcNAcb-β-1,4GlcNAca-Asn (3, Figure 1), via β-1,2 linkage that is a basic structure of human complex-type oligosaccharides. There are two forms of natural triantennary structures 1 and 2, which have an additional Gal-β1,4-GlcNAc branch connected to the terminal mannosides of the core pentasaccharide through either β-1,6 or β-1,4 linkage, respectively (Figure 1). We envisioned that the desired triantennary forms could be synthesized by a single glycosylation reaction of a specific hydroxy group of the biantennary structure. Once such a semi-synthesis of triantennary oligosaccharides from the biantennary oligosaccharide is performed, the number of chemical synthetic steps will be dramatically reduced compared to that of the conventional step-wise chemical syntheses, and we will be able to easily obtain a practical amount of triantennary oligosaccharides. In order to accomplish this semi-synthesis, however, we need to develop selective protection/deprotection protocols toward 24 hydroxy groups in the biantennary oligosaccharide 3. So far, the concept of semi-synthesis has been demonstrated in the synthesis of many useful compounds,25,26 while large oligosaccharides bearing many hydroxy groups as well as other functional groups have never been used as substrate in semi-synthesis. Semi-syntheses of antibiotics and modified proteins usually employ regioselective reactions toward a specific functional group exposed on a substrate that is isolated from a natural source.27 In terms of sugar derivatives, semi-synthesis has been performed on mono- and di-saccharide units.28-32 The specific protection/deprotection protocols for mono- and di-saccharides have been developed, and

2 ACS Paragon Plus Environment

Page 3 of 10

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

Journal of the American Chemical Society

this chemistry enabled us to perform the stepwise construction of large oligosaccharides.

Scheme 1 3

Here we describe the new semi-synthesis of the two naturally occurring triantennary complex-type oligosaccharides 1 and 2 by 9- and 10-step conversions, respectively, by manipulating 24 hydroxy groups as well as other functional groups of the biantennary oligosaccharide 3 isolated from a natural source.

i O

O O

HO

HO

HO O HO

NHAc O

O

O

O

O HO

OH O

O HO O HO O O

RESULTS AND DISCUSSION

O

O O

HO

Our synthesis began with the sequential selective protections of 24 hydroxy groups of asialononasaccharide 3, which was prepared from chicken egg yolk as previously reported.23 A specific hydroxy group needs to be retained by selectively protecting other hydroxy groups, thus allowing the conversion of this oligosaccharide 3 into suitable glycosyl acceptors. Such acceptors can then be used for a glycosylation reaction in order to yield a triantennary oligosaccharide. First, we examined the selective protection of primary hydroxy groups, but these attempts were not successful because of the low solubility of asialononasaccharide 3 in organic solvent and the similar reactivity of the many hydroxy groups of the molecule. Trityl, tert-butyldimethyl silyl, and tert-butyldiphenyl silyl groups were tested for 6-OH protection (Figure S15-1 and -2). When these reactions were conducted with a few equivalents of reagents to primary alcohols, they did not yield suitable results. Although most of the substrate remained as a precipitate, a small amount of the substrate dissolved into the solvent, allowing the reaction to proceed. Diluting the solvent did not improve any conditions. Furthermore, using an excess of reagents caused over-protection of the substrate dissolved in the solvent. As a result, we could not regulate any selective protection of primary and secondary hydroxy groups among 24 hydroxy groups. Next, we examined the selective protections with benzylidene acetal toward primary 6-OH and secondary 4-OH groups of both Mand,g and Galf,i of oligosaccharide 3. After intensive optimization of the reaction conditions, we succeeded in obtaining tetrabenzylidene derivative 4 in which eight hydroxy groups out of twenty four were selectively protected, including two hydroxy groups (Mand-6-OH and Mang4-OH), which we intended to use for future glycosylation (Scheme 1, Figure S1 and S2). Benzylidenation reaction usually promotes an equilibrium state, giving a thermodynamically stable product with the protection of specific 1,3-diol among multiple hydroxy groups. In the case of the oligosaccharide, we presumed that 1,3-diol protection with an equilibrium system was more applicable than the protection of primary alcohol

O HO

HO

NHAc

HO O HO

O HO

O

O

OH O

OH O

4

O

H N

O

AcHN

O OH

FmocHN

NHAc

ii, iii

5

Reagents and conditions: (i) PhCH(OMe)2, CSA, DMF, rt, 17 h. (ii) Phenacyl Br, iPrN2Et, DMF, rt, 3 h. (iii) Ac2O/Pyridine (1:1), DMAP, rt, 3 h, 18% (over 3 steps). (a) H

O

e

O

O AcO O AcO AcO

AcO

f

NHAc O

O

d

H O O AcO

c

O O AcO O

AcO O O H

O

H

O O

AcO

AcO

O AcO

i

OAc O O

O

OAc O AcO O AcO

b

a

NHAc O

O AcO

OAc O H N

O

AcHN

g

O O

FmocHN

O

5

NHAc

h

(b) ppm

HMBC

HSQC

62

Mand-5 Mang-5 Galf,i-5

64 66 68

Mand,g-6 Mand,g-6’ Galf,i-6,6’ Galf,i-4

70 72 74 76

Mand,g-4 5.20 5.15 5.10 ppm 5.0 4.8 4.6 4.4 4.2 4.0 3.8 3.6 3.4 3.2 3.0

78 80 ppm

Figure 2. (a) Structure of fully protected oligosaccharide 5. The observed HMBC correlation signals are shown by red arrows. (b) Selective regions of HSQC (right square) and HMBC (left square) of 5. The HMBC correlation signals between the benzyl protons of benzylidene acetals and the C-4 and -6 were observed in Mand,g, and similar correlation signals were also confirmed in Galf,i.

3 ACS Paragon Plus Environment

Journal of the American Chemical Society

(a) H

5 (peak A)

O

H

O

O AcO O AcO AcO AcO

60% aq. AcOH, rt, 9-11 h, 9-17% 6, 13-23% 7, 8-18% 8, 10-17% 5.

O

O

HO HO

f

d

O

AcO

AcO O AcO

R1

NHAc O

O AcO

OAc O H N AcHN

O

H

O

O

H

O O

OPac

AcO

FmocHN

AcO

O AcO

H

O O

AcO

O

O

AcO O

HO R 2O

g H

O

O O

AcO

NHAc

i

6 (peak B)

AcO

O AcO

O

OAc O

O O AcO O

OR1

OAc O

OR1

AcO

g

H

O

OAc O O

HO HO

O O

AcO

NHAc

AcO

O

OAc O O

O AcO

7 (peak C) : R2 = H

BzCN, iPr2NEt, CH2Cl2, 48%

O

AcO

AcO

OAc O O

NHAc O

HO HO

c

O AcO O

OR1

AcO O AcO

d

O O AcO

O

O

NHAc

H

c

i

(b)

O

f OAc O

AcO O AcO O O

O

O AcO O AcO AcO AcO

NHAc

O

NHAc

8 (peak D)

9 : R2 = Bz

(c) A

ppm

HMBC

Relative absorbance at 220 nm

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

Page 4 of 10

HSQC

Mand-6,6’’

62

Mand-5 Mang-5 Galf,i-5

2h

64 66 68

C

Mang-6 Galf,i-6,6’

A

D

Mang-6’

70 72

B

Galf,i-4 Mand-4 Mang-4

76

9h 0

Retention time/ min

20

74

78 5.5 5.4 5.3 5.2 ppm

5.0

4.8

4.6

4.4

4.2

4.0

3.8

3.6

3.4

3.2

80 ppm

Figure 3. (a) The structures of partially debenzylidenated oligosaccharides 6, 7, and 8. The observed HMBC correlation signals are shown by red arrows. Mang-6-OH of diol 7 was further protected with BzCN in CH2Cl2 to afford glycosyl acceptor 9 (48%). (b) The monitoring of selective debenzylidenation reaction by RP-HPLC analysis. (c) Characterization of glycosyl acceptor 6. The selective regions of HSQC (right square) and HMBC (left square) spectra of 6 are shown. among 24 hydroxy group. Subsequent phenacyl esterification of the carboxylic acid in the asparagine moiety, followed by the peracetylation of residual hydroxy groups, afforded fully protected oligosaccharide 5 (18% over 3 steps). In order to determine the positions of the benzylidene acetals, 2D NMR experiments were conducted. The HSQC and HMBC spectra of oligosaccharide 5 are shown in Figure 2. HMBC correlation signals between the benzyl protons of benzylidene acetals and the C-4 and -6 of Galf,i were observed. Similar correlation signals were found in Mand,g as well. These correlation patterns indicated that benzylidene acetals were exclusively introduced to the 4- and 6-OH of Galf,i and Mand,g. In terms of phenacyl ester of asparagine, this ester was found to form a cyclic product (aspartimide) via a nucleophilic attack of N-glycosylated side-chain nitrogen under basic conditions, and on purification with ammonium acetate buffer. This aspartimide for-

mation was also observed during glycopeptide syntheses by use of oligosaccharyl-asparagine derivatives.33 To prevent this side reaction, fractions of purified oligosaccharide 5 were kept on ice during the HPLCpurification step and were then lyophilized as soon as possible. Moreover, we took care to keep the reaction from strong basic conditions in order to prevent aspartimide formation. In addition to this protection protocol yielding oligosaccharide 5, in which Galf,i and Mand,g were protected with benzylidene acetals, we also examined the selective protection of only Mand,g with benzylidene acetal in order to yield suitable glycosyl acceptors. We conducted this step because the third Gal-β-1,4-GlcNAcantenna links with Mand-6-OH or Mang-4-OH in the target triantennary oligosaccharides 1 and 2, however, we could not find any selective protection protocol toward Mand and Mang residues. There were multiple products during benzylidenation of 5 (Figure S2a).

4 ACS Paragon Plus Environment

Page 5 of 10

O O HO

Ph

(a)

O

Ph

70

OH O

O O HO

60

Percentage of hydrolysis

80

ManA OMe

50 40

Ph

30

O O

20 10

O

HO 0

5

10

15

Reaction time/ h

(c)

20

25

GalA GalA

GlcB

90 80

Ph

70

ManB

40

Ph

30 20

O O

10 0

5

10

15

Reaction time/ h

20

AcO OMe

GalB GlcB 24 h

12 h

12 h

9h

6h

6h

3h

3h

5.74 5.72 5.70 5.68 5.66 5.64 5.62 5.60 5.58 5.56

1

O

AcO

25

GalB ManB

GlcA

1h

1h

< 5 min

< 5 min

H/ ppm

OAc O OMe

50

(d)

ManA

O O AcO

60

0

HO OMe

O AcO OMe

100

GlcA

90

O O AcO

Ph

(b)

HO OMe

100

Percentage of hydrolysis

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

Journal of the American Chemical Society

5.74 5.72 5.70 5.68 5.66 5.64 5.62 5.60 5.58 5.56

1

H/ ppm

Figure 4. (a) Hydrolysis rate of benzylidene group of methyl 4,6-O-benzylidene-α-D-glycosides. (b) Hydrolysis rate of benzylidene group of methyl 2,3-di-O-acetyl-4,6-O-benzylidene-α-D-glycosides. (c) Monitoring of the competitive reaction using mixture of methyl 4,6-O-benzylidene-α-D-glycosides. GlcA and ManA show faster decrease of benzyl proton of benzylidene group. (d) Monitoring of the competitive reaction using mixture of methyl 2,3-di-O-acetyl4,6-O-benzylidene-α-D-glycosides. GlcB and ManB show faster decrease of benzyl proton of benzylidene group. Therefore, we examined the selective deprotection of benzylidene acetals of fully protected oligosaccharide 5 in order to convert it into suitable acceptors for the syntheses of triantennary oligosaccharides. During the extensive investigation for the selective removal of benzylidene groups, we found that two benzylidene acetals of Mand,g residues are prone to deprotection in comparison to those of two Galf,i residues under mild acidic conditions. This unexpected finding enabled us to obtain biantennary oligosaccharyl acceptors 6 and 7 that were suitable for the semi-syntheses of triantennary oligosaccharides 1 and 2, respectively (Figure 3a and b). As shown in Figure 3b, this deprotection of benzylidene groups in mannosides gradually proceeded under the optimized condition using 60% aqueous acetic acid. The increase of two tri-benzylidene derivatives (peak B and C) and concomitant decrease of the tetra-benzylidene derivative 5 (peak A) were observed by HPLC/mass. When reaction time was extended, all four benzylidene acetals were removed. The maximum conversion yield of a mixture of two tribenzylidene derivatives was found to be 46% based on the peak areas of the HPLC chromatogram, and individual isomers were isolated in 9-17% (6, peak B, 12% in average) and 13-23% (7, peak C, 17% in average) yield by RP-HPLC.

In order to confirm which hydroxy groups were made free under acidic removal of a benzylidene acetal, the structures of these two isomers (peak B and C) were individually determined by 2D NMR experiments (Figure 3c and S4). According to these NMR assignments, we concluded that the structures corresponding to peak B and C were tri-benzylidene derivatives 6 and 7, respectively (Figure 3b). We also checked for the possibility of acetyl migration to newly formed hydroxy groups under this acidic condition, and we concluded that acetyl rearrangement did not occur based on NMR and HPLC analyses. For the synthesis of triantennary oligosaccharide 1, we decided to use diol 6 as a glycosyl acceptor to examine selective glycosylation toward primary Mand-6-OH, which is more reactive than Mand-4-OH. For the synthesis of triantennary oligosaccharide 2, primary Mang-6-OH in diol 7 was selectively benzoylated by treatment with BzCN to afford 48% yield of glycosyl acceptor 9, which had a free hydroxy group at the Mang-4-OH. This structure was also confirmed by NMR experiments, in which we observed that the chemical shift of H-6 and C-6 of Mang changed (Figure S10). To confirm the dependence of the benzylidene group on the sugar stereochemistry for selective deprotection, we investigated the deprotection of the benzylidene

5 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Page 6 of 10

Scheme 2 6 AcO AcO

9 AcO

OAc O

OAc O O SPh AcO AcO TrocHN

AcO

i

OAc O

OAc O O SPh AcO AcO TrocHN

10

10

Ph O

vi

Ph O

O

NHAc O AcO AcO O O O AcO AcO AcO OAc NHTroc O AcO O AcO O O d AcO AcO O HO j AcO O OAc O AcO AcO O O AcO AcO O Ph O Ph O O O OAc O O O O 11 AcO AcO NHAc AcO

O

O AcO O AcO AcO

AcO

NHAc O

Ph

O

O

O

O AcO AcO NHAc O O AcO

OAc O H N AcHN

AcO O

O OPac

j’

OAc O

OAc O AcO O

O AcO AcO TrocHN Ph O O

FmocHN

O AcO

AcO

O AcO

O AcO O O

BzO

OAc O AcO O AcO

NHAc O

O AcO

OAc O H N AcHN

g

O

O OPac

FmocHN

OAc O O

12

NHAc

ii, iii, iv, v

vii, viii, ix, x

1

2

Reagents and conditions: (i) 10, NIS, TfOH, CH2Cl2, 0 oC, 1 h, 47%. (ii) Zn, THF/AcOH/Ac2O (3:2:1), 0 oC to rt, 16 h. (iii) MeOH/aq. NaOH, 0 oC to rt, 2.5 h. (iv) FmocOSu, aq. NaHCO3, acetone, 0 oC to rt, 3.5 h. (v) aq. TFA, 0 oC, 10 min, 63% (over 4 steps). (vi) NIS, TfOH, CH2Cl2, 0 oC to rt, 2 h, 44%. (vii) Zn, THF/AcOH/Ac2O (3:2:1), 0 oC to rt, 18 h. (viii) MeOH /aq. NaOH, 0 oC to rt, 2 h. (ix) FmocOSu, aq. NaHCO3, acetone, 0 oC to rt 3 h. (x) aq. TFA, 0 oC, 10 min, 44% (over 4 steps). (a) ppm 66

HSQC

HMBC

Mand-6,6’

68

tested competitive hydrolysis reactions using an acidic solution containing three benzylidene derivatives. These competitive reactions also showed the same selectivity (Figure 4c, 4d, S13, and S14). We presumed that the selectivity towards oligosaccharide 5 was due to different stereochemistry at the 4-position.

70 72 74 4.8

4.7

4.6

4.5

4.4

HSQC

4.3 ppm

4.1

3.9

3.7

3.5

ppm

100

GlcNj-1 4.8

4.7

4.6

4.5

4.4

102 4.3 ppm

Figure 5. The selective regions of HSQC and HMBC spectra of 11. The HMBC correlation signal between the C-6 of Mand and the H-1 of GlcNj (dotted line) was observed. The newly synthesized β-linkage was determined by the coupling constant observed for H-1 and H-2 in GlcNj (8.5 Hz). group of monosaccharides (Figure 4). Six types of methyl 4,6-O-benzylidene-α-D-glycosides with and without Ac groups were prepared for this study. In both cases, the reactions using glucosides and mannosides were all faster than those using galactosides. We also

We then examined the glycosylation of suitably protected biantennary oligosaccharyl acceptors 6 and 9 with thioglycoside 10 (Scheme 2), which was prepared from lactose in 8 steps via an azidonitration reaction.34 For the synthesis of the triantennary oligosaccharide 11, the diol 6 was coupled with glycosyl donor 10 by TfOH/NIS activation. We found the best condition to employ acceptor 6 (5 mM in CH2Cl2), an excess of donor 10 (10 equiv), and NIS (10 equiv) for 1 h reaction time at 0 oC. This condition yielded the desired triantennary oligosaccharide 11 in 47% yield, while small amounts of byproducts, such as diglycosylated product, were also observed. The formation of this di-glycosylated product was accelerated when the amount of NIS was increased. In addition, this glycosylation reaction was prone to forming aspartimide derivative through intramolecular cyclization during long reaction time. For example, a condition using donor (2 equiv for acceptor) and NIS (2 equiv for donor) at -20 oC for 3.5 h did not complete the reaction and gave aspartimide (Figure S15-3).

6 ACS Paragon Plus Environment

Page 7 of 10

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

Journal of the American Chemical Society

Scheme 3

removed by treatment with NaOH in MeOH, and Fmoc groups were reintroduced to the asparagine moieties, respectively. The removal of remaining benzylidene acetals was carefully conducted with a brief treatment of 90% aq. TFA at 0 ºC, and the desired triantennary oligosaccharides 1 and 2 were obtained in 63% and 44% yield, respectively. High resolution mass and 2D NMR analyses were conducted (SI) and established beyond doubt that homogeneous triantennary oligosaccharides were obtained.

1 Isolated yield ca. 23% (Conversion yield 78%) OH

HO HO AcHN

O

O OH

HO

OH O

OH

O

O HO

OH OH O HO

OH

O HO

OH

HO AcHN

HO O HO

NHAc O

O

AcHN O HO O HOO O HO HO

O

HO

O HO O

HO HO AcHN

OH O O OH

CMP-Neu5Ac, α-2,6 sialyltransferase HEPES buffer

OH OH O O O HO HO HO

HO HO HO OH O

OH O

HO O HO

NHAc O O HO

OH O

H N

O

AcHN

O

O OH

FmocHN O

NHAc

13

Moreover, in this reaction, we also observed another byproduct exhibiting the same mass as the desired product 11 by HPLC/MS analysis. This byproduct was thought to possibly be an α-linked isomer or a regioisomer, however due to its negligible amount, we could not isolate and assign this byproduct. Contrary to the glycosylation of Mand-6-OH for the synthesis of triantennary oligosaccharide 11, the glycosylation of 4-OH of Mang for the synthesis of triantennary oligosaccharide 12 was slow (Scheme 2). Although the desired triantennary oligosaccharide 12 was obtained, the acceptor 9 (5 mM in CH2Cl2) still remained even when an excess of donor 10 (20 equiv) was employed and the reaction temperature was elevated to room temperature. Unreacted acceptor 9 was recovered during the purification step and used for the glycosylation reaction repeatedly. After optimization of this reaction, the desired triantennary oligosaccharide 12 was obtained in 44% yield after HPLC purification.

In addition, we examined enzymatic sialylation toward triantennary oligosaccharide 1 (Scheme 3), because mature complex-type oligosaccharides have sialic acid residues at the non-reducing termini, and sialic acid is known to be involved in many biological events. Enzymatic sialylation usually proceeds in a stereoand regio-selective manner in a single step.35 Because α-2,3 and α-2,6 sialyl linkages with the terminal galactoside are found in the structures of natural oligosaccharides, enzymatic sialylation by specific α-2,3 or α-2,6 sialyltransferase seems to be effective in making the desired sialyl-oligosaccharides. Wong and coworkers reported a combined enzymatic and chemical strategy for the syntheses of sialylated multiantennary complex-type oligosaccharides with their regeneration system of sugar nucleotides.18 In our application, sialic acid residues were installed to 1 by using α-2,6 sialyltransferase (P. damsela, [EC 2.4.99.1])36 in the presence of CMP-Neu5Ac. This sialylation reaction proceeded in 78% conversion yield. The structure of triantennary sialy-oligosaccharide 13 was also confirmed by NMR experiments and mass spectrometry. CONCLUSION

The structures of triantennary oligosaccharides 11 and 12 were confirmed by 2D NMR experiments. HMBC correlation between the anomeric proton of GlcNj and the C-6 of Mand was observed in 11 (Figure 5: dotted line). Similarly, HMBC correlation between the anomeric proton of GlcNj’ and the C-4 of Mang was also observed in 12 (Figure S6: dotted line). These HMBC correlation signals clearly indicate that we indeed obtained the protected triantennary oligosaccharides 11 and 12 by the semi-synthesis methods.

In summary, we have established a strategy for the semi-synthesis of complex-type triantennary oligosaccharides from a biantennary oligosaccharide isolated from a natural source. The sequential protection and deprotection of 24 hydroxy groups of this biantennary oligosaccharide successfully yielded two glycosyl acceptors bearing free hydroxy groups at the desired positions, which are essential for the semi-syntheses of two types of natural triantennary oligosaccharides. Because selective introduction of a protecting group is sometimes difficult for a complex molecule, the idea to regulate reactivity during the deprotection step was critical.

Finally, we examined deprotection reactions of 11 and 12 in order to obtain the desired intact triantennary asialo-undecasaccharides 1 and 2, respectively (Scheme 2). First, the Troc group was removed by treatment with Zn powder, followed by acetylation to yield acetamide derivatives (S3 and S6). Phenacyl group was also removed during this zinc treatment. Subsequently, all acyl groups and Fmoc group were

This new strategy significantly reduced the number of synthetic steps and overcame limitations of traditional oligosaccharide synthesis that relies on repetitive protection/deprotection and multiple glycosylation steps. In terms of the synthesis of triantennary oligosaccharide 1 having three Gal-β1,4-GlcNAc antennae through a β-1,6 and two β-1,2 linkages with the terminal core mannosides, more than 50 synthetic steps

7

ACS Paragon Plus Environment

Journal of the American Chemical Society

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

were eliminated compared to the previous conventional chemical oligosaccharide synthesis18. Because biantennary oligosaccharide can be isolated over gram scales and the synthetic routes of triantennary oligosaccharides are established once based on rigorous structural analyses, the synthesis of triantennary oligosaccharides can be repeatedly examined to promptly provide sufficient amounts, such as dozens of milligrams scale. Our current experiments demonstrated that biantennary nonasaccharide 3 (ca. 600 mg) could be converted into triantennary undecasaccharide 1 (4 mg) and 2 (2 mg). Isolation of biantennary nonasaccharide 3 from egg yolk can be easily performed over 20 g scale, and we consider there is no limitation to synthesize triantennary undecasaccharides. The work on the syntheses of glycoproteins bearing multi-antennary oligosaccharides prepared by the semi-synthesis reported here is currently in progress.

ASSOCIATED CONTENT Supporting Information All experimental details of the synthesis of disaccharide donor and triantennary oligosaccharides and investigation of the debenzylidenation reactions, including general procedures, spectroscopic and analytical data. The Supporting Information is available free of charge on the ACS Publications website.

AUTHOR INFORMATION Corresponding Author [email protected]

Present Addresses 3

GlyTech, Inc., 134, Chudoji Minamimachi KRP1-109 Shimogyoku, Kyoto-city, Kyoto 600-8813, Japan

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT The financial support from the Japan Society for the Promotion of Science (Grant-in-Aid for Creative Scientific Research No.26248040 and 23245037 to Y.K.) is acknowledged and appreciated. A fellowship for Y.M. from JSPS Research Fellowships for Young Scientists is also gratefully acknowledged. The authors thank GlyTech, Inc., and Dr. Takeshi Yamamoto (Japan Tobacco Inc.) for the support to prepare the biantennary oligosaccharide and sialyltransferase, respectively.

REFERENCES (1) Varki, A.; Cummings, R. D.; Esko, J. D.; Freeze, H. H.; Stanley, P.; Bertozzi, C. R.; Hart, G. W.; Etzler, M. E., Eds.; Essentials of Glycobiology; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, 2009.

Page 8 of 10

(2) Apweiler, R.; Hermjakob, H.; Sharon, N. Biochim. Biophys. Acta 1999, 1473, 4. (3) Lowe, J. B. Cell 2001, 104, 809. (4) Dwek, R. A. Chem. Rev. 1996, 96, 683. (5) Rudd, P. M.; Dwek, R. A. Crit. Rev. Biochem. Mol. Biol. 1997, 32, 1. (6) Boltje, T. J.; Buskas, T.; Boons, G. J. Nat. Chem. 2009, 1, 611. (7) Guo, J.; Ye, X. S. Molecules 2010, 15, 7235. (8) Toshima, K.; Tatsuta, K. Chem. Rev. 1993, 93, 1503. (9) Zhu, X.; Schmidt, R. R. Angew. Chem. Int. Ed. Engl. 2009, 48, 1900. (10) Unverzagt, C.; Kajihara, Y. Chem. Soc. Rev. 2013, 42, 4408. (11) Schmaltz, R. M.; Hanson, S. R.; Wong, C.-H. Chem. Rev. 2011, 111, 4259. (12) Blixt, O.; Head, S.; Mondala, T.; Scanlan, C.; Huflejt, M. E.; Alvarez, R.; Bryan, M. C.; Fazio, F.; Calarese, D.; Stevens, J.; Razi, N.; Stevens, D. J.; Skehel, J. J.; van Die, I.; Burton, D. R.; Wilson, I. A.; Cummings, R.; Bovin, N.; Wong, C. H.; Paulson, J. C. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 17033. (13) Stevens, J.; Blixt, O.; Paulson, J. C.; Wilson, I. A. Nat. Rev. Micro. 2006, 4, 857. (14) Higuchi, M.; Oh-eda, M.; Kuboniwa, H.; Tomonoh, K.; Shimonaka, Y.; Ochi, N. J. Biol. Chem. 1992, 267, 7703. (15) Dennis, J. W.; Laferte, S.; Waghorne, C.; Breitman, M. L.; Kerbel, R. S. Science 1987, 236, 582. (16) Walczak, M. A.; Hayashida, J.; Danishefsky, S. J. J. Am. Chem. Soc. 2013, 135, 4700. (17) Walczak, M. A.; Danishefsky, S. J. J. Am. Chem. Soc. 2012, 134, 16430. (18) Shivatare, S. S.; Chang, S. H.; Tsai, T. I.; Ren, C. T.; Chuang, H. Y.; Hsu, L.; Lin, C. W.; Li, S. T.; Wu, C. Y.; Wong, C. H. J. Am. Chem. Soc. 2013, 135, 15382. (19) Wang, Z.; Chinoy, Z. S.; Ambre, S. G.; Peng, W.; McBride, R.; de Vries, R. P.; Glushka, J.; Paulson, J. C.; Boons, G. J. Science 2013, 341, 379. (20) Unverzagt, C.; Gundel, G.; Eller, S.; Schuberth, R.; Seifert, J.; Weiss, H.; Niemietz, M.; Pischl, M.; Raps, C. Chem.-Eur. J. 2009, 15, 12292. (21) Ogawa, T.; Sugimoto, M.; Kitajima, T.; Sadozai, K. K.; Nukada, T. Tetrahedron Lett. 1986, 27, 5739. (22) Wu, B.; Hua, Z.; Warren, J. D.; Ranganathan, K.; Wan, Q.; Chen, G.; Tan, Z.; Chen, J.; Endo, A.; Danishefsky, S. J. Tetrahedron Lett. 2006, 47, 5577. (23) Kajihara, Y.; Suzuki, Y.; Yamamoto, N.; Sasaki, K.; Sakakibara, T.; Juneja, L. R. Chem.-Eur. J. 2004, 10, 971. (24) Seko, A.; Koketsu, M.; Nishizono, M.; Enoki, Y.; Ibrahim, H. R.; Juneja, L. R.; Kim, M.; Yamamoto, T. Biochim. Biophys. Acta 1997, 1335, 23. (25) Kirschning, A.; Hahn, F. Angew. Chem. Int. Ed. 2012, 51, 4012. (26) Altmann, K.-H.; Gaugaz, F.; Schiess, R. Mol. Divers. 2011, 15, 383. (27) Muir, T. W. Annu. Rev. Biochem 2003, 72, 249. (28) Kumar H. V, R.; Naruchi, K.; Miyoshi, R.; Hinou, H.; Nishimura, S.-I. Org. Lett. 2013, 15, 6278. (29) Lopin, C.; Jacquinet, J.-C. Angew. Chem. Int. Ed. 2006, 45, 2574. (30) Vibert, A.; Lopin-Bon, C.; Jacquinet, J.-C. Chem.-Eur. J. 2009, 15, 9561. (31) Kurita, K.; Matsumura, Y.; Takahara, H.; Hatta, K.; Shimojoh, M. Biomacromolecules 2011, 12, 2267. (32) Laezza, A.; Iadonisi, A.; Castro, C. D.; De Rosa, M.; Schiraldi, C.; Parrilli, M.; Bedini, E. Biomacromolecules

8

ACS Paragon Plus Environment

Page 9 of 10

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

Journal of the American Chemical Society

2015, 16, 2237. (33) Yamamoto, N.; Takayanagi, A.; Sakakibara, T.; Dawson, P. E.; Kajihara, Y. Tetrahedron Lett. 2006, 47, 1341. (34) Kajihara, Y.; Kodama, H.; Wakabayashi, T.; Sato, K.-i.; Hashimoto, H. Carbohydr. Res. 1993, 247, 179. (35) Izumi, M.; Wong, C.-H. Trends Glycosci. and Glycotechnol. 2001, 13, 345. (36) Yamamoto, T.; Nakashizuka, M.; Kodama, H.; Kajihara, Y.; Terada, I. J. Biochem. 1996, 120, 104.

9

ACS Paragon Plus Environment

HO HO

1 2 3

OH O HO

HO OH O HO HO

HO

Journal of the American ChemicalPage Society 10 of 10

HO O HO

NHAc

HO HO HO

HO HO HO O HO

OH O

AcHN

HO

O

O

Chemical Synthesis

O O HO O O O

OH O

HO O HO

NHAc O

O HO

OH O

H N

O

HO HO

OH O

HO

HO O HO

NHAc

O

AcHN OH O HO O HOO O HO HO HO

O

O O HO O

OH O

HO O HO

ACS Paragon Plus Environment AcHN

FmocHN

Isolation from a natural source

O

OH

HO HO

OH O HO

O HO

HO HO HO OH O

O

O

NHAc

NHAc O HO

O

OH O AcHN

H N

O

FmocHN

O OH