Chemical Synthesis of Oligosaccharides Related to the Cell Walls of Plants and Algae

Downloaded from orbit.dtu.dk on: Sep 26, 2021

Chemical Synthesis of Oligosaccharides related to the Cell Walls of Plants and Algae

Kinnaert, Christine; Daugaard, Mathilde; Nami, Faranak; Clausen, Mads Hartvig

Published in: Chemical Reviews

Link to article, DOI: 10.1021/acs.chemrev.7b00162

Publication date: 2017

Document Version Peer reviewed version

Link back to DTU Orbit

Citation (APA): Kinnaert, C., Daugaard, M., Nami, F., & Clausen, M. H. (2017). Chemical Synthesis of Oligosaccharides related to the Cell Walls of Plants and Algae. Chemical Reviews, 117(17), 11337-11405. https://doi.org/10.1021/acs.chemrev.7b00162

General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

 Users may download and print one copy of any publication from the public portal for the purpose of private study or research.  You may not further distribute the material or use it for any profit-making activity or commercial gain  You may freely distribute the URL identifying the publication in the public portal

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Chemical Synthesis of Oligosaccharides related to the Cell Walls of Plants and

Algae

Christine Kinnaert, Mathilde Daugaard, Faranak Nami, Mads H. Clausen*

Center for Nanomedicine and Theranostics, Department of Chemistry,

Technical University of Denmark, Kemitorvet, Building 207, 2800 Kgs. Lyngby, Denmark

[email protected]

Abstract

Plant cell walls are composed of an intricate network of polysaccharides and proteins that varies during the developmental stages of the cell. This makes it very challenging to address the functions of individual wall components in cells, especially for highly complex glycans. Fortunately, structurally defined oligosaccharides can be used as models for the glycans, to study processes such as cell wall biosynthesis, polysaccharide deposition, protein-carbohydrate interactions, and cell-cell adhesion. Synthetic chemists have focused on preparing such model compounds, as they can be produced in good quantities and with high purity. This review contains an overview of those plant and algal polysaccharides, which have been elucidated to date. The majority of the content is devoted to detailed summaries of the chemical syntheses of oligosaccharide fragments of cellulose, hemicellulose, pectin, and arabinogalactans, as well as glycans unique to algae. Representative synthetic routes within each class are discussed in detail and the progress in carbohydrate chemistry over recent decades is highlighted.

Contents

Introduction ...... 3 Architecture and composition of plant cell walls ...... 6 2.1 Polysaccharides in plant and algal cell walls ...... 8 2.2 The study of cell wall composition, function and evolution ...... 10 Cellulose ...... 12 Hemicellulose ...... 18 4.1 Xyloglucans ...... 20 4.2 β-(1→3)-D-Glucans ...... 28 4.3 Mixed-linkage [β-(1→3), β-(1→4)] glucans ...... 35 4.4 Xylans (arabinoxylans, glucuronoxylans and glucurono-arabinoxylans) ... 38 4.5 Mannans ...... 47 Pectin ...... 56 5.1 Homogalacturonan ...... 59 5.2 Xylogalacturonan ...... 66 5.3 Apiogalacturonan ...... 66 5.4 Rhamnogalacturonan I ...... 70 RG-I backbone ...... 71 RG-I side chains ...... 80 5.5 Rhamnogalacturonan II ...... 81 RG-II side chain A ...... 83 RG-II side chain B ...... 89 RG-II side chain C ...... 94 RG-II side chain D ...... 95 RG-II side chain E/F ...... 95 (Arabino)galactans and arabinans ...... 96 6.1 Type I AGs and galactans ...... 97 6.2 Type II AGs ...... 107 6.3 Arabinans ...... 119 Polysaccharides exclusive to algae ...... 124 7.1 Glucuronans ...... 124 7.2 Carrageenans ...... 125 7.3 Agarans ...... 127 7.4 Porphyrans ...... 127 7.5 Alginates ...... 128 7.6 Fucans ...... 137 7.7 Ulvans ...... 150 Conclusion ...... 151 References ...... 154 TOC Graphic ...... 183

Introduction

Plant cell walls are highly sophisticated fiber composite structures. They are dynamic systems, which have evolved to fulfill the wide range of biological functions necessary to support plant life. The wall is the outer layer of the cell, providing a strong and protective supporting boundary, and defining the size and shape of the cell. Beyond its structural function, the cell wall also plays an important role in cell expansion during plant growth, and in plant-microbe interactions, in particular the defense response against pathogens.1,2 The cell wall is not just a boundary between plant cells, but forms the interface to neighboring cells, contributing to intercellular communication.

Plant cell walls are highly heterogeneous, with a wide range of constituent components. Table 1 compares the monosaccharides found in mammalian cells with the ones occurring in plant and/or algal cell wall, together with the symbols used here and in the CFG nomenclature. The diversity can make it challenging to distinguish the function of individual components and how they interact. Researchers have therefore used well-defined oligosaccharides as model compounds to represent domains from the larger, more complex polysaccharides. Many groups have succeeded in synthesizing such oligosaccharides using different strategies. This review will introduce the glycans found in plant cell walls and describe all relevant chemical oligosaccharide synthesis, up to and including February 2017.

The material is organized in tables showing the structures according to glycan nomenclature given in

Figure 1. With the exception of some rare disaccharide combinations found in pectin, we present only trisaccharides and longer structures, and only oligosaccharides that have been deprotected successfully. Representative syntheses are discussed in detail, in order to include different approaches to plant cell wall glycan assembly. Some of the syntheses described here have been covered in

3 previous reviews; we are endeavoring to provide a comprehensive overview of the field, and therefore have chosen not to exclude these.

Table 1. Comparison between mammalian monosaccharides and plant/ algae cell wall monosaccharides.

Monosaccharide CFG symbol Symbol used here Plant/algal cell wall Mammalian glycans (existing or glycans suggested)

Glucose D D

Galactose D and L D

N-Acetyl- D glucosamine N-Acetyl- D galactosamine

Mannose D D

Galacturonic acid D D

Xylose D D

Fucose L L

Sialic acid D

Iduronic acid L L

Rhamnose L

Gulose L

Arabinose L

Apiose D

3-Deoxy-lyxo- heptulosaric acid D (Dha)

2-Keto-3-deoxy- manno- D octulosonic acid

(Kdo) Me Aceric acid L

L-rhamnose L-iduronic acid OAc β-notation

COOH α D-galactose L-arabinose -notation

COOMe 6 O L-fucose = D-3,6-anhydrogalactose 4 1 3 2 SO3H Me L-aceric acid less common enantiomer (D/L) D-glucose

HO Example: COOH D D-apiose O -xylose HO OMe HO = O D-3-deoxy-D-lyxo- HO D-mannose acid O heptulosaric (Dha) OH O OMe O HO OH L-gulose D-2-Keto-3-deoxy-D-manno- α- → α → - octulosonic acid (Kdo) = D-GalpA-(1 2)-[ -L-Araf-(1 4)]-β L-RhapOMe Figure 1. Glycan notation.

Architecture and composition of plant cell walls

As a simple model, plant cell walls can be grouped in two types: the primary and the secondary cell wall. The primary cell wall is a thin and flexible structure deposited when the cell is growing, while the secondary cell wall, much thicker and stronger, is deposited on certain cells when they have ceased to grow. Primary cell walls are typically composed of around 70% water. The dry material is made up of polysaccharides (approximately 90%) and (glyco)proteins (~ 10%). In the primary cell wall, cellulose microfibrils are cross-linked by non-cellulosic polysaccharides (often referred to as hemicellulose), and these are intertwined with pectic polysaccharides. The chemical composition of the main polysaccharides found in land plants is summarized in Table 2.

There is strong evidence that hemicellulose, mostly xyloglucans, bind to the cellulose microfibrils via two different mechanisms: being entwined in the microfibril during crystallization, which occurs immediately after synthesis; and by hydrogen bonding between the glycan chains.3–5 The polysaccharide network in the cell wall helps the plant resist turgor pressure and other deformations.

Less is known about the interaction of pectic polysaccharides with the other wall components, but there is evidence that neutral glycans side chains also interact with cellulose, a factor that is usually not accounted for in models describing plant cell wall architecture.6–8 Pectic polysaccharides are usually divided into four different domains (HG, RG-I, RG-II, and XGA / AGA, see Table 1), which are believed to be covalently linked to form the pectin macromolecule. Evidence suggests that some pectic polysaccharides are also crosslinked to hemicelluloses, phenolic compounds and wall proteins, which provides an even higher degree of complexity in the structure and function of the wall.5,9 A

2009 paper detailing the structure, function and biosynthesis of plant cell wall pectic polysaccharides by Mohnen & Caffall provides a good overview.6

Table 2. Cellulose, hemicellulose and pectin structures.

Cellulose β-(1→4)-Glup Hemicellulose Xyloglucans (XyG) β- Glucans Mixed-linkage Xylans Mannans (1→4)-Glup with β-(1→3)-Glup glucans (MLGs) β-(1→4)-Xylp β-(1→4)-Manp and mainly Xylp with β-(1→6) β-(1→3)-β-(1→4)- (potentially with Araf β-(1→4)-Glup/Manp branching branching Glup and GlupA branching) with Galp branching Pectin Homogalacturonan Rhamnogalacturonan I Rhamnogalacturonan Xylogalacturonan Apiogalacturonan (HG) (RG-I) II (RG-II) (XGA) (AGA) α-(1→4)-GalpA α-(1→4)-GalpA-α- α-(1→4)-GalpA α-(1→4)-GalpA α-(1→4)-GalpA (1→2)-Rhap backbone with conserved side with Xylp branching with Apif branching with side chains of chains of mainly rare Galp and Araf sugars

In addition to polysaccharides, the primary cell wall also contains about 10% protein, which is mostly glycosylated. Arabinogalactan proteins (AGPs) are some of the most abundant and most studied glycoproteins in plant cell walls and they are thought to be involved in recognition and signaling events at the cell surface.10 However, as the focus of this review is on polysaccharides, we will not describe the plant cell wall proteome. We invite the reader to look at some comprehensive reviews covering this subject.11,12

In secondary cell walls, the embedding material is not pectin, but rather the phenolic polymer lignin, which constitutes up to 30% of the dry weight of the wall.13 The secondary cell wall also contains cellulose and a network of connecting hemicellulosic polysaccharides with a lower degree of backbone branching (mainly xylan and mannans). The deposition of lignin results in dehydration of the wall, rendering it more hydrophobic and less dynamic. Other components such as the aliphatic polyesters cutin and suberin can also be found in the plant cell wall as minor components.14 Plant cell walls display a considerable degree of diversity in their composition and molecular architecture depending on plant species but also related to cell type and cell wall microstructure.15 An interesting

7 review of the heterogeneity in the chemistry, structure and function of plant cell walls has been published by Fincher and co-workers recently.15

2.1 Polysaccharides in plant and algal cell walls

The primary cell walls of seed plants and especially angiosperms (flowering plants) are by far the most studied. However, since land plants have algal ancestors, the cell walls of algae can be interesting from an evolutionary point of view. Land plants originate from freshwater green algae of the

Charophytes class, which emerged from their aquatic habitat approximately 450 million years ago.16,17,18 As shown in the phylogenetic tree of eukaryotic organisms (Figure 2), land plants

(Embryophytes), red algae (Rhodophytes) and green algae (Chlorophytes) all belong to the

Archaeplastida clade, while brown algae (Phaeophytes) belong to the separate Chromalveolata clade.

The phylogenetics can explain why the cell walls of brown algae share polysaccharides with both plants (cellulose) and animals (sulfated fucans), but also with some bacteria (alginates).19

Figure 2. Phylogenetic tree of eukaryotic organisms.

The nature of the different cell wall components from the major plant and algal families has been the topic of several reviews and is summarized in Table 3.16,20,21 The unique presence of different sulfated polysaccharides in the Rhodophytes and Phaeophytes is noteworthy. Agar, carrageenan and porphyran are the most important sulfated polysaccharides found in the red algae, while brown algae produce homofucans. The structure of these polysaccharides will be described in detail later in this review.

Table 3. Major cell wall polymers present in different plant and algal taxa.22

Taxa Chloroplastida Rhodophytes Phaeophytes

Embryophytes Charophytes Chlorophytes (red algae) (brown algae) (land plants) (green algae) (green algae) Polymer cellulose Crystalline mannan cellulose cellulose cellulose cellulose polysaccharides xylan (1→3)-β-D-xylan xyloglucan sulfated xyloglucan xyloglucan mannan mannans xylofucoglucan mannans mannans sulfated MLG Hemicellulose xylans sulfated xylofuco- xylans glucuronan (1→3), (1→4)-β-D- MLG glucuronan glucan (callose) glucan (callose) xylan glucan (callose) glucan (laminarin) Matrix carboxylic pectins pectins ulvans alginates polysaccharides Matrix agars sulfated ulvans carrageenans fucans polysaccharides porphyran

2.2 The study of cell wall composition, function and evolution

The study of the plant cell wall requires a multi-disciplinary approach and there are many challenges in trying to elucidate cell wall architecture, function, and evolution. The field is being advanced by technologies such as atomic force microscopy, immunofluorescence microscopy and the development of molecular probe sets.23 In a 2012 review, Willats and co-workers discussed a multiple-level strategy for correlating the occurrence of cell wall polysaccharides with genetic diversity in order to get an insight into underlying evolutionary mechanisms.24 Many of the most efficient techniques employ molecular probes that interact specifically with glycans, such as monoclonal antibodies (mAbs) and carbohydrate-binding modules (CBMs). Well-defined oligosaccharides provide the highest resolution possible when characterizing protein-carbohydrate interactions, and they contribute significantly to

10 the toolkit available for plant research. Figure 3 shows an example of in situ mapping of cellulose, rhamnogalacturonan I and methyl esterified homogalacturonan in seed coat epidermal cells and mucilage in Arapidopsis thaliana.

A B

Figure 3. Labelling of polysaccharidic components of seed coat epidermal cells and mucilage in Arabidopsis thaliana. (A) Composite image of a section through whole seed and mucilage. Scale bar = 100 µm. (B) Magnification of region in (A). Scale bar = 50 µm. Blue: calcofluor-labeled cellulose; red: INRA-RU1-immunolabelled rhamnogalacturonan; green: JIM7-immunolabelled highly methyl esterified homogalacturonan.23 Images used with permission from Adeline Berger.

Cellulose

Figure 4. Structure of cellulose.

Cellulose is the main component of the primary cell wall in plants. It is the most abundant organic compound on earth and consists of an unbranched homopolymer of β-(1→4)-glucan units (Figure 4).

The biopolymer is currently used in a wide variety of applications including the production of biofuels, paper products, and textile fibers.25 Cellulosic chains are held together tightly by both intra- and intermolecular hydrogen bonds facilitated by the conformation of the polymer. The chain length depends on the source of the cellulose; the degree of polymerization (DP) is believed to reach up to

8,000 glucose residues in primary cell walls and 15,000 in secondary cell walls. This makes cellulose one of the longest bio-macromolecules known.

Table 4 shows all chemically synthesized β-(1→4)-glucans with at least three glucose residues. It is worth noting that the majority of the reports on the preparation of β-(1→4)-glucans detail enzymatic procedures.

Table 4. The cellulose fragments synthesized chemically. Structure R Year Reference

26 OR Me 1981 Takeo et al.

27 1983 Takeo et al.

27 H 1983 Takeo et al.

28 (CH2)3(CH)2O 1990 Rodriguez and Stick

29 CH2CH2C10H7 1999 Xu and Vasella

30 (CH2)5NHCOO(CH2)3C8F17 2010 Chang et al.

31 (CH2)5NH2 2016 Dallabernardina et al.

27 OR Me 1983 Takeo et al.

32 (S)-CH-CH3CH2OH 1983 Sugawara et al. 1986 Sugawara et al.33

32 (R)-CH-CH3CH2OH 1983 Sugawara et al.

1986 Sugawara et al.34

32 (S)-CH2CH-OHCH3 1983 Sugawara et al.

1985 Nicolaou et al.33

1986 Sugawara et al.34

32 (R)-CH2CH-OHCH3 1983 Sugawara et al

1985 Nicolaou et al.33

1986 Sugawara et al.34

27 OR H 1983 Takeo et al.

35 1996 Nishimura and Nakatsubo

30 (CH2)5NHCOO(CH2)3C8F17 2010 Chang et al.

29 CH2CH2C10H7 1999 Xu and Vasella

33 OR (S)-CH2CH-OHCH3 1985 Nicolaou et al.

33 (R)-CH2CH-OHCH3 1985 Nicolaou et al.

31 OR (CH2)5NH2 2016 Dallabernardina et al. 2

33 OR (S)-CH2CH-OHCH3 1985 Nicolaou et al. 2 33 (R)-CH2CH-OHCH3 1985 Nicolaou et al.

35 OR H 1996 Nishimura and Nakatsubo 5 1999 Xu and Vasella29

In 1981 Takeo and co-workers reported the synthesis of β-methylcellotrioside using a Koenigs-Knorr glycosylation36 between building blocks 1 and 2 (Scheme 1).26 This produced the trisaccharide in 39% yield alongside unreacted 1 and 2. The trisaccharide was converted into the bromide 4 before methanolysis in the presence of mercuric cyanide, which gave 84% of 5. Zemplén deacetylation37 then resulted in the targeted cellotrioside 6 in 92% yield.

Scheme 1. Synthesis of β-methyl cellotrioside reported by Takeo and co-workers in 1981.26

OAc AcO OAc O AcO HO + AcO O O AcO OAc O AcO AcO AcO OAc Br 1 2

AgOTf OAc AcO OAc (CH3)2NCON(CH3)2 O AcO AcO O O O AcO O AcO OAc CH2Cl2, AcO OAc AcO 39% 3

OAc AcO OAc HBr in AcOH O AcO AcO O O O AcO O AcO 84% AcO OAc AcOBr 4

OAc AcO OAc Hg(CN)2 AcO O AcO O O O AcO O AcO OCH3 CH3OH:C6H6 1:10 AcO OAc AcO 84% 5

OH NaOMe OH OH HO O HO O O O OCH HO O HO 3 MeOH OH OH OH 92% 6

Nishimura and Nakatsubo published the first synthesis of cellooctaose using a convergent synthetic strategy in 1996.35 The synthesis is depicted in Scheme 2. The tetrasaccharide 8 (derived from allyl

3-O-benzyl-4-O-(p-methoxybenzyl)-2,6-di-O-pivaloyl-β-D-glucopyranoside 7 using a linear synthetic method described by Nishimura et al.38) was used as a common precursor for the tetrasaccharide acceptor 9 and Schmidt39 donor 11. These two were coupled under standard conditions

(BF3⋅OEt2) giving the octasaccharide 12 in 95% yield. The glycosylation was conducted in a high-vacuum system allowing for better control of the anhydrous reaction conditions. Protecting group manipulations afforded the fully protected oligosaccharide 14. The allyl group was removed using a one-pot procedure first reported by Kariyone and Yazawa40 before the anomeric position was acetylated. Catalytic hydrogenolysis followed by acetylation yielded the fully acetylated

15 octasaccharide 18. After purification, deacetylation afforded cellooctaose 19 in 87% yield. In a similar sequence, tetrasaccharide 8 was converted into cellotetraose 25 (see Scheme 2). More recently,

Pfrengle and co-workers produced linear β-(1→4)-glucans using solid-phase oligosaccharide synthesis.31 This work was part of a larger study of xyloglucan synthesis and will be described in section 4.1.

Scheme 2. Synthesis of cellotetraose and cellooctaose published by Nishimura and Nakatsubo.35

OPiv PivO OPiv PivO OPiv O BnO PMBO O BnO O NaOMe BnO OAll AcO O O O O BnO O BnO OAll MeOH PivO PivO OPiv PivO 8 OPiv 7 quant.

RO OR RO OR SeO2, AcOH, dioxane DBU, MeOH BnO O BnO RO O O O O O BnO O BnO OAll 79% 65% RO OR OR RO 20: R = H Ac2O, py, 82% 21: R = Ac PivO OPiv PivO OPiv BnO O BnO SeO2 HO O O O O O BnO O BnO OAll AcOH, dioxane PivO OPiv OPiv PivO 54% 9 AcO OAc AcO OAc R2O 2 O O R O O OPiv AcO 2 O O PivO PivO OPiv O R O O R2O BnO O BnO AcO OR1 AcO O O O OAc OAc AcO O BnO O O BnO , 1 2 PivO OR Pd/C, H EtOH:AcOH 4:1 22: R = H, R = Bn OPiv OPiv PivO 2 92% 23: R1 = R2 = H Ac 72% 10: R = H 24: R1 = R2 = Ac 2O, py, DBU, CCl3CN, CH2Cl2, 96% 11: R = C(NH)CCl3 DBU MeOH:CH2Cl2 1:4

75%

· OH BF3 OEt2, CH2Cl2 OH OH OH HO O HO HO O O O O 95% O HO O HO OH OH OH OH OH 25 PivO OPiv PivO OPiv BnO O BnO AcO O O O O O BnO O BnO OAll PivO OPiv OPiv PivO 3 12 NaOMe MeOH 78%

RO OR RO OR BnO O BnO RO O O O O O BnO O BnO OAll RO OR OR RO 3 o 13: R = H Ac2O, py, 50 C, quant. 14: R = Ac

SeO2 AcOH, dioxane

77%

AcO OAc AcO OAc R2O 2 O O R O O AcO 2 O O O R O O R2O AcO OR1 OAc OAc AcO 3 15: R1 = H, R2 = Bn Ac O, py, quant. 2 16: R1 = Ac, R2 = Bn , Pd(OH)2/C, H2 THF 17: R1 = Ac, R2 = H

Ac2O, py, 64% 18: R1 = R2 = Ac

DBU MeOH:CH2Cl2 1:4

87% OH OH OH OH OH OH OH OH HO O HO HO HO O O O O O HO O O HO O O O O O HO O HO O HO OH OH OH OH OH OH OH OH OH 19

Hemicellulose

Hemicellulose was first described in 1891 by Ernst Schulze, as the components of the cell wall that are easily dissolved in hot diluted mineral acids.41 Such a definition based on extractability was not, however, found to be useful, and these days the term is commonly used to refer to a group of non- starch, non-pectic polysaccharides found in association with cellulose in the cell walls of higher plants.

Based on current knowledge, hemicellulose is generally divided into four classes of structurally different cell-wall polysaccharides: xylans, mannans, xyloglucans and glucans including mixed linkage glucans (β-(1→3),(1→4)-glucans, MLGs) as shown in Figure 5.42 These are the types of polysaccharides described in most reviews about hemicellulose, but Peter Ulvskov and Henrik

Scheller, in their review of hemicellulose, further narrow down the definition to glycans sharing a β-

(1→4)-linked backbone with an equatorial configuration at the C-4 position. This excludes MLGs, callose and laminarin (β-(1→3)-linked glucans) from the hemicelluloses.4,42 Other polysaccharides, such as galactans, arabinans, and arabinogalactans are sometimes included in this group as well, but these will be discussed separately in section 6.42

L-rhamnose L-iduronic acid OAc β-notation

COOH α D-galactose L-arabinose -notation

COOMe 6 O L-fucose = D-3,6-anhydrogalactose 4 1 3 2 SO3H Me L-aceric acid less common enantiomer (D/L) D-glucose

D D-xylose -apiose

D-3-deoxy-D-lyxo- D-mannose heptulosaric acid (Dha)

L-gulose D-2-Keto-3-deoxy-D-manno- octulosonic acid (Kdo)

X X S G L X F G

Xyloglucans β-1,3-Glucans

n = 0,1,2

Mixed Linkage Glucans (MLGs)

Xylans Mannans

Arabinoxylan (AX) Galactomannans

MeO MeO

Glucuronoxylan (GX) Galactoglucomannans

MeO

Glucuronoarabinoxylan (GAX)

Figure 5. Overview of the different families of hemicellulose polysaccharides.

4.1 Xyloglucans

X X S G L X F G

Figure 6. Structure of XyGs with the most common side chains.

Xyloglucans (XyGs) constitute the most abundant class of hemicellulose in the primary cell wall of higher plants. They have been found in every species of land plant that has been analyzed, except charophytes.4 An exemplary structure of XyG with the different side chains is shown in Figure 6.

However, this schematic representation does not indicate the frequency of each side chain and many variations in this general pattern can be observed. XyGs have a backbone of β-D-(1→4)-glucans substituted by α-D-(1→6)-linked xylose residues. A special one-letter code is used to label the different XyG side chains. G denotes an unbranched Glcp residue, while X corresponds to an α-D-

Xylp-(1→6)-Glcp. The xylosyl residues can be further branched at O-2 with α-L-Araf (S) or β-Galp

(L), which can be further substituted at O-2 with α-L-Fucp (F). The galactose residues can be acetylated. The branches dictate the physical properties of the molecules; the fewer the branches, the less soluble the XyG. Only a few syntheses of XyGs with substituted xylose (side chains S, L or F) have been reported to date, including an LG fragment produced by Jacquinet and co-workers43 and an XXFG fragment synthesized by Ogawa and co-workers.44 In 2016, Pfrengle and co-workers achieved the automated solid phase synthesis of a variety of XyG oligomers without xylose substitution.31 Table 5 shows all the XyG fragments synthesized to date.

Table 5. The synthetic XyG fragments. Structure R Year Reference

31 (CH2)5NH2 2016 Dallabernardina et al.

Pr 1990 Wotovic et al.43

31 (CH2)5NH2 2016 Dallabernardina et al.

H 1990 Sakai et al.44

Me 2000 Watt et al.45

H 1990 Sakai et al.44

31 (CH2)5NH2 2016 Dallabernardina et al.

In 1990, Jacquinet and co-workers synthesized tetrasaccharide LG (38) by condensation of the two key disaccharides galactoxylose and cellobiose 33 and 36 respectively, as shown in Scheme 3. The building blocks 33 and 36 were synthesized through standard protection group manipulations and glycosylation reactions. Coupling of donor 33 and diol acceptor 36 afforded tetrasaccharide 37 in 50% yield as a single regio- and stereoisomer. However, the authors showed that steric hindrance around the O-6’ position of the acceptor had a dramatic impact on the regioselectivity as well as the yield of the reaction. Indeed, when using a diol acceptor with benzyl protecting groups in lieu of acetyl groups, only 10% of the undesired β-linked tetrasaccharide could be isolated, while none of the desired stereoisomer was observed.

Scheme 3. Synthesis of LG tetrasaccharide by Jacquinet and co-workers.43

1) 2,6-dimethylpyridinium AcO O BnO O perchlorate,ClCH2CH2Cl, AcO 1) NaOMe, MeOH BnO AllBr, PhCl BnO O O O BnO OAll O O OH Me 2) NaH, BnBr, DMF Me 2) NaOMe, MeOH OEt OEt 79% 74% 26 27 28

OBn BnO 1) 95% aq. AcOH BnO OBn 2) Ac2O, py O O BnO BnO O BnNH , Et O O 3) 2 2 AcO , DMF, CH Cl Cl Me 4) (COCl)2 2 2 OMe 29 57% 30

O O 1) Ac2O, pyridine O BnO OAll BnO O BnO BnO BnO 2) HgCl2/HgO BnO 28 O 1) NaOMe, MeOH O acetone/water O Cl BnO OBn BnO OBn BnO OBn 30 AgOTf O KOtBu, DMSO O 3) (COCl)2 O BnO 2) BnO BnO ClCH2CH2Cl ClCH2CH2Cl, DMF OAc OH OAc 80% 83% 31 32 70% 33

HO AcO AcO Ph O HO AcO 1) NaOMe, MeOH O 1) Ac2O, py AcO O O O HO O O O O O OAll O AcO OAll HO HO AcO OAll AcO AcO 2) benzaldehyde dimethylacetal OH 2) 60% aq. AcOH AcO AcO OAc OH OAc DMF 85% 34 35 36 75%

BnO O HO O BnO HO 36 O NaOMe, MeOH O BnO OBn 1) HO OH 33 O AcO O HO O Pd/C, H O AgOTf BnO HO O O 2) 2 HO HO O O O O OPr ClCH2CH2Cl OAc AcO AcO OAll OH HO HO AcO OAc 88% OH OH 50% 37 38 (LG)

The same year, Ogawa and co-workers44 reported the synthesis of both the heptasaccharide XXXG

(48) and the branched nonasaccharide XXFG (57) using a convergent approach as shown in Scheme

4. To reach the two targets, they used the key glycohexaosyl acceptor 46 (XXGG) as a common precursor. This resulted from a coupling between fragment 43 (XX) and cellobiose acceptor 44 (GG).

Glycosylation with xylopyranose donor 40 and hexose acceptor 46 afforded the heptasaccharide 47 in 64% yield as a 4:3 ratio of separable α:β stereoisomers. Finally, saponification and hydrogenolysis

23 afforded the desired unprotected XXXG target 48. In the same manner, coupling between trisaccharide donor 55 and 46 afforded the nonasacharide 56 as a 3:1 mixture of stereoisomers which, after separation and deprotection, gave the XXFG xyloglucan oligosaccharide 57. Even though the crucial glycosylation steps were plagued by low yields and poor stereoselectivity, both XyG targets could be reached by way of a common advanced intermediate using this convergent strategy.

Scheme 4. Synthesis of XXXG and XXFG oligomers by Ogawa and co-workers.44

BnO O BnO SMe OBn RO O O AcO O O 40 RO RO 1) Ac2O, py AcO AcO HO HO RO AcO RO 2) NH2NH2·AcOH, DMF AcO BnO O O RO O AcO O O O O BnO BnO OBn RO O AcO O BnO CuBr2, Bu4NBr, HgBr2 O O 3) CCl3CN, DBU, CH2Cl2 O O OBn CH NO RO OR AcO O CCl3 3 2 RO RO AcO AcO OR 72% OAc 62% NH 39 41: R = Bn 43 42: R = H AcO O PMBO BnO 43 AcO O AcO AcO HO O O BF3·OEt2 AcO BnO O O AcO BnO OBn O RO BnO 52% AcO O BnO OBn O O O AcO AcO O O AcO BnO O OBn OAc BnO BnO OBn 44 = CAN, CH3CN, H2O 45 R PMB 56% 46 R = H O , AcO O O 40, CuBr 2 Bu4NBr, HgBr2 AcO AcO AcO CH NO AcO AcO 3 2 AcO AcO O O AcO O O 64% AcO O BnO AcO O O AcO O O AcO BnO O OBn OAc BnO BnO OBn 47 α:β 4:3 - 25% α isolated

O HO HO O HO O HO HO HO NaOMe, MeOH HO HO HO 1) O O O HO O HO H O O 2) 2, Pd/C, MeOH HO O O O HO HO O OH OH HO OH 63% OH OH

48 (XXXG)

O Me O SMe O O BnO BnO OAll AcO OAll OBn BnO BnO OBn BnO O OBn AcO OH BnO OAll BnO OBn O OAc OSMe O BnO BnO AcO 50 53 7 steps BnO BnO OBn O O O AcO BnO AcO O NH O BF3·OEt2 MeOTf O O BnO 31% CCl RO Me O Me O 3 73% 81% OBn OAc OBn OAc NaOMe, MeOH 51: R = Ac BnO AcO 94% 52: R = H 49 54 55

Me AcO OAc AcO OAc 46 CuBr , Bu NBr, , 2 4 HgBr2 O OAc Et O 2 O O OAc 55 AcO OAc 26% AcO O O AcO AcO O O AcO AcO O AcO O O AcO O O BnO AcO O O AcO O O AcO BnO O OBn OAc BnO BnO OBn 56 α:β 3:1 - 20% α isolated

OH OH Me O OH HO HO O HO O HO HO HO O O HO OH 1) NaOMe, MeOH HO O HO HO O O O O 2) H2, Pd/C, MeOH, H2O O HO O HO HO O O HO O O O 54% HO HO OH OH HO HO OH 57 (XXFG)

More recently, Pfrengle and co-workers reported an automated glycan assembly of five XG oligosaccharides using monosaccharide 58 and disaccharide 59 as building blocks (Scheme 5).31 This solid phase synthesis also enabled the production of linear cellotriose and cellopentaose (see Table 4).

The authors used a Merrifield resin46 with a photocleavable linker47 and two phosphate glycosyl donors48 with either Fmoc or a Lev groups as temporary O-4 protection. Using this strategy, the five

XyG oligomers (60–64) were produced in overall yields of 2–16%.

Scheme 5. Automated glycan assembly of five XyG oligosaccharides by Pfrengle and co-workers.31

Merrifield resin

BnO O BnO O BnO OBn FmocO O OBu O BnO O O P LevO O OBz OBu O OBu O BnO HO N O 58 OBz P OBu × O 1 or 2 O 1a) 3.7 equiv. 58, 59 TMSOTf, CH2Cl2 O N or 2 × 2 or 1b) 3 3.7 equiv. 59, TMSOTf, CH2Cl2 Start

R2O R2O HO O R1O O BnO O BnO O OBn OBz n n

2 α 1 R = Bn, -xylopyranosyl R = Fmoc, Lev 2 α R = Bn, -xylopyranosyl

× Et N in DMF 2a) 3 20% 3 or N H 2b) 2 4·HOAc py, AcOH, H2O

Cleavage and global deprotection ν 3) h , CH2Cl2 4) NaOMe, MeOH, THF 5) Pd/C, H2, EtOAc, MeOH, H2O, HOAc HO O HO HO HO O HO O HO O O O HO HO O O NH OH HO 2 OH OH 3

60: 15% (GXG)

O HO O HO HO O HO O HO HO HO O HO HO HO HO HO HO HO HO HO O O HO O O HO O O HO O HO O O O HO O O HO O O HO HO O O HO O O O OH HO O NH OH HO O OH HO O 2 OH HO O NH2 OH OH HO OH 3 OH OH 3

61: 16% (GXXG) 62: 2% (GXXXG)

HO O HO O HO HO HO O HO O HO HO HO HO HO O O HO HO O HO HO O O O O HO O HO HO O O HO O O HO HO O O O HO O O O OH OH HO HO O O NH OH HO HO O O NH OH HO 2 OH HO 2 OH OH 3 OH OH 3

63: 10% (GXGXG) 64: 9% (XXGG)

4.2 β-(1→3)-D-Glucans

Figure 7. Structure of β-(1→3)-D-glucan (callose).

β-(1→3)-D-Glucans with various short β-(1→3)-D-glucan chains branched at the 6-position of the backbone are found in the cell walls of land plants, algae and also fungi.49 Callose is in this class of polysaccharide and has been isolated from all multicellular green algae as well as some higher plants

(embryophytes).50 This polymer plays an important role in maintaining cell wall integrity as well as in plant defense. In response to pathogen attack, callose is deposited between the plasma membrane and the pre-existing cell wall at the sites of attack.51 Another member of this class, laminarin, has been found in brown algae. As certain types of β-glucans containing (1→3) and (1→6) linkages exhibit biological activity by stimulating the innate immune system in man, their synthesis has attracted much interest.52 Several groups have successfully accomplished syntheses of linear as well as branched oligosaccharides mostly with the motivation of synthesizing fungal glycans. The synthesized fragments are summarized in Table 6.

Table 6. The β-(1→3)-D-glucan fragments synthesized to date. Structure R Year Reference

53 OR Me 1993 Takeo et al.

2011 Adamo et al.54

H 2005 Jamois et al.55

54 (CH2)3S(CH2)2NH2 2011 Adamo et al.

56 Cy-(C5H9) 2013 Jong et al.

Me 2011 Adamo et al.54 OR

All 2003 Zeng et al.57 OR

53 OR Me 1993 Takeo et al.

2 2003 Zeng et al.57

H 2005 Jamois et al.55

53 OR Me 1993 Takeo et al.

3 58 CH2CHOCH2 2005 Huang et al.

H 2005 Jamois et al.55

59 (CH2)3NH2 2016 Yashunsky et al. OR

53 OR Me 1993 Takeo et al.

4 H 2009 Mo et al.60

54 (CH2)3S(CH2)2NH2 2011 Adamo et al.

61 (CH2)5COOCH3 2015 Elsaidi et al.

62 (CH2)2NHCO(CH2)3COOSu 2015 Liao et al.

53 OR Me 1993 Takeo et al.

6 63 (CH2)6NH2 2010 Tanaka et al.

62 (CH2)2NHCO(CH2)3COOSu 2015 Liao et al.

n = 0, 2 64 (CH2)2NH2 2015 Liao et al. OR n

63 OR (CH2)6NH2 2010 Tanaka et al.

n = 4, 8, 12

65 OR (CH2)5NH2 2017 Weishaupt et al. n = 1, 4, 8

62 OR (CH2)2NHCO(CH2)3COOSu 2015 Liao et al.

63 OR (CH2)6NH2 2010 Tanaka et al.

10 66 (CH2)5NH2 2013 Weishaupt et al.

62 (CH2)2NHCO(CH2)3COOSu 2015 Liao et al.

63 OR (CH2)6NH2 2010 Tanaka et al.

63 OR Me 2010 Tanaka et al.

14 63 (CH2)6NH2 2010 Tanaka et al.

Several strategies have been developed to produce both linear and branched β-(1→3)-D-glucan oligosaccharides through solution phase synthesis. For oligosaccharides up to the pentasaccharide

30 level, an iterative approach has been used, which has enabled the synthesis of tri-, tetra- and pentasaccharides in the same sequence, as exemplified by the preparation of β-(1→3)-D-glucan oligosaccharides reported by Vetvicka and co-workers (Scheme 6).55,56 The group used an iterative deprotection-glycosylation approach with monosaccharide 65 to elongate the chain. This key monosaccharide could act successfully as both donor and acceptor to afford linear glucans 66–69. In common with most other reported syntheses of linear β-(1→3)-D-glucans, the glucoside building block 65 carried a benzoyl group at the 2-position to invoke neighboring group participation and a

4,6-benzylidene acetal as a permanent protecting group. The short length oligosaccharides were easily deprotected using Zemplén conditions and catalytic hydrogenolysis to afford the corresponding reducing oligosaccharides.

Scheme 6. Synthesis of protected di, tri, tetra- and penta-β-D-(1→3)-glucans by Vetvicka and co- workers.55

1) NIS, BnOH, TESOTf Ph Ph O Ph O O O 89% O O O O O OBn NAPO SEt NAPO O , OBz OBz 2) DDQ, CH2Cl2 MeOH OBz 85% 65 66 3) 65, NIS, TESOTf 83%

, 1) DDQ, CH2Cl2 MeOH Ph Ph O 86% Ph O O O O O O O OBn O O O NAPO OBz 2) 65, NIS, TESOTf OBz 78% OBz n

, : n = 1) DDQ, CH2Cl2 MeOH 67 1 2) 65, NIS, TESOTf 70% 68 : n = 2 , 1) DDQ, CH2Cl2 MeOH 2) 65, NIS, TESOTf 64% 69 : n = 3

For the synthesis of longer β-D-glucan oligosaccharides in solution, the use of convergent strategies employing di-53,59–62,64, tri-54,57–59,64, tetra-52,64 and pentasaccharide52,59 building blocks has been demonstrated to be advantageous, as it reduces the number of steps. This type of approach has been

31 used to synthesize long linear52,62 and branched52,59,64 oligosaccharides. In 2015, Guo and co-workers62 reported a convergent synthesis of tetra-, hexa-, octa-, deca- and dodeca-β-(1→3)-D-glucans, which they achieved through pre-activation-based iterative glycosylation with p-tolylthioglycosides as donors and disaccharide 70 as the key building block. The latter was used as a donor to elongate the chain from disaccharide to tetra-, hexa-, octa-, deca- and finally dodecasaccharides.

Scheme 7. Convergent linear synthesis of tetra-, hexa-, octa-, deca-, and dodecasaccharides by Guo and co-workers.62

Ph Ph O O O O O O STol NAPO O OBz OBz 70

1) 2-azidoethanol, AgOTf, TTBP , ο p-TolSCl, CH2Cl2 - 78 C , 2) DDQ, CH2Cl2 H2O

91% 1) 70, AgOTf, TTBP, p-TolSCl, , ο CH2Cl2 - 78 C Ph O Ph Ph O Ph Ph O O O O O O O O O O O O O O O O HO O HO O OBz 2) DDQ BzO OBz OBz , BzO 2 CH2Cl2 H2O N3 N3 71 72 90%

70 Ph Ph O HO HO Ph O O O O HO O HO O O O O O O HO O O O HO O O HO O HO BzO OBz OH OH BzO n OH n N H2N 3 73a-d n = 4, 6, 8, 10 74a-d n = 4, 6, 8, 10

Using another highly convergent and efficient synthesis, the same group prepared two branched oligosaccharides consisting of a nonaglucan backbone carrying a β-(1→3)-D-glucodiose or -tetraose at the 6’’’’-position.64 The backbone was assembled in a [4 + 1 + 4] fashion, in which each tetrasaccharide building block was generated using an iterative one-pot glycosylation as shown in

Scheme 8.

Scheme 8. Retrosynthesis of branched glucans by Guo and co-workers.64

RO HO HO O HO HO O O O HO O NH2 O OH O OH 4 H OH 4 75a-b HO HO HO O 1) glycosylation of the branch R HO O = O HO O 2) Zemplén deprotection and hydrogenolysis R HO O HO OH OH n = 2 OH 0,

OH Ph O O O Ph O O O O BnO O N O O BzO 3 O OBz 4 Bz BzO 4 76

1) 2 glycosylations 2) selective removal of Lev

Ph OLev Ph O Ph O O Ph Ph O O Ph O O O O O O O O O STol + O O O O O O O BnO O + O BzO O STol HO OBz BzO OBz HO BzO BzO 2 BzO 2 OBz N3 77 78 79 Preactivation-based iterative Preactivation-based iterative one-pot glycosylation one-pot glycosylation

Ph O Ph O Ph O Ph O O O O O O O O O BzO STol HO STol NAPO STol HO STol OBz OBz OBz OBz 80 81 82 81

Takahashi and co-workers achieved the impressive syntheses of linear hexadeca- and branched heptadeca-saccharides by conducting a convergent synthesis employing a tetrasaccharide as the key building block.52 In this case, 2,3-diol glycosides with a 4,6-O-benzylidene protecting group were effective as glycosyl acceptors, made possible by the higher reactivity of the 3-position over the 2- position. This afforded the desired (1→3)-linked regioisomer in each glycosylation reaction.

As shown by these examples, the use of pre-assembled building blocks is efficient but only gives access to glucans of limited chain lengths. In 2013, the group of Seeberger developed an iterative automated solid-phase synthesis of linear β-D-glucan oligosaccharides giving great flexibility in terms of the length of the targeted oligosaccharides. They showed the efficiency of the approach by synthesizing a linear β-(1→3)-D-glucan dodecasaccharide, as shown in Scheme 9.66 Glycosyl

33 phosphate building block 83 in combination with the resin-linker construct 84 (see also Scheme 5) proved to be a suitable choice for the automated synthesis of linear glucans. The efficiency of this glycosylation sequence was proven by the average glycosylation yield over 12 cycles: 89%. In 2017, the group further optimized the conditions to synthesize a series of branched β-(1→3)-D-glucans.65

Scheme 9. Automated solid-phase synthesis of β-glucan dodecasaccharide 87 by Seeberger and co- workers.66

Merrifield resin O2N

BnO O BnO O FmocO O OBu HO NCBz OPiv P OBu O 83 84 × 3 3 of 83 × equiv. 12 TMSOTf, CH Cl , dioxane × 2 2 3 piperidine, DMF

O2N

BnO O BnO BnO O BnO BnO O O NCBz BnO O O O OPiv HO OPiv OPiv 10 85

ν h , CH2Cl2

BnO BnO BnO O BnO BnO O O NHCBz BnO O O O OPiv HO OPiv OPiv 10 86

NaOMe, MeOH, CH Cl 1) , , 2 2 2) Pd(OH)2 H2 THF, H2O, AcOH

HO HO HO O HO HO O O NH2 HO O O O OH HO OH OH 10 87

4.3 Mixed-linkage [β-(1→3), β-(1→4)] glucans

n = 0,1,2

Figure 8. Structure of MLGs.

Mixed-linkage glucans (MLGs) are usually composed of short stretches of β-(1→4)-glucans connected through β-(1→3)-linkages. They are a type of hemicellulose characteristically found in both the primary and secondary cell walls of grasses. These glucans have also been isolated from bryophytes67 and from the horsetail plant (Equisetum).68 More recently, in 2000, Lahaye and co- workers discovered MLG 6-sulfate in the cell wall of the red alga Kappaphycus alvarezii. No chemical synthesis of MLG oligosaccharides had been reported before the automated synthesis achieved by

Pfrengle and co-workers in 2017.69

Table 7. The synthetic MLG fragments prepared by Pfrengle and co-workers. Structure R Reference

69 OR (CH2)5NH2 Dallabernardina et al.

69 OR (CH2)5NH2 Dallabernardina et al. 2

69 OR (CH2)5NH2 Dallabernardina et al.

69 OR (CH2)5NH2 Dallabernardina et al. 2

69 OR (CH2)5NH2 Dallabernardina et al.

This group performed the automated synthesis of seven MLG oligosaccharides using glucose building blocks 58 and 88 as shown in Scheme 10. After optimizing the conditions for trisaccharide 89, they concluded that two glycosylation cycles were required for the formation of β-(1→3)-linkages while one coupling sufficed to form the β-(1→4)-linkages. After completion of the solid-phase synthesis, the oligomers were cleaved from the resin in a continuous flow reactor, and deprotected under standard conditions to afford the targets (89–95).

Scheme 10. Automated synthesis of seven MLG oligosaccharides by Pfrengle and co-workers.69

HO HO O HO HO O HO O HO O O HO O NH2 OH OH OH 3 89: 13%

HO HO HO HO O O HO BnO BnO HO O O HO O HO O O FmocO O BnO O OH HO O NH2 O OBu O OBu OH OH BnO FmocO OH 3 OBz P OBu OBz P OBu 90: 26% O O 58 88 HO HO O HO HO O HO O HO HO O O O HO HO × HO O O HO O 1a) 2 3.7 equiv. of 58 OH OH HO O OH HO O O NH2 TMSOTf, CH Cl OH 2 2 OH 3 or 91: 12% OH × 1 1b) 3.7 equiv. of 88 TMSOTf, CH Cl HO HO 2 2 O HO HO HO HO HO O HO O HO HO O O O O HO O O O O OH OH HO HO O NH2 Start OH OH OH Cleavage and OH 3 84 92: 12% global deprotection × ν 3 20% Et N 3) h , CH2Cl2 2) 3 HO DMF 4) NaOMe O HO HO O HO MeOH, THF HO O O HO HO HO O O HO Pd/C, H EtOAc OH HO O O HO O O N 5) 2, OH HO O O 2 OH O NH2 MeOH, H2O, AcOH OH HO OH OH 93: 34% 3 O HO NCBz HO O HO HO O HO HO HO O O HO HO HO O O HO O OH O O O HO OH HO HO O O NH2 84 = Merrifield resin OH OH OH OH 3 94: 18%

HO HO HO HO O HO O HO HO O O HO O O HO HO O O O O HO HO HO HO O O HO O OH OH OH HO O OH OH HO O O NH2 OH OH OH 3 95: 23%

4.4 Xylans (arabinoxylans, glucuronoxylans and glucurono-

arabinoxylans)

Arabinoxylan (AX)

MeO MeO

Glucuronoxylan (GX)

MeO

Glucuronoarabinoxylan (GAX)

Figure 9. Schematic representation of different common heteroxylans.

Xylan, the most prevalent non-cellulosic polysaccharide in plant cell walls, binds to cellulose microfibrils and to lignin in order to strengthen the cell wall.4 Recently, Dupree and co-workers used

13C solid-state magic-angle spinning (MAS) nuclear magnetic resonance (NMR) spectroscopy to show that xylans, which form a threefold helical screw in solution, flatten into a twofold helical screw ribbon to bind intimately to cellulose microfibrils in the plant cell wall.70 Xylans have a backbone of

β-(1→4) linked D-xylose and they are known to have highly complex and heterogeneous structures due to the number, distribution and differentiation of side chains. Xylans are usually classified into homoxylans and heteroxylans, where the latter includes arabinoxylans (AX), glucuronoxylans (GX) and glucuronoarabinoxylans (GAX) as shown in Figure 9. Arabinoxylans present L-arabinofuranosyl substituents α-(1→2) and/or α-(1→3) linked to the xylose residues of the backbone. Additionally, it

38 has been found that in grasses, the O-5 position of arabinose can be esterified with for example ferulic acid that can form covalent bonds to lignin.71,72 In glucuronoxylans, 4-O-methyl-D-glucuronic acid and/or D-glucuronic acid residues are generally α-(1→2)-linked to xylose substituents, with an average substitution ratio of 1:10.73 Finally, GAX has 4-O-methyl-D-glucuronic acid in the O-2- postitions and arabinofuranosides in the O-3-position with a ratio of xylan:substituents of approximately 10:1.4.73 In addition to these substituents, most xylans are acetylated to a varying degree. Esterification is mainly found in the O-3 position, and more rarely in the O-2 position. An overview of the xylan fragments that have been synthesized is presented in Table 8.

Table 8. Synthetic xylan fragments. Structure R Year Reference

74 OR H 1982 Hirsch et al.

75 OR Me 1981 Kováč et al. 2 H 1982 Hirsch et al.74

Bn 1995 Takeo et al.76

77 (CH2)5NH2 2015 Schmidt et al.

74 OR H 1982 Hirsch et al. 3 Bn 1995 Takeo et al.76

77 OR (CH2)5NH2 2015 Schmidt et al. 4 Bn 1995 Takeo et al.76

76 OR Bn 1995 Takeo et al. 5

77 OR (CH2)5NH2 2015 Schmidt et al. 6 Bn 1995 Takeo et al.76

76 OR Bn 1995 Takeo et al. 7

76 OR Bn 1995 Takeo et al. 8

78 OR H 2001 Oscarson et al.

MeO

79 OR Me 1978 Kováč et al.

77 OR (CH2)5NH2 2015 Schmidt et al.

80 OR (CH2)5NH2 2017 Senf et al.

79 OR Me 1980 Kováč et al.

MeO

77 OR (CH2)5NH2 2015 Schmidt et al.

80 OR (CH2)5NH2 2017 Senf et al.

77 OR (CH2)5NH2 2015 Schmidt et al.

80 OR (CH2)5NH2 2017 Senf et al.

77 OR (CH2)5NH2 2015 Schmidt et al.

80 OR (CH2)5NH2 2017 Senf et al.

77 OR (CH2)5NH2 2015 Schmidt et al.

The chemical synthesis of well-defined homoxylans has been reported over recent decades by several groups using a range of synthetic strategies.74–77 Fewer have described the synthesis of heteroxylans,

77–79 probably due to the difficulty in producing differentially-protected xylose building blocks needed to introduce side chains. Xylobiose was first synthesized by Myrhe and Smith in 1961, and the reaction was carried out via Helferich coupling81 between benzyl 2,3-di-O-benzyl-D-xylopyranoside and a

82 peracetylated xylosyl bromide in the presence of Hg(CN)2. Twenty years later, Hirsch and Kováč synthesized a series of oligoxylans in the same fashion by using a sequential approach to elongate the

41 chain and reach up to a pentaxylan as shown in Scheme 11.74,75 1,2,3-Tri-O-acetyl-4-O-benzyl-β-D- xylopyranose 96 was used as common building block to form both the bromide donor 97 and acceptor

98. Mercury cyanide-mediated coupling of the two afforded the disaccharide product as a 1:1.5 mixture of α:β stereoisomers, which could be separated by column chromatography to afford 23% of the desired β-(1→4)-linked disaccharide 99. Subsequent hydrogenolysis afforded the 4’-OH disaccharide acceptor 100 ready for coupling with donor 97. Repetition of the two last steps enabled the researchers to reach tri-, tetra- and pentaxylans (103a-c) after saponification. Using this procedure, the yields of the glycosylation reactions were all quite low (< 25%). The same iterative strategy was subsequently employed to obtain a series of methyl β-xylosides up to a xylohexaoside.83

Scheme 11. Hirsch and Kováč strategy for the synthesis of oligohomoxylans.75

HBr BnO O toluene AcO AcO CH2Cl2 Br 97

O BnO O Hg(CN)2 RO O O AcO OAc AcO AcO OAc OAc MeCN AcO OAc

96 Pd/C 99: R = Bn Pd/C 23% O MeOH, acetone : R = H MeOH HO 100 AcO OAc acetone OAc 98

1) 97, Hg(CN)2 2) Pd/C, MeOH acetone RO O O O O O AcO AcO OAc AcO AcO AcO OAc n = 1,2,3 Pd/C R = Bn 101a-c MeOH, acetone R = H 102a-c

NaOMe HO O O O O O MeOH HO HO OH HO HO HO OH

103a-c n = 1,2,3

More recently, Takeo and co-workers developed a blockwise approach to a series of linear xylooligosaccharides up to xylodecaose using suitably protected thio-xylobiosides as building blocks and

N-iodosuccinimide (NIS) in combination with silver triflate as promoting agent. This strategy afforded glycosylation yields of around 80%.76

The first synthesis of heteroxylan oligosaccharides was reported by Hirsch and Kováč in 1982. 75,79

The fragment was a xylotriose backbone bearing a β-(1→3)-linked xylose unit and an α-(1→2)-linked glucuronic acid. The researchers started a stepwise synthesis from methyl 2,3-anhydro-β-D- ribopyranoside 104 as shown in Scheme 12. This epoxide was coupled with a xylosyl bromide donor

105 and the resulting disaccharide 106 was then selectively deprotected at positions O-3 and O-4 and further glycosylated with xylosyl bromide 108 to afford the tetrasaccharide 109 in 40% yield. The epoxide was then opened selectively with benzyl alcohol following the Fürst-Plattner rule84, in order to produce the xylose tetrasaccharide 112. The free hydroxyl group in position 2 was then glycosylated with glucoronate donor 113 in the presence of silver perchlorate and 2,4,6-collidine affording the desired α-branched pentasaccharide in 57% yield.

Scheme 12. Synthesis of a branched 4-O-methyl-α-D-glucuronic acid-containing xylotetraose by Hirsch and Kováč.75,79

AcO O AcO O AcO AcO BnO AcO Br Br 105 108 O BnBr OMe O O OMe RO O OMe HO O Hg(CN)2 RO O Hg(CN)2 RO O O O NaH RO RO O BnO O BnO O MeCN O MeCN RO O DMF RO 104 NaOMe, MeOH 106: R = Ac RO NaOMe, MeOH 109: R = Ac 62% 40% 83% 98% 107: R = H 98% 110: R = H MeOOC O MeO MeOOC BnO Cl MeO O BnO HO 113 HO O O BnO BnO O 1) AgClO4, Et3N HO O OH O O O OMe O O O , O O HO BnO O BnONa BnO O CH Cl 57% HO O BnO O BnO O OMe 2 2 OH O O BnO BnO BnO BnO O O BnO O OH , HO O OH BnO BnOH BnO 2) H2 Pd/C, H2O HO BnO 111 BnO 112 acetone OH 114

In 2001, Oscarson and Svahnberg reported the synthesis of two glucuronoxylan fragments based on xylobiose. Using dimethyl(methylthio)sulfonium trifluoromethane sulfonate (DMTST) in diethyl ether, they were able to couple the glucuronic acid donor 116 with complete α-selectivity to the disaccharide acceptor 115 (Scheme 13).78

Scheme 13. Synthesis of a glucuronoxylan trisaccharide by Oscarson and Svahnberg.78

O MeO O MeO O MeO SEt O OMe BnO MeO OH OBn BnO O O O 116 OMe BnO O O BzO OPMB O OBz O O OMe DMTST O O OPMB Et O O BzO 2 OBz 115 OMe 117 O 89% HO MeO O deprotection HO HO O 32% HO O O HO O HO OH OH 118

Recently, Pfrengle and co-workers developed an automated synthesis of arabinoxylans.77,80 They first reported the synthesis of a set of xylan oligosaccharides up to an octasaccharide decorated in different positions with either one or two α-(1→3)-linked arabinofuranosides as shown in Scheme 14.65

Scheme 14. Automated synthesis of arabinoxylan fragments by Pfrengle and co-workers.76

SEt SPh FmocO O FmocO O O O BnO O OBu NapO O OBu OBz OBz OBz P OBu OBz P OBu O O BzO BzO OBz OFmoc 119 120 121 122

Automated solid-phase synthesis

Coupling: TMSOTf, CH2Cl2 (119 , 120) , dioxane NIS, TfOH, CH2Cl2 (121, 122) Et N in DMF Deprotection: 20% 3 (Fmoc) DDQ, ClCH2CH2Cl (NAP) Ac Capping: 2 O, py hν, Cleavage: CH2Cl2 MeOH Deprotection: NaOMe, CH 2Cl2, Pd/C, H2, EtOAc, MeOH, H2O, AcOH

HO O O O O O HO HO O NH2 O OH HO HO O O O HO OH 3 HO O O NH O HO HO 2 6 HO O O O OH HO HO O NH OH 3 125 : 43% HO HO O 2 4 OH HO OH 3 : 2 124 20%

123 : 27% O HO O O O O HO O HO O O O O NH2 HO O O OH HO O O NH2 O OH OH 3 OH HO O OH OH 3 OH OH HO O O HO 128 : 42% O O O O HO 126 : 23% O HO O NH2 OH O OH HO O OH OH 3 OH OH

HO : HO OH 127 19% OH O HO O O O HO HO O O O O OH O O O O OH HO HO O NH2 O OH OH HO OH OH OH 3 O HO 129 : 8% HO O O O OH HO O O O O O HO O O NH OHO OH O HO 2 OHO OH O OH OH 3 HO O O O OH HO O O O HO O O O HO OH O O NH2 OH HO 130 : 15% OHO OH HO O OH OH 3 OH OH OH

HO HO 131 : 21% OH OH HO O O O O O O HO O O O O HO HO O O OH O OH O O NH2 OH OH HO OH O OH OH 3 OH HO OH HO 132 : 7% OH

To achieve their synthesis, two xylosides 119 and 120, differentially protected at their O-3 position, were used as building blocks for the assembly of the linear xylan backbone, while perbenzoylated and

2-O-Fmoc-L-arabinofuranosides 121 and 122 were used for substitution. The sequential synthesis provided, within short timeframes and with overall yields of 7-43%, a collection of arabinoxylan

45 fragments (123–132) as shown in Scheme 14. Very recently, they optimized this strategy in order to supplement this set of α-(1→3)-substituted glycans with α-(1→2)-substituted arabinoxylan oligosaccharides 135 to 139 (see Figure 10). By using the 2-O-(azidomethyl)benzoyl (Azmb)- protected donor 133 the team could make oligosaccharides containing monosubsituted xylose residues. With the inclusion of the orthogonally protected donor 134, synthesis of the disubstituted arabinoxylan 135 was possible.

FmocO O FmocO O BnO O OBu NapO O OBu AzmbO P OBu AzmbO P OBu O O 133 134

HO O O O HO O O O HO O O O O O NH2 HO O O OH HO HO HO O O O O O O OH 3 OH HO O NH OH O HO HO O 2 OH OH OH O O OH 3 HO OH OH OH 135 : 4% 136 : 7% HO HO OH OH

HO O O O O O HO O O O HO HO O O NH HO O O HO 2 HO HO O O O O OH O O HO O NH OH 3 OH OH HO HO O 2 O O OH OH OH OH OH 3

HO HO OH OH 137 : 13% 138 : 4%

O O HO O HO O O HO O O O HO O O O HO O NH2 HO HO O O O O O HO OH HO O NH OH OH OH O HO HO O 2 O 3 OH OH OH O OH 3 OH HO HO OH 139 : 41% HO 140 : 11%

Figure 10. A set of substituted (arabino)xylan oligosaccharides synthesized in 2017 by Pfrengle and co-workers.80 The strategy is presented in Scheme 14.

4.5 Mannans

Galactomannans

Galactoglucomannans Figure 11. Schematic representation of different mannans.

Mannans are widely distributed and the main hemicellulose in Charophytes. 67 They have been intensively studied due to their role as seed storage compounds, but they are found in variable amounts in all cell walls.4 The mannan class of polysaccharides in higher plants can be divided into two categories: galactomannans and glucomannans. Galactomannans have a homologous backbone made up of β-(1→4)-linked mannose residues only, and can be substituted at some of the O-6 positions by

β-D-galactose units. The glucomannan backbone is composed of β-(1→4)-linked glucose and mannose units assembled in a random pattern with a higher number of mannose than glucose monosaccharides. Galactoglucomannans are further substituted with galactose at the O-6 positions of some backbone mannose residues. Furthermore, mannans are often acetylated at C-2, C-3 and more rarely at C-6 positions depending on the plant species.85 Mannans and glucomannans have been highly conserved throughout plant evolution, however their role and function in the plant cell wall remain largely unclear. The exception is the galactoglucomannans which are often described as seed storage compounds.4 The formation of β-linked mannosides is an exceedingly difficult glycosylation reaction since neighboring group participation, the steric hindrance from a protecting group at the O-2 position and the anomeric effect uniformly favor the formation of α-D-mannopyranosides. Nevertheless,

47 several groups have developed strategies to circumvent the challenges and have successfully synthesized certain mannan fragments, as shown in Table 9. Numerous reviews dealing with advances in the construction of β-D-mannosides have been published. Among those, a review by Gridley and co-workers86 deals with most of the strategies. A more recent one, by Lichtenthaler, presents the use of 2-oxoglycosyl donors in assembling oligosaccharides with β-D-mannose, β-L-rhamnose, N-acetyl-

β-D-mannosamine, and N-acetyl-β-D-mannosaminuronic acid residues.87

Table 9. Synthetic mannan fragments. Structure R Year Reference

88 OR Bn 2003 Twaddle et al.

89 OR Me 2008 Kim et al. 2

90 OR Me 2004 Crich et al. 4 H 2013 Ohara et al.91

Pr 2015 Tang et al.92

91 OR H 2013 Ohara et al.

91 OR H 2013 Ohara et al. 5

91 OR H 2013 Ohara et al. 2

90 OR C6H11 2004 Crich et al. 6 H 2013 Ohara et al.91

91 OR H 2013 Ohara et al. 2 2

93 OR Me 2012 Ekholm et al.

Me 2012 Ekholm et al. 93

In 2004, Crich and co-workers developed a direct approach for the synthesis of β-mannosides using

4,6-O-benzylidene-directed glycosylation via in situ generation of α-mannosyl triflates from their thio- or sulfoxidoglycoside precursors. They employed this strategy in combination with a three step iterative glycosylation protocol to synthesize a β-linked hexamannoside as shown in Scheme 15.90

Benzylidene deprotection, followed by selective Piv protection of the primary alcohol in each case provided the acceptor ready for the glycosylation with 4,6-O-benzylidene-thiomannoside 141. High glycosylation yields were obtained in each case (71–82%) and very good β-selectivity was achieved

(α:β ~1:10).

Scheme 15. Three-step iterative glycosylation protocol to synthesize hexamannoside 146d by Crich and co-workers.90

1) neopentyl glycol, BSP, Tf O CSA,CH Cl 141, BSP Ph O OBn 2 2 2 OPiv OPiv OBn Tf O, TTBP O O TTBP Ph O 89% OBn 2 Ph O OBn OBn O O O BnO OMe HO O O O O BnO OMe CH Cl BnO OMe SPh MeOH 2) PivCl, DMAP, BnO 2 2 BnO 141 142 Et3N, CH2Cl2 143 144 82% 98% 80%

OPiv OPiv OBn Ph O OBn OBn O O O O O O O BnO BnO OMe N 3-step BnO S protocol: n N A) removal of benzylidene N 145a-d n = 1, 2, 3, 4 B) Piv protection with donor 141 C) glycosylation 2-benzenesulfonyl 2,4,6-tri-tert-butyl piperidine pyrimidine (BSP) (TTBP)

1) NaOMe, MeOH OPiv OPiv OH OH OH Ph O OBn OBn 94% OH OH O OBn O OH O O O O HO O O BnO O , O O BnO BnO OMe 2) H2 Pd/C, MeOH HO HO HO OMe 4 87% 4 145d 146d

In 2008, Jeon and co-workers reported an efficient one-pot, three-step glycosylation strategy starting from a reducing sugar. A 4,6-O-benzylidene mannopyranose was treated with phthalic anhydride in the presence of DBU, and the α-mannosyl phthalate intermediate was further activated with triflic anhydride to generate a mannosyl triflate. Like Crich and co-workers, the group used 4,6-O- benzylidene protected mannose to impart stereoselectivity and form the desired β-mannosidic linkage.89

Another approach was taken by Pohl and co-workers in an automated solution phase synthesis of a β-

(1→4)-linked hexamannoside, shown in Scheme 16.92 Inspired by the work of van der Marel and co- workers in their formation of β-mannosidic linkages using a mannuronate donor (section 7.5), Pohl and co-workers used a β-directing C-5 carboxylate strategy and synthesized the protected β-(1→4)- hexamannoside 149 in an average yield of 75% per step. Post-assembly, the mannuronic acid residues were reduced to mannans with DIBAL-H.

Scheme 16. Automated solution phase synthesis of β-hexamannoside by Pohl and co-workers.92

C8F17 MeO2C OBn TBSO O HO O BnO O CCl3 147 148 NH

Start Coupling 1) TMSOTf, 148 148, TMSOTf CH2Cl2 ⋅ CH2Cl2 TBAF, Et N HF 2) 3 3 DMSO, THF

C8F17 C8F17 R MeO2C OBn R MeO2C OBn O O O O BnO O O BnO O O n n R = H R = TBS

Deprotection · Reduction TBAF, Et3N 3HF DMSO, THF , DIBAL-H, CH2Cl2 toluene

C F Grubbs' cat. 2nd gen. HO OBn 8 17 1) HO OH H H O O ethylene, CH2Cl2 O O BnO O O HO OPr , 6 2) H2 Pd/C, MeOH 6 149 150

73%

Indirect approaches requiring epimerization at C-2 in readily available β-glucosides have been used by other groups to synthesize linear β-(1→4)-linked mannans.88,91 Nikolaev and co-workers accessed

β-(1→4)-mannotriose from commercially available and cheap D-cellobiose octa-acetate 151 via an

88 epimerizing SN2 reaction as shown in Scheme 17. After some protecting group manipulations required to reach benzyl 4’,6’-O-benzylidene-cellobioside diol 153, simultaneous epimerization of C-

2 and C-2’ was performed via triflation followed by reaction with Bu4NOBz, affording β-(1→4)- linked mannobiose 154 in 76% yield. Mannobiose 156 was further coupled to glucose donor 157, forming a trisaccharide, which was then epimerized to give mannotriose 160 in a reasonable overall yield.

Scheme 17. Synthesis of β-mannotriose through C-2 epimerization of glucoside by Nikolaev and co- workers.88

1) PhCH(OMe)· 2, HBr, AcOH, TsOH H O, OAc 1) OH 2 CH2Cl2 DMF OAcOAc OH O OH O AcO 2) BnOH, Hg(OAc)2 O HO 68% Ph O BzO AcO HO O OBn O O OBn AcO O O HO O BzO O OAc 3) NaOMe, MeOH OH 2) BzCN, Et3N OH OAc OH OBz 151 152 33% 153 70%

OH OBz 1) Tf2O, py BzCN, Et3N, O OBz AcOH OBz OBz CH2Cl2 Ph O BzO O BzO MeCN, CH2Cl2 O BzO O O OBn HO O OBn HO O OBn BzO O BzO O BzO O 76% 2) Bu4NOBz, OBz OBz 92% OBz OBz OBz OBz toluene 154 155 156

76% OAc

AcO O AcO AcO O NH 157 OAc OBz 1) HCl, OBz CCl3 OBz MeOH, 61% O OBz O Ph O BzO AcO O BzO O O O O O OBn BzO O OBn TMSOTf, CH2Cl2 AcO O BzO O AcO BzO 2) PhCH(OMe)· 2, OH 47% OBz TsOH H O, OBz OBz 2 OBz 158 MeCN 159 3) BzCN, Et3N MeCN 65% (two steps)

1) Tf py OBz OH OH 2O, OBz OH CH Cl O OBz TFA, H CH Cl OH 2 2 Ph O BzO 1) 2O, 2 2 O HO O O O HO O O BzO O OBn HO O OBn Bu NOBz, BzO O NaOMe, MeOH, THF HO O 2) 4 OBz 2) OH toluene 160 OBz 161 OH 70% 88%

A combination of direct and indirect strategies was used by Wong and co-workers in their synthesis of β-(1→4)-linked hexa- to octamannans, where some of the targets were further acetylated at one or two positions along the backbone.91 An example of the convergent synthetic strategy taken to the three hexamannan targets 162, 163, 164 is presented in Scheme 18. The three oligosaccharides can be derived from a common protected hexamannan intermediate 165. As direct β-selective mannosylation remains challenging, the group chose to synthesize the hexamannoside 165 by an indirect method involving β-selective glucosylation of trimannoside acceptor 167 with mannoglucose donor 166. The subsequent epimerization of the glucose C-2 hydroxyl group was achieved via an oxidation-reduction sequence affording the desired β-mannoside. This methodology simultaneously allowed for

52 acetylation at the designated positions. The trisaccharide fragments 166 and 167 could be obtained from monosaccharides 168 and 171 respectively by repeating direct β-selective mannosylations using

169 or 170 as the donors.

Scheme 18. Retrosynthetic analysis of hexamannans 162–164 by Wong and co-workers.91

OH OH OH OH OR1 OH OH HO O HO O O HO O O HO HO O O HO O O HO O O OH OR2 OH OH OH OH

162: R1 = R2 = H

163: R1 = Ac, R2 = H

164: R1 = R2 = Ac

OBn OBn OBn OBn OH Ph O OBn O O BnO O O BnO O O BnO BnO O O BnO O O BnO O O OBn OPMB OBn OBn OBn OBn 165

OBn OBn OBn OBn Ph O OBn O O BnO O O O + BnO O O BnO BnO O BnO HO O OBn BzO O O BnO O OPMB OBn OBn OBn OBn 166 Cl3C NH 167

OBn OBn OR Ph O O O BnO HO O BnO HO O BnO OAll OBn SPh BzO OBn 168 169: R = Bn 171 170: R = PMB

The authors used the same strategy, with a late β-glucosylation in [4 + 3] and [4 + 4] couplings followed by epimerization to get the hepta- and octamannans shown in Figure 12.

OH OH OH OH OR1 OH OH HO O O HO O O HO O O HO HO O HO O O HO O O HO O O OH OH OR2 OH OH OH OH OH

172: R1 = R2 = H

173: R1 = Ac, R2 = H

174: R1 = R2 = Ac

OH OH OH OH OH OR OH OH HO O O HO O O HO O O HO O O HO O HO O O HO O O HO O O HO OH OH OH OH OH OH OH OH OH

175: R = H 176: R = Ac Figure 12. Hepta- and octamannans synthesized by Wong and co-workers.91

To date, only one synthesis of a galactoglucomannan fragment has been reported. The tetrasaccharide fragment 185 was synthesized by Leino and co-workers in 2012, using a linear synthesis involving a

β-selective mannosylation with 4,6-O-benzylidene thiomannosyl donor 177, Scheme 19.93 A β- selective glucosylation afforded trisaccharide 181, which was further selectively deprotected at the C-

6’’ position. Finally, and following Ando and co-workers’ discovery,94 α-selective galactosylation with di-tert-butylsilylene galactoside donor 183 afforded tetrasaccharide 184 ready for deprotection to reach the target 185.

Scheme 19. Synthesis of galactoglucomannan tetrasaccharide by Leino and co-workers.93

BnO BnO OBn HO O O AcO SPh HO OAc BnO 178 180 OMe BSP, TTBP, Tf O NIS, TMSOTf, OBn BnO 2 OBn BnO Ph O BnO OBn OBn CH Cl Ph O CH Cl O O O Ph O 2 2 O O O 2 2 O O O O O O BnO AcO SPh BnO AcO SPh BnO BnO 53% 77% AcO OAc OMe 177 179 181

Si O O O Si O BzO SPh O OBz 183 O , · BzO Cu(OTf)2 BH3 THF NIS, TMSOTf, BzO CH Cl HO OBn BnO CH Cl O OBn BnO 2 2 O O BnO OBn 2 2 O BnO OBn BnO O O BnO O O O BnO O BnO O 85% AcO BnO 68% AcO BnO AcO AcO OMe OMe 182 184

OH HO O 1) HF·py, THF, 68% HO 2) NaOMe, MeOH, THF, 96% HO O OH HO O HO OH , HO O O O 3) H2 Pd/C, MeOH 96% HO O HO HO HO 185 OMe

Pectin

Pectin was first discovered in 1790 by Vauquelin,95 being first described as ‘pectin’ in 1825 by

Braconnot.96 The name derives from the Greek word pektikos, meaning ‘congeal, solidify or curdle’, and accurately describes the main characteristic of this group of polysaccharides. As a result of their ability to form gels, pectic polysaccharides have many industrial applications. They are used as gelling and stabilizing ingredient in a wide range of food products. They also fulfil important biological roles, including imparting mechanical strength and flexibility to the cell wall. Pectin therefore has multiple functions in plant growth, morphology, development, and defense.97 Pectic polysaccharides also play an important role in cell-cell adhesion.6

Pectins make up around 35% of the dry weight of the primary cell wall in dicots and non-graminaceous monocots, 2–10% in grasses and other commelinoid primary walls, and up to 5% in woody tissue.98

Although pectin is structurally and functionally the most complex and heterogeneous class of polysaccharides in nature, it is relatively well-defined, being characterized by a high content of galacturonic acid (see Figure 13). It is widely believed that four to five pectic domains, namely, homogalacturonan (HG), rhamnogalacturonan I (RG-I), rhamnogalacturonan II (RG-II), and xylogalacturonan (XGA) and/or apiogalacturonan (AGA) are covalently linked to one another to form the pectin complex in muro. As yet, there is no sufficiently compelling evidence, as to how the different blocks are assembled in the macromolecule. Therefore, the fine-structure of pectin is still a matter of controversy and several models have been proposed. A review by Yapo in 2011 describes the two traditional models, one with alternating ‘smooth’ (HG) and ‘hairy’ (XGA, RG-I/RG-II) regions, the other proposing a backbone of RG-I, with side chains of HG, XGA and RG-II. The author argues that in the face of more recent analysis, the traditional models become less and less convincing

56 as comprehensive descriptions of pectin.99 Irrespective of which model of polysaccharide structure one subscribes to, it is undisputed that pectins consist of distinct domains. HG and RG-I represent the two major kinds of backbone in pectin. While HG is a linear homopolymer of partially methyl esterified and/or acetylated α-(1→4)-linked D-galacturonic acid, the RG-I backbone is made up of alternating galacturonic acids and L-rhammnose monosaccharides, with various L-arabinan, D- galactan and arabinogalactan side chains. Three more domains of pectin, xylogalacturonan (XGA), apiogalacturonan and rhamnogalacturonan II (RG-II) can be seen as structurally modified HG as they share the same backbone. While 25–75% of the galacturonic acid residues of XGA bear xylose substitution, RG-II is a complex branched HG having four different side chains, with a total of up to

29 different monosaccharides, including a number of rare sugars.100 More detail on the structure of the five pectic domains will be given further on in this section. Only few pectic oligosaccharides are available from the chemical or enzymatic degradation of polysaccharides. Although degradation techniques have been widely used for elucidating the structure of pectin (especially RG-I), the oligosaccharide fragments prepared in this way required extensive purification and were generally only obtained in analytical quantities. Furthermore, this approach has mostly been limited to isolation of linear backbone fragments since the techniques generally involve the removal of side chains.97 This fact has prompted chemists to devise synthetic approaches to pectic oligosaccharides, albeit to date only for a limited number of structures. A review by Nepogodiev and co-workers summarized the fragments of HG, RG-I and RG-II, which had been synthesized before 2011.101 These are included in the following discussion.

L-rhamnose L-iduronic acid OAc β-notation

COOH L-arabinose D-galactose α-notation

COOMe L-fucose 6 D-3,6-anhydrogalactose O = 4 1 Me 3 2 L-aceric acid SO3H D-glucose less common enantiomer (D/L)

D D-xylose -apiose

D-3-deoxy-D-lyxo- D-mannose heptulosaric acid (Dha)

L-gulose D-2-Keto-3-deoxy-D-manno- octulosonic acid (Kdo)

Rhamnogalacturonan I Homogalacturonan

MeO

OMe

Rhamnogalacturonan II Xylogalacturonan

Figure 13. Schematic representation of the four domains of pectin.

5.1 Homogalacturonan

Figure 14. Homogalacturonan (HG) structure.

Homogalacturonan (HG), which is referred to as the ‘smooth region’ of pectin constitutes approximately 65% of the macromolecule. HG consists of a linear chain of α-(1→4)-linked galacturonic acid, which can be methyl esterified or acetylated at the C-2 and C-3 positions depending on the plant as well as the type of tissue. HG has been shown to be present in stretches of approximately 100 GalpA residues.102 It is believed that HG is synthesized as a highly methyl esterified form in the Golgi apparatus and then de-esterified enzymatically, once it has been deposited in the cell wall.98,103,104 The degree of methylation is well known to influence gelling capability as well as other physical properties of pectin. C-6 carboxylates of HG residues can interact with calcium ions to form a stable gel when more than 10 consecutive un-esterified galacturonic acid residues are present.6 Furthermore, the methyl esterified GalpA residues of HG can associate through hydrophobic interactions.100

Efforts to synthesize different fragments of HG, in some cases with a distinct pattern of methyl esterification have provided a variety of oligomers ranging from tri- to dodecasaccharides, presented in Table 10.

Table 10. Overview of the synthesized homogalacturonan fragments. Structure R Year Reference

H 1990 Nakahara et al.105 OR

Pr 2011 Pourcelot et al.106 OR

H 1999 Kester et al.107 OR

2001 Clausen et al.108

H 1999 Kester et al.107 OR

2001 Clausen et al.108

H 1999 Kester et al.107 OR

2001 Clausen et al.108

H 2003 Clausen and Madsen.109 OR

Pr 1989 Nakahara et al.110 OR 8

H 1990 Nakahara et al.111 OR 10

As is typical for the synthesis of glycuronides, two distinct approaches have been employed to reach oligomers of α-(1→4)-linked D-GalpA: a post-glycosylation oxidation and a pre-glycosylation oxidation. A comprehensive microreview by van der Marel and co-workers explains the rational behind these two strategies.112 With post-glycosylation oxidation, often the preferred approach, the backbone of the target is assembled from aldose building blocks, followed by oxidation of the primary alcohols to carboxylic acids prior to or after global deprotection. The alternative pre-glycosylation oxidation strategy involves the glycosylation between protected uronic acid donors and acceptors.

Due to the electron-withdrawing carboxylic acid group, galacturonic acid building blocks are substantially less reactive than the corresponding galactose derivatives, which makes this second approach challenging. In fact, only two syntheses of trigalacturonides have been reported using the pre-glycosylation oxidation method, and no oligomer longer than this has been produced.

The first was performed by Anker and co-workers in 1997 using thiophenyl galacturonides as glycosyl donors, as shown in Scheme 20.113 Good stereoselectivity and a high yield was obtained for the first glycosylation affording disaccharide 188. However, attempts to elongate the chain to a trisaccharide by coupling the thiogalacturonide 190 with the disaccharide acceptor 189 only afforded the target molecule in 45% yield. The trisaccharide 191 was not deprotected. Similar results were obtained by

Kramer et al., also by glycosylating a disaccharide acceptor with a thiogalacturonide donor. Despite slight variations in the protecting group pattern, this group obtained the targeted trisaccharide in a similarly disappointing 48% yield.114

Scheme 20. Protected trigalacturonide synthesized through a pre-glycosylation oxidation strategy by Anker and co-workers.113

HO BnO CO2Bn CO Me O 2 BnO O CO2Me BnO OBn BnO SPh O OBn OBn BnO BnO 187 PMBO HO 190 O CO2Bn CO2Bn CO2Bn NIS, TfOH O DDQ O NIS, TfOH O PMBO BnO , BnO BnO CO2Bn CH Cl CH Cl H O CH2Cl2 O 2 2 BnO 2 2 2 BnO BnO BnO SPh O O O CO2Bn CO2Bn CO2Bn 78% O 86% O 45% O OBn α α:β > 95:5 BnO OBn BnO OBn :β = 95:5 BnO OBn OBn OBn OBn 186 188 189 191

These results encouraged other groups to turn their attention to the post-glycosylation oxidation approach. Galactopyranose-based building blocks used for the synthesis of the neutral α-(1→4)-linked galactan backbone require a temporary protecting group at the C-6 position, which can release the primary alcohol prior to oxidation at a late stage of the synthesis. This method includes early work done by Nakahara and Ogawa in the 1990s.105,110,111 Among their elegant syntheses, the pinnacle remains the total synthesis of dodecagalacturonic acid using a highly convergent approach as shown in Scheme 21.111

They chose galactosyl fluorides as donors, which afforded highly stereoselective glycosylations and gave good yields (62–95%). Acetyl groups were used as temporary protecting groups for all the primary alcohols, which were then selectively unmasked to afford the α-(1→4)-linked dodecagalactan

102. The alcohols were further oxidized to the corresponding carboxylic acids using a Swern115 followed by a Lindgren-Kraus-Pinnick oxidation116–118 affording the dodecagalacturonic acid product in 50% yield. Final TBDPS deprotection and hydrogenolysis gave the target (203) in a good yield overall.

Scheme 21. Synthesis of dodecagalacturonic acid by Nakahara and Ogawa.111

OAc HO BnO OAc O O BnO BnO BnO OAc BnO O O OAc OAc BnO O O BnO OTBDPS O BnO BnO BnO BnO OAc BnO O O OAc 193 O O BnO BnO , , SnCl2 AgClO4 Et2O BnO OAc BnO F O 192 92% O 194: R = OTBDPS BnO TBAF, AcOH, THF BnO R : = , 195 R OH (90%) Et2NSF3 MeOH 196: R = F (97%)

HO OAc O O O O O BnO O O BnO OAc BnO BnO O BnO BnO O OAc O OAc O O BnO O OTBDPS O O BnO BnO BnO O BnO 1) TBAF, AcOH, THF BnO BnO 193 O OAc , O OAc 2) Et2NSF3 MeOH BnO OAc O O O BnO BnO O , 93% SnCl2 AgClO4, Et2O BnO BnO BnO O OAc O OAc BnO F 77% O O 197 198 BnO OTBDPS 199 BnO BnO BnO F 1) 80% aq. AcOH 2) AcCl, py

86% HO OAc O BnO BnO O OAc O BnO BnO O OAc O BnO BnO O OAc 200 O BnO OTBDPS BnO

, 1) Me2SO, (COCl)2 i-Pr2EtN HO OAc BnO OH HO COOH CH2Cl2 O O NaClO , O BnO BnO 2) 2 2-methyl-2-butene HO t , BnO BnO BuOH, NaH2PO4 H2O HO , O O O 1) 200, SnCl2 AgClO4 , 50% 196, SnCl Et2O OAc 1) 2 AgClO4 OH COOH 3) TBAF, AcOH, THF 80% aq. AcOH Et2O 2) O O 75% O 62% 64% HO 199 BnO BnO BnO BnO HO O O O 3) AcCl, py 6 2) NaOMe, MeOH 10 4) 10% Pd/C in 80% aq. MeOH 10 81% 82% OAc OH 75% COOH O O O BnO OTBDPS BnO OTBDPS HO 201 BnO 202 BnO 203 HO OH

In order to synthesize some partially methyl-esterified hexagalacturonates, Madsen and co-workers introduced two orthogonal protecting groups at the C-6 positions of D-galactose: acetyl and para- methoxyphenyl (PMP). This enabled them to oxidize the C-6 position to the carboxylic acid or to the methyl ester selectively, and thereby synthesize tri- and hexagalacturonans with different methylation patterns.108,119 Their methodology was based on the repeated coupling of galactose mono- and disaccharide donors onto a galactose acceptor to produce hexagalacturonan backbones. They chose n- pentenyl donors, which provided good yields of the desired α-anomers. An example of their synthesis of two partially-esterified hexamers is shown in Scheme 22. Tetrasaccharide acceptor 211, formed from disaccharide 208 by sequential glycosylation with pentenyl donor 209 (ClAc = chloroacetate) was coupled with disaccharide donor 207 to form the pure α-(1→4)-linked hexagalactoside, which was further deacetylated to afford 212, the hexamer precursor of both targets. The oxidation to esters was carried out in a one-pot three-step procedure. The primary alcohols were first oxidized by Dess-

Martin periodinane120 to the aldehydes, which underwent a sodium chlorite oxidation to afford the corresponding carboxylic acids. The resulting acids were then esterified with either TMSCHN2 affording 213 or with PhCHN2 giving 214. The benzyl protection masked the positions destined to become carboxylic acids, and aided the purification of the intermediates. To complete the synthesis, the PMP-ethers were removed with cerium(IV) ammonium nitrate (CAN) and the primary alcohols were oxidized and esterified using the same one-pot three-step procedure, affording both benzyl and methyl esters. Hydrogenolysis afforded the two targets 215 and 216.

Scheme 22. Example of two partially methyl esterified hexagalacturonic acids prepared by Madsen and co-workers.109

HO OAc O BnO O OBn BnO OAc 206 , O BnO OAc NBS, Et2NSF3 BnO OAc SnCl2 AgClO4 BnO CH2Cl2 CH2Cl2 BnO O O OAc BnO O BnO O 55% OBn 74% BnO F O BnO OH BnO O O OBn 204 205 207 BnO BnO O OH OPMP ClAcO HO OPMP O BnO O O BnO O OPMP BnO BnO HO O OPMP OBn BnO HO OPMP O O OPMP O 209 BnO BnO O BnO O BnO 209, NIS, TESOTf OPMP 209, NIS, TESOTf BnO BnO 1) O 1) 1) 207, NIS, TESOTf O OPMP CH Cl CH Cl BnO 2 2 2 2 BnO OPMP CH2Cl2 O OAc O O O , BnO , BnO thiourea, NaHCO thiourea, NaHCO THF O 2) 3 BnO 2) 3 O 2) NaOMe, MeOH, BnO BnO OBn OAc BnO OPMP Bu4NI, THF O Bu4NI, THF 71% O BnO OBn 88% 90% OAc 208 210 O 211 O 212 O BnO OBn BnO OBn O BnO OBn BnO O OH OBn O BnO OBn OBn BnO HO COOR COOR1 O O BnO HO BnO HO O O COOR COOR1 O 1) CAN, MeCN, H2O O BnO HO BnO OPMP HO a) Dess-Martin periodinane, CH2Cl2 O 2a) Dess-Martin periodinane, CH2Cl2 O 2 , t , t COOR then NaClO2 NaH2PO4, THF, BuOH, H2O O then NaClO2 NaH2PO4, THF, BuOH, H2O O , BnO , HO then TMSCHN2 THF, MeOH then TMSCHN2 THF, MeOH (for 214) BnO HO O OPMP O COOR2 or O or O 212 BnO HO BnO HO b) Dess-Martin periodinane, CH2Cl2 OPMP 2b) Dess-Martin periodinane, CH2Cl2 O O 2 , t , , t COOR then NaClO2 NaH2PO4, THF, BuOH, H2O then NaClO2 NaH2PO4 THF, BuOH, H2O , O , O then PhCHN2 EtOAc BnO then PhCHN2 EtOAc (for 213) HO BnO HO O O , 1 2 1 a: 213: R = Me (44%) COOR 3) H2 Pd/C, THF, MeOH, H2O 215: R = Me, R = H COOR O (48%) O : = 1 2 b: 214 R Bn (59%) BnO OBn 216: R = H, R = Me (30%) HO OBn HO OH

5.2 Xylogalacturonan

Figure 15. Example of XGA structure. Xylogalacturonan (XGA) is similar to HG except that the backbone is substituted with single β-

(1→3)-linked xylose residues or short β-(1→4)-linked xylan side chains as shown in Figure 15.121

XGA-based glycans have not yet received attention from synthetic chemists.

5.3 Apiogalacturonan

Figure 16. Example of AGA structure showing various types of branching points.

Apiogalacturonan (AGA) is found in the cell walls of aquatic plants such as duckweeds (Lemnaceae) and marine seagrasses (Zosteraceae).122 In AGA, β-D-apiofuranosyl residues are attached to the 2- and/or 3-positions of selected GalpA monosaccharides in the HG chain in the form of mono- or sometimes short oligo-saccharides, as shown in Figure 16.6 While several groups have focused on the synthesis of apiose-containing oligosaccharides found in the RG-II region of pectin (see section 5.5), only two groups have reported the synthesis of oligosaccharides related to apiogalacturonan (see Table

11).

Table 11 Three different synthetic AGA oligosaccharides. Structure R Year Reference

Me 2004 Buffet et al.123 OR 124 2011 Nepogodiev et al.

Me 2011 Nepogodiev et al.124 OR

As D-apiose is a rare sugar, all syntheses of D-apiose-containing oligosaccharides start with the preparation of the appropriate apiose derivative, usually from D-mannose, L-arabinose or D-xylose.125–

128 Reimer and co-workers chose a pre-glycosylation oxidation approach to synthesize AGA disaccharide 225 as shown in Scheme 23. They started by preparing 2,3-O-isopropylidene apiofuranose 219 from D-mannose in five steps.126,127 Four additional transformations afforded thioapiofuranoside donor 222. Glycosylation of the 2-OH galacturonate acceptor 223 proceeded smoothly, giving disaccharide 224 in 83% yield. Deprotection afforded the disaccharide 225 in 76% yield.

Scheme 23. Synthesis of disaccharide 225 by Reimer and co-workers.123

H2SO4, aq. MeOH O HO BnO BnO 37% aq. HCOH then NaBH4 BnBr, NaH 1) 80% HCOOH O , O OH O OBn O R K2CO3 MeOH then NaIO4 DMF 2) Ac2O, py O O O O O O O OH O O O O AcO OAc O OH 86% 79% 92% 81% OH . 217 218 219 220 TolSH, BF3 Et2O 221: R = OAc 90% 222: R = STol

O , HO CO Me 1) Pd(OAc)2 H2 CO H 222, NIS, 2 2 O AgOTf O EtOAc, MeOH, AcOH O CO2Me CH Cl O HO O 2 2 90% O O O BnO OMe HO OMe 84% O 2) NaOMe, MeOH O HO OMe + 3) H2O, Amberlite H AcO OAc 4) NaOH (aq.) HO OH 85% steps) 223 224 (three 225

Some years later, in 2011, Field and co-workers synthesized the AGA fragments 225, 234 and 237 shown in Scheme 24.124 Like Reimer and co-workers, they prepared apiofuranoside donor 226 from

D-mannose according to previously published procedures. Treatment of arylthio glycoside 226 with

90% TFA followed by acetylation afforded triacetate 227 as an anomeric mixture. Coupling between

227 and 223 afforded the β-linked disaccharide 228 exclusively in 79% yield, which was further deprotected to give the target disaccharide 225. Galacturonate acceptors with unprotected 3-OH or both 2- and 3-OH groups (239 and 235, respectively) were prepared and coupled with 227, leading to di- and tri-saccharides 233 and 236 in good yields. Cleavage of the ester protecting groups afforded the targeted apiogalacturonans 234 and 237.

Scheme 24. Synthesis of di- and trisaccharide AGA fragments by Field and co-workers.124

O CO2Me O O HO OMe 223 O HO 1) 90% TFA CO Me 1) 80% aq. AcOH CO H AcO STol 2 2 2) Ac2O, DMAP NIS, HClO4/SiO2 O 2) NaOMe, MeOH O O AcO O HO py O CH2Cl2 3) NaOH, H2O STol O O O O AcO OMe HO OMe 76% AcO OAc 79% O 80% O

AcO OAc HO OH 226 227 228 225

227, NIS, AcO , CO Me HO O O AcO HClO4/SiO2 2 CO Me MeC(OEt)3 TsOH CO Me CO Me CO Me O 2 2 BzCl, py 2 80% AcOH 2 CH Cl O THF EtO O EtO O O 2 2 O HO O O HO AcO O BzO HO 73% HO 78% BzO 79% BzO 74% OMe OMe OMe OMe OMe 229 230 231 232 AcO OAc 233

80% AcOH 1) NaOMe, MeOH 95% 2) NaOH, H2O, MeOH

AcO 66% CO2Me O HO HO HO CO2H OMe O 235 O HO , HO 227, NIS, HClO4/SiO2 CH2Cl2 O OMe 84% HO OH

234 AcO HO CO2Me CO2H O O O O AcO 1) NaOMe, MeOH HO O O O O OMe 2) Et3N, MeOH, H2O OMe

AcO OAc 73% HO OH AcO HO O O

AcO OAc HO OH 236 237

5.4 Rhamnogalacturonan I

Figure 17. Representation of RG-I structure.98

RG-I (Figure 17) makes up 20–35% of pectin. Its average length has not been determined, since isolated RG-I domains are generally accompanied by HG. However, the RG-I backbone has generally been found to be shorter than HG. Furthermore, the RG-I backbone is predicted to have a three-fold helix conformation.129 RG-I polysaccharides have repeating disaccharide units of (1→2)-α-L- rhamnopyranosyl-(1→4)-α-D-galactopyranuronic acid. The GalpA residues may be O-acetylated on

C2-O and/or C3-O as seen for HG, but no methyl esterification has been observed.98 The diversity of

RG-I stems from the variety of side chains that can decorate the C-4 positions of rhamnose. Depending on the plant source and the method of isolation, 20–80% of the rhamnosyl residues are found to be substituted. These side chains consist of three different types of branched oligosaccharides of varying length (1-20 residues): galactans, arabinogalactans (AG), and arabinans.98,130 Pectic arabinogalactans are usually divided into two main groups.131 Type I arabinogalactan (AG-I) is the most prevalent neutral pectic glycan and is especially abundant in citrus fruits and apples. It is characterized by a

70 linear β-(1→4)-linked D-Galp chain, which can be substituted at the C-3 position by L-Araf residues and occasionally by arabinosyl oligosaccharides. It sometimes also carries galactan at the C-6 position. Type II arabinogalactan (AG-II) are highly branched polysaccharides composed of a backbone of β-(1→3) or (1→6)-linked D-galactan, which can be (arabino)galactan-substituted at the

C-6 or C-3 positions, respectively.97 Galactans are simply β-(1→4)-linked D-galactan chains and have been observed in lengths of up to 40 residues. Finally, the arabinans consist of α-(1→5)-linked L-

Araf residues with some degree of (1→3)-branching. The glycosyl residues α-L-fucosyl, β-D- glucuronosyl and 4-O-methyl β-D-glucuronosyl may also be present.98 The (arabino)galactans and arabinans will be discussed separately in section 6.

RG-I backbone

Figure 18. RG-I backbone structure.

Several syntheses of fragments of the RG-I backbone (Figure 18) have been carried out and an overview of the different oligosaccharides produced is shown in Table 12.

Table 12. Synthetic fragments of RG-I backbone. Structure R Year Reference

Pr 2013 Ma et al.132 OR

Pr 2000 Maruyama et al.133 OR 2008 Nemati et al.134

H 2010 Scanlan et al.129

Me 2008 Reiffarth & Reimer135 OR

Pr 2013 Ma et al.132 OR

H 2013 Zakharova et al.136 OR

Pr 2013 Ma et al.132 OR

As with HG fragment syntheses, and more generally all oligosaccharides containing uronic acids, both pre-glycosylation oxidation and post-glycosylation oxidation approaches have been investigated by different groups. The former approach was chosen by Reiffarth & Reimer135, also Vogel and co-

72 workers134 and Yu and co-workers.132 In 2008, Reiffarth and Reimer135 used a block synthesis approach, coupling two disaccharide repeating units derived from the same disaccharide precursor, to form their target rhamnogalacturonan tetrasaccharide 244 as shown in Scheme 25. The C-4 positions of the rhamnosyl residues were orthogonally protected with an allyl group in order to allow potential introduction of RG-I side chains. An acetyl group was used to protect their C-2 positions to enable elongation of the backbone. Rhamnosyl thioglycoside donor 237 and methyl galacturonate acceptor

238 were coupled to form the key disaccharide intermediate 239. This was then further transformed in a three-step sequence to generate a new disaccharide acceptor 240, by removal of the 2-acetyl group, and a new trichloroacetimidate disaccharide donor 241. However, glycosylation of 240 with donor 241 turned out to be more difficult than expected. Tetrasaccharide 242 was obtained in only

39% yield and attempts to optimize the reaction by changing the promoter and donor types were unsuccessful. The fully deprotected tetrasaccharide methyl ester 243 was then reached over four steps in 33% overall yield and ester hydrolysis using NaOH followed by acidification afforded the desired rhamnogalacturonic acid tetrasaccharide 244 in 77% yield.

Scheme 25. Synthesis of an RG-I backbone tetrasaccharide fragment by Reiffarth & Reimer.135

HO COOMe O OAc OH OAc BnO BzO BzO BzO BnO AllO AllO AllO SEt OMe O HCl, O O 238 MeOH, CH2Cl2 O O O O AllO COOMe COOMe COOMe BzO NIS, TfOH O 80% O O OAc CH2Cl2 BnO BnO BnO 78% BnO BnO BnO O CCl3 OMe OMe 237 239 240 241 NH H , Ac 1) 2SO 4 AcOH, 2O (81%) 4Å 2) MS, CH2Cl2, MeOH (89%) 3) CCl3CN, DBU, CH2Cl2 (81%)

OAc OH OH BzO HO HO AllO HO HO O O O Cl, DABCO O 1) Rh(PPh3)3 O O COOMe EtOH, toluene, H2O COOMe COOH O O O BnO 2) HgO, HgCl2 HO HO AgOTf acetone, H2O aq. NaOH, then BnO NaOMe, MeOH HO HO CH2Cl2 O 3) O aq. HCl O 240 + 241 BzO HO HO , AllO H HO HO 39% O 4) 2 Pd(OAc)2 O 77% O EtOAc, MeOH, AcOH O O O COOMe 33% steps) COOMe COOH O (4 O O BnO HO HO BnO HO HO OMe OMe OMe 242 243 244

In 2008, Vogel and co-workers134 used a similar synthetic approach. Unlike Reiffarth & Reimer, trimethylsilyl trifluoromethanesulfonate-promoted glycosylation with a similar trichloroacetimidate disaccharide donor and disaccharide acceptor (O-3’ and O-4’ protection was modified compared to

240 and 241),provided them with the desired tetrasaccharide in 60% yield. More recently, in 2013,

Yu and co-workers132 reported the synthesis of tri-, penta-, and heptasaccharide fragments (251, 254 and 257) of the RG-I backbone using the pre-glycosylation oxidation approach with an iterative elongation of the chain, using disaccharide building block 248 (Scheme 26). The syntheses feature highly stereoselective formation of the α-D-GalpA-linkage, with very good yields, when using N- phenyltrifluoroacetimidates as donors.

Scheme 26. Synthesis of tri-, penta- and heptasaccharide fragments of RG-I backbone by Yu and co- workers.132

OAll HO COOMe BnO O AcO O AcO COOMe BnO OAll BnO COOMe O AcO OH O BnO OBn COOMe 246 BnO 1) PdCl2 O THF, MeOH BnO 249 BnO BnO O O BnO BnO O CF3 TBSOTf BnO 2) PhN=C(Cl)CF3 BnO TBSOTf CH Cl BnO O 2 2 O K2CO3, acetone NPh O CF3 CH2Cl2

245 68% 247 OAll 248 87% NPh 88%

AcO HO COOMe COOH O O BnO HO BnO LiOH HO O 1) O BnO MeOH, H2O, 87% HO BnO HO O , O 2) H2 Pd(OH)2/C O MeOH, EtOAc, 92% O COOMe COOH O O BnO OAll HO OPr OBn OH 250 251

NaOMe AcO HO MeOH COOMe COOH O O HO 87% BnO BnO HO O O BnO HO HO BnO HO COOMe O O O BnO 248 O O BnO TBSOTf COOMe 1) LiOH COOH O O BnO toluene O MeOH, H2O, 82% BnO HO BnO , HO O 99% BnO H /C O O 2) 2 Pd(OH)2 HO BnO MeOH, EtOAc, 99% O HO COOMe BnO O O O BnO OAll O O OBn COOMe COOH O O BnO OAll HO OPr 252 253 OBn 254 OH

NaOMe MeOH

86% AcO HO COOMe COOH O O HO HO COOMe BnO O BnO HO BnO O O BnO HO BnO HO O BnO O BnO O BnO O O O COOMe COOH O O 248 HO O BnO 1) LiOH COOMe TBSOTf HO O BnO MeOH, H2O, 82% toluene O BnO O HO BnO , BnO H /C O 92% BnO 2) 2 Pd(OH)2 HO BnO O MeOH, EtOAc, 87% O BnO O O O COOMe COOH O O O COOMe BnO HO O BnO HO BnO OAll O O BnO HO OBn BnO HO O O O O COOMe COOH O O BnO OAll HO OPr OBn OH 255 256 257

The three other groups who have synthesized RG-I backbone fragments used the post-glycosylation approach, i.e. synthesizing a neutral backbone and oxidizing the primary hydroxyl groups to the desired carboxylic acid functionality afterwards. 129,133,136 In 2000, Takeda and co-workers prepared the unprotected propyl RG-I backbone tetrasaccharide 269 by employing trichloroacetimidate glycosyl donors (see Scheme 27).133 Selected rhamnose residues bore PMB protective groups at C-4, allowing for potential introduction of RG-I side chains.

Scheme 27. Synthesis of fully unprotected propyl glycosyl tetrasaccharide of RG-I backbone by Takeda and co-workers.133

O CCl3

AcO O AcO OAc OR OR2 RO O R3O O 259 RO R1O HO OBn O RO O AgOTf 2,2-dimethoxypropane O O CH Cl TsOH 80% AcOH BnO OAll 2 2 O OBn O OBn O OBn OBn 97% O 93% 7a 67% O BnO OAll O BnO OAll BnO OAll 7b 94% OBn OBn OBn

258 NaOMe, MeOH 260: R = Ac 262: R = H : 1 = 2 = 3 = PMBCl, NaH, DMF 1) Bu2SnO, benzene 264a R PMB, R H, R H 70% 261: R = H 65% 263a: R = PMB 2) BnBr, TBABr : = 93% 1 2 3 BnBr, NaH, DMF 262 R H 265a: R = PMB, R = H, R = Bn 82% = Ac 71% 263b R Bn 2O, py, 266a: R1 = PMB, R2 = Ac, R3 = Bn OAc BnO 1 2 3 1) Bu2SnO, benzene 264b: R = Bn, R = H, R = H , RO PdCl NaOAc, O 2) BnBr, TBABr 1) 2 AcOH 91% 1 2 3 90% aq. OBn 265b: R = Bn, R = H, R = Bn O Ac O, py, 99% 2 266b: R1 = Bn, R2 = Ac, R3 = Bn CCl CN, DBU O 2) 3 BnO CH2Cl2 BnO O CCl3 NH 267a: R = PMB (17%) 267b: R = Bn (30%)

OAc OH BnO HO BnO HO O O

AgOTf O OBn 1) NaOMe, MeOH, 93% O Pd/C, H , MeOH, 74% COOH CH2Cl2 O 2) 2 O 265a + 267b BnO HO 67% BnO 3) TEMPO, KBr, NaClO BnO O , O BnO aq. NaHCO3 50% HO PMBO HO O O

O OBn O COOH O O BnO OAll HO OPr OBn OH 268 269

AgOTf-catalyzed glycosylation between allyl galactoside acceptor 258 and trichloroacetimidate donor 259 afforded the α-linked disaccharide 260 in 97% yield as a single stereoisomer. Protecting group manipulations afforded four differently protected disaccharides: two trichloroacetimidate donors 267a and 267b (PMB and Bn protected at the Rhap C-4, respectively) and two 2’-OH acceptors

265a and 265b. One noteworthy aspect is that the two-step conversion into trichloroacetimidate donors 267a and 267b resulted in modest yields (17% and 30%, respectively). Coupling of 265a and

267b gave the α-linked tetrasaccharide 268 in 67% yield. Deprotection followed by selective oxidation

77 of the primary hydroxyl groups with 2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPO), KBr and NaClO afforded the desired unprotected propyl tetrasaccharide 269 in 34% yield over the three steps.

In the same year, Davis and co-workers used an orthogonal approach, combined with a late-stage oxidation strategy, to synthesize a fully unprotected RG-I backbone tetrasaccharide 281 and its corresponding methyl ester (see Scheme 28).129 Surprisingly, initial attempts to couple a galactorhamnosyl disaccharide donor to Galp C4’-OH of a disaccharide acceptor failed. This was probably due to low reactivity of the acceptor, and forced the authors to change strategy and assemble the RG-I tetrasaccharide through galactosylation instead of rhamnosylation. The group used both trichloroacetimidates and ethylthioglycosides as donors, affording good glycosylation yields and α- stereoselectivity.

Scheme 28. Davis and co-workers’ synthesis of an RG-I backbone tetrasaccharide 281.129

HO OBz O BnO SEt OBn OAc NH 271 BnO BnO TMSOTf O O CCl 3 CH2Cl2 O OBz BnO O 65% O BnO OAc BnO SEt OBn 270 272 HO OBz O BnO OBn OBn OR 274 BnO NIS, TMSOTf BnO SEt O CH2Cl2 O BnO O OBz BnO 75% OAc O BnO OBn OBn

: = 273 NaOMe, MeOH, THF 275 R Ac 84% 276: R = H

OR1 OH BnO HO BnO HO O O

O O 276 R2 COOH NIS, TMSOTf O O BnO , HO CH2Cl2 Pd/C, H2 H2O 272 BnO HO O O 83% BnO 62% HO BnO HO O O

O O R2 COOH O O BnO OBn HO OH OBn OH 281 1 2 277: R = Ac, R = CH2OBz NaOMe, MeOH, THF 78% 1 2 278: R = H, R = CH2OH TEMPO, NaClO , NaClO 2 279: R1 = H, R2 = COOH , 1 2 Cs2CO3 BnBr, DMF 280: R = H, R = COOBn 72% over 2 steps

In 2013, Clausen and co-workers synthesized a fully unprotected hexasaccharide backbone of RG- I using a strategy relying on iterative coupling of a common pentenyl disaccharide glycosyl donor

284.136 The latter was prepared by an efficient chemoselective armed-disarmed coupling of a thiophenyl rhamnoside donor 282, with a pentenyl galactoside acceptor 283 bearing the strongly electron-withdrawing pentafluorobenzoyl ester (PFBz) protective group, as shown in Scheme 29. Two iterative glycosylations using pentenyl disaccharide 284 afforded hexasaccharide 289 in reasonable yields. Removal of the PFBz groups under Zemplén conditions afforded the Galp primary alcohols,

79 which were oxidized with Dess-Martin periodinane to the aldehydes and then with NaClO2 to the desired carboxylic acids. As in many other cases, the carboxylic acids were protected with benzyl esters in order to facilitate purification. Palladium catalyzed hydrogenolysis gave the fully unprotected hexasaccharide 291.

Scheme 29. Fully unprotected hexasaccharide backbone of RG-I synthesized by Clausen and co- workers.136

OR HO OPFBz BnO BnO O O BnO O OPFBz OBn ONAP OR O BnO BnO 283 BnO 284 O ONAP BnO , O BnO BnO NIS, TESOTf O 1) Br2 CH2Cl2 NIS, TESOTf BnO Et O 2) BnOH, TBABr OPFBz Et O BnO O 2 O OPFBz O 2 O BnO O SPh 78% O 90% 71% BnO BnO O BnO O OBn BnO OBn O OPFBz , 282 284 DDQ, H2O, CH2Cl2 MeOH 285: R = NAP O 74% 286: R = H BnO BnO OBn , DDQ, H2O, CH2Cl2 MeOH 287: R = NAP 76% 288: R = H

ONAP ONAP OH BnO BnO HO BnO BnO HO O O O OH O O O COOBn COOH O O O BnO BnO HO DMP, CH Cl BnO 1) 2 2 BnO HO O NaClO NaH PO O BnO 2) 2, 2 4 BnO O , HO 1) 284, NIS, TESOTf, BnO 2-methyl-2-butene BnO H2 Pd/C, MeOH HO O t O O Et2O BuOH, H2O, THF THF, H2O O OH O COOBn O MeONa, MeOH , 95% COOH 2) O 3) PhCHN2 EtOAc O O BnO BnO HO 40% BnO BnO HO O 60% O O BnO BnO HO BnO BnO HO O O O OH O O O COOBn COOH O O O BnO BnO 289 290 291 HO BnO BnO OBn OBn HO OH

RG-I side chains

As explained in the introduction to rhamnogalacturonan I (section 5.4), RG-I has three main types of side chain: galactans, arabinogalactans and arabinans. Since (arabino)galactans are also found in

80 arabinogalactan proteins (AGPs), oligosaccharides related to these structures will be discussed separately in Chapter 6.

5.5 Rhamnogalacturonan II

D C α-stereochemistry discovered in 2017

Rare sugar structures:

HO O D-apiose OH HO OH

HOOC OH O L-aceric acid OH Me OH

OH MeO HO D-2-keto-3-deoxy-D-manno- OH octulosonic acid O (Kdo) HO COOH A B OH

HO OMe COOH D-3-deoxy-D-lyxo- O heptulosaric acid (Dha) HO COOH OH

Figure 19. Schematic representation of RG-II based on the findings of Gilbert and co-workers.137

Rhamnogalacturonan II (RG-II) constitutes ~10% of pectin in the cell wall of higher plants and is a structurally complex branched pectic polysaccharide.97 It is present in above-average quantities in wine and other beverages obtained by fermentation of fruits and vegetables (one liter of red wine contains between 100 and 150 mg of RG-II while around 20 mg is present in a liter of white wine).138

Therefore, red wine has been used as a source of RG-II in several studies of this polysaccharide.139

Unlike RG-I, the RG-II structure has been regarded to have remarkably little variation and to be evolutionarily conserved in all vascular plants. Variations are minor and are limited to methylation and/or acetylation of the side chains.140 However, recent discoveries have shown evidence of variation also in the monosaccharide composition of the side chains.141 RG-II has an HG backbone, made up of

81 seven to nine residues, and substituted by four different side chains (A–D), which can contain up to

29 monosaccharides, including a number of rare sugars (see Figure 19).100,101,142 These comprise L- aceric acid (AcefA; 3-C-carboxy-5-deoxy-L-xylose), 2-keto-3-deoxy-D-manno-octulosonic acid

(Kdop), 2-keto-3-deoxy-D-lyxo-heptulosaric acid (Dhap), D-apiose (Apif; 3-C-(hydroxymethyl)-β-D- erythrose), L-galactose, 2-O-methyl-D-xylose and 2-O-methyl-L-fucose. The monosaccharides are connected by over 20 different glycosidic linkages. Side chains A and B are the largest and have been described as highly ramified octa- and heptasaccharides, respectively.143,144 The C and D side chains are disaccharides, and two extra side chains E and F, composed of a single L-arabinose residue linked to a mono- or di-substituted galacturonic acid residue respectively, have been isolated from red wine

RG-II recently.100,137,145

Before 2000, knowledge of the primary structure of RG-II was mostly based on the determination of the glycosidic linkage composition (using methylation analysis) and identification of various oligosaccharide fragments by mass spectroscopy (after acid and/or enzymatic hydrolysis of RG-II).

Recently, several NMR solution studies have helped determine the fine- and three-dimensional structure of the oligosaccharide.144,146,147 In 2000, Vidal and co-workers published a convincing NMR study of an RG-II fragment, which contained a hexasaccharide homogalacturonan backbone substituted by side chains B and D.146 Their structural studies showed that the original configurations postulated for the anomeric centers of two residues in the B-chain were incorrect. This led some other groups to look in more detail at the structure of RG-II. In 2003, Pérez and co-workers reported a thorough NMR investigation of RG-II. They reassigned most of the RG-II primary structure and used

NOESY data to propose a conformational model of the oligosaccharide in solution.145 Nevertheless, several NMR assignments remained ambiguous. In 2017, Gilbert and co-workers published a revised

82 model of the RG-II structure based on an impressive work including crystal structure of side chain B in complex with a glycosyl hydrolase combined with enzymatic degradation studies. This revised structure is the one shown in Figure 19.137 The main difference compared to previously proposed structures of RG-II is highlighted by pink circles: the L-Rhap-D-Apif linkage in chains A and B is α and not β. It has been observed that more than 90% of the RG-II in the cell wall self-assembles into a covalently cross-linked dimer in the presence of boron. RG-II is dimerized by borate ester formation involving the apiosyl residues of side chain A belonging to two different RG-II monomers, which results in networks of pectic polysaccharides.100 This secondary structure has been confirmed by 11B-

NMR.148 Several studies support the idea that RG-II and its borate crosslinking support cell adhesion and wall mechanical strength and are therefore significant for plant growth and development.149,150

All the knowledge about RG-II structure and function accumulated before 2004 is summarized in a comprehensive review by Darvill and co-workers.100

Most synthetic efforts have focused on side chain A and B, which also constitute most of the RG-II structure, as mentioned previously. Since many of the synthesized fragments are only disaccharides, the goal of synthesizing them was often not so much to study RG-II function, but rather to develop synthetic methods. Furthermore, as the chemical synthesis of some of the rare sugars present in RG-

II is not straightforward, we will refer the reader to selected articles reporting methods that have been developed for their preparation.

RG-II side chain A

Despite the fact that RG-II side chains were thought to be highly conserved in the plant kingdom, some variations in the composition of side chain A were reported by Léonard and co-workers in

2013.141 They showed that in wild-type Arabidopsis plants, up to 45% of L-fucose was replaced by L- galactose (see Figure 20). The study also provided evidence that a large proportion of the GlcpA residues were methyl esterified and some of the β-GalpA residues were O-methylated at positions 3 and 4 (these variations are highlighted in red on Figure 20). Furthermore, until the work published by

Gilbert and co-workers in 2017, it was believed that the Rhap-Apif linkage (shown in pink-dot circle on Figure 20) was a β-linkage, which was proved to be the wrong stereochemistry by Gilbert’s group.

Or COOMe Or COOMe

MeO MeO

MeO MeO A A OMe OMe

α-stereochemistry discovered in 2017

Figure 20. Schematic representation of side chain A including the variations across different plant species observed by Léonard and co-workers.141,145

Side chain A is attached to the homogalacturonan backbone residue via a D-Apif residue.151,152 One- and two-dimensional 1H NMR spectroscopic analysis showed that apiose is β-(1→2)-linked to

GalpA.146 The synthesized fragments belonging exclusively to side chain A are reported in Table 13, while fragments found in both side chain A and B are presented in Table 14.

Table 13. Synthetic fragments found exclusively in RG-II side chain A. Structure R Year Reference

153 OR Me 1990 Backman et al.

154 OR Me 2005 Chauvin et al.

Jansson and co-workers synthesized a Fucp-Rhap disaccharide by MeOTf-promoted glycosylation between 2,3,4-tri-O-benzyl-1-thio-β-L-fucopyranoside 292 and methyl 2,3-isopropylidene-α-L- rhamnopyranoside 293 affording the desired α-linked disaccharide 294 in 77% yield as shown in

Scheme 30.153

Scheme 30. α-L-Fucp(1→2)-α-L-RhapOMe fragment synthesized by Jansson and co-workers.153

OMe Me O HO O O

293 OMe OMe MeOTf 90% aq. AcOH Me O , Me O Me O SEt Et2O O then H2 Pd/C O OBn O HO O OH OBn 77% 72% BnO Me O Me O OBn OH OBn OH BnO HO 292 294 295

In 2005, Nepogodiev, Field and co-workers achieved the synthesis of branched tetrasaccharide fragment 308 following the route shown in Scheme 31. As this tetrasaccharide contains a combination of two 6-deoxy sugar residues (Rhap and Fucp) and two uronic acids (GalpA), they chose a post-

85 glycosylation oxidation strategy in order to introduce the carboxylic acids in the very last step of their synthesis. The method was based on the use of the differentially protected rhamnoside 299. Sequential glycosylation afforded the neutral tetrasaccharide 307 in reasonable yields and stereoselectivities overall. Following deprotection of 307, the primary hydroxyl groups were successfully oxidized using

TEMPO affording the target 308 in 47% yield.

Scheme 31. Synthesis of a branched tetrasaccharide fragment from side chain A by Nepogodiev, Field and co-workers.154

OBz BzO BnO OBn O O BzO SMe BnO SMe OBz OMe OBn Me 300 O 303 OMe PMBO OMe MeC(OEt)3 OMe NIS, TfOH NIS, TMSOTf 80% aq. AcOH O TsOH Me CH2Cl2 OBz OR CH Cl Me O O Me O O 2 2 HO RO PMBO BzO HO MeCN O HO 70% OBz OH O OAc 87% Me BzO OEt 296 PMBCl, NaH, 297: R = H 299 0.4 M HCl in MeOH 301: R = Ac MeCN 298: R = PMB (70% 3 steps) 45% 302: R = H

Me SPh O OBn OBn BnO OMe OMe OMe 306 Me 1) 0.05% NaOMe, MeOH Me NIS equiv.), O O , O O Me O (2 then H2 Pd/C, EtOH RO TfOH equiv.) OBn (2 Me O O 61% Me O O OBn O O OH O R OBz O OBn O Et2O, CH2Cl2 OBn OBz O TEMPO, NaOCl, KBr OH O BzO BnO O BnO OBn 2) HO O HO OH BzO BnO H O HOR HO OBz O OBz 2 BzO BnO BnO OBn 87% 47% OH BzO HO

CAN, MeCN, H2O 304: R = PMB 307 308: R = COOH 93% 305: R = H

The linear trisaccharide α-L-Rhap-(1→3’)-β-D-Apif-(1→2)-α-D-Galp is a fragment that is common to both side chain A and B. However, since the linkage between the rhamnose and the apiose residues was believed to be β until 2017, chemists focused on synthesizing fragments with this configuration.

Reimer and co-workers123 and Nepogodiev and co-workers155 both reported the synthesis of disaccharide elements of the fragment in 2004 (β-D-Apif(1→2)-α-D-Galp and β-L-Rhap-(1→3’)-β-D-

Apif, respectively). In 2011, Nepogodiev, Field and co-workers124 reported the synthesis of the full structure (see Table 14).

Table 14. Synthetic fragments found in both side chain A and side chain B of RG-II. Structure R Year Reference

Me 2004 Buffet et al.123 OR Me 2011 Nepogodiev et al.124

OR Me 2004 Chauvin et al.155

Me 2011 Nepogodiev et al.124 OR

All these syntheses required access to the rare sugar apiose. As explained in section 5.3, D-apiose derivatives are usually produced from D-mannose, L-arabinose or D-xylose.125–128 The approach by

Reimer and co-workers was described earlier (see Scheme 23). The two other fragments synthesized, required the challenging β-rhamnosylation to form β-L-Rhap-(1→3)-β-D-Apif disaccharide. Unlike for β-mannosylation, only few methods for β-rhamnosylation have been reported.156,157 The first report by Hodosi and co-workers in 1997 described the SN2 coupling between a sugar triflate and an unprotected 1,2-O-stannylene acetal of L-rhamnopyranose.157 In 2003, Crich and Picione reported a second method enabling the direct formation of a β-L-rhamnoside by means of thiorhamnoside donors protected with a 2-O-sulfonate ester and, ideally, a 4-O-benzoyl ester.156 After attempting in vain to apply Hodosi’s method for the synthesis of β-L-Rhap-(1→3)-β-D-Apif, Nepogodiev and co-workers decided to use a cyclic 2,3-carbonate rhamnosyl bromide donor as shown in Scheme 32.124 It is worth mentioning that the synthesis of thiophenyl apiofuranoside residue 310 required extensive

87 optimization and was achieved through a tedious 9-steps synthesis from L-arabinose. The authors did test simpler protecting group schemes, but these strategies were unsuccessful. Koenigs-Knorr β- rhamnosylation between 313 and apiogalacturonate acceptor 312 afforded the desired core trisaccharide 314 in 59% yield, which was deprotected in four steps to reach the target molecule. In

2012, Yasomanee and Demchenko demonstrated a third approach by forming a β-rhamnoside employing a picoloyl protecting group in the 3-position of a thioethyl rhamnose donor.158

Scheme 32. Synthesis of trisaccharide β-L-Rhap-(1→3’)-β-D-Apif(1→2)-α-D-Galp by Nepogodiev, Field and co-workers.124

Br ClAcO STol O Me AcO O O O O O

Ph H O O O , 310 COOMe COOMe 1) H2 Pd/C, EtOAc 313 HO O O 2) 80% aq. AcOH COOH NIS, TfOH O Ag O O O 2 3) NaOMe, MeOH O COOMe CH Cl O CH Cl Me O HO O 2 2 RO 2 2 AcO O O 56% O OMe O OMe Me O O O O HO O 62% 59% O O OMe HO 4) Et3N, H2O, MeOH HO OMe O O O O 78% OH O HO OH Ph H Ph H 309 311: R = ClAc 314 315 NaOMe, MeOH, 96% 312: R = H

RG-II side chain B

α-stereochemistry discovered in 2017

OMe

not conserved

Figure 21. Side chain B of RG-II.

The most common structure of side chain B is represented in Figure 21. However, some variations have been observed depending on the plant species, which mainly involve the presence or absence of substituents at the O-2 and/or O-3 positions of the Arap residue.100 The different sugar substituents, which have been observed at these two positions by Pérez and co-workers, are summarized in Table

15. Furthermore, mono-O- or di-O-acetylation can occur at the O-3 position of the AcefA residue as well as the O-3/O-4 positions of 2-O-Me-Fucp.145

Table 15. Plant dependent variations observed by Pérez and co-workers for side chain B of RG-II.145 Glycose linked to Arap Plant Source

O-2 O-3

β-L-Araf-(1→2)-α-L-Rhap α-L-Rhap Grape (wine)

Ginseng

Sycamore

3-O-Me α-L-Rhap 3-O-Me α-L-Rhap P. bifurcatum

P. nudum

α-L-Rhap 3-O-Me α-L-Rhap C. thalictroides

L. tristachyum

P. bifurcatum

α-L-Rhap OH A. thaliana

E. hyemale

S. kraussiana

The synthetic fragments of side chain B that have been reported to date are summarized in Table 16 and Table 14.

Table 16. Synthetic fragments found exclusively in RG-II side chain B. Structure R Year Reference

159 OR C8H17 2000 Mukherjee et al.

All 2001 Amer et al.160 OMe

161 OR (CH2)3NH2 2008 Rao et al.

162 OR (CH2)3NH2 2007 Rao & Boons

OMe

162 OR (CH2)3NH2 2007 Rao & Boons

OMe

Although aceric acid (AcefA) has only been found in RG-II, and the development of synthetic routes to this unusual sugar started over five years ago, no RG-II fragments including aceric acid have yet been reported. Two main approaches have been developed for its synthesis: a 4-step approach from

L-xylose163 and a 6-step synthesis from D-arabinose.164 Both Hindsgaul and co-workers159 and Kosma and co-workers160 reported the synthesis of 2-O-Me-α-L-Fucp(1→2)-β-D-Galp early in the 21st

Century. The synthetic route employed by Hindsgaul and co-workers159 afforded the desired β-linked disaccharide 321 in a straightforward manner (Scheme 33).

Scheme 33. Synthesis of 2-O-Me-α-L-Fucp(1→2)-β-D-Galp by Hindsgaul and co-workers159

O SEt OPMB OBn BnO 317 BnO OBn OBn DMTST, DTBMP O BnO BnO CH CH2Cl2 O(CH2)7 3 O O BnO O(CH2)7CH3 67% OH O OR OBn BnO 316 318: R = PMB CAN, MeCN, H2O, 70% 319: R = H MeI, NaH, DMF, 86% 320: R = Me

HO OH , H2 Pd(OH)2 O MeOH HO O(CH2)7CH3 O 89% O OMe OH HO 321

In 2007 Rao & Boons reported the synthesis of two highly challenging targets: tetra- and hexasaccharide portions of side chain B containing the central L-Arap (338 and 337, respectively,

Scheme 34).162 The tetrasaccharide 338 was chosen as a second target because some plant species,

91 such as Arabidopsis, lack the β-L-Araf-(1→3)-L-Rhap substituent on the 2-position of the L-Arap unit.

This chemical synthesis was challenging because it involved a number of unusual monosaccharides, due to the steric hindrance around the central arabinopyranoside as well as the presence of both

(1→2)-cis-linked 2-O-methyl-α-L-fucoside and β-L-arabinofuranoside residues, which are difficult to install stereoselectively. The authors first attempted a convergent [2+2+2] approach. However, the final glycosylation between tetrasaccharide acceptor 331 and a β-L-Araf-(1→3)-L-Rhap donor gave an unacceptably low yield (8%). The strategy was therefore revised and the final two monosaccharides were added sequentially as shown in Scheme 34.

Glycosylation between 2-O-methyl thioethylfucoside donor 322 and galactosyl acceptor 323 carrying an electron withdrawing benzoyl ester group at the C-3 position afforded the desired α-(1→2)-linked disaccharide 324 in high yield and with high α-selectivity. Reductive opening of the benzylidene acetal followed replacement of the ester protecting groups with benzyl ethers, in order increase the reactivity of acceptor 326.

Scheme 34. Tetra- and hexasaccharide portions of side chain B synthesized by Rao & Boons.162

Ph O O Ph O BzO O(CH2)3N3 O OBn OH O HO 323 O O R1O N BnO N O(CH2)3 3 , O(CH2)3 3 NIS,TfOH O NaCNBH3 THF O , O SEt CH2Cl2 Et2O 2 M HCl, Et2O OMe O O OMe OMe OAc 2 85% AcO 88% 2 OR R O BnO OBn

322 1) NaOMe, MeOH 324: R1 = Bz, R2 = Ac 326 NaH, BnBr, TBAI, DMF 1 2 2) 325: R = R = Bn (80%)

OBz SPh O HO SPh OBz BzO O OLev O OBn LevO 328 O O OBz OBz OR O 332 326 O O AcO BnO O(CH2)3N3 NIS, TfOH AcO O O CCl3 O SPh NIS, TfOH TMSOTf, CH2Cl2 AcO O CH2Cl2 AcO OLev CH2Cl2 OAc OAc NH 91% AcO O O 71% OMe 60% AcO OAc BnO OBn ⋅ 327 329 NH NH AcOH 330: R = Lev 2 ,2 CH2Cl2 MeOH 331: R = H (85%)

NH OBz OBz tBu2Si O O O O O OBn O CCl3 O OBn O O O O O OBn O O O BnO N O BnO N AcO O(CH2)3 3 335 AcO O(CH2)3 3 AcO O AcO O TBAF, THF OAc TMSOTf, CH2Cl2 OAc 1) O O O OMe 92% O OMe BzO BzO 2) NaOMe, MeOH OBn tBu2Si O OBn RO BnO O BnO , t ⋅ OBz OBz 3) Pd/C, H2 BuOH, AcOH, H2O NH NH AcOH 333: R = Lev O O 2 2 , : R = H OBn CH2Cl2 MeOH 334 (90%) 336 51%

OH O OH O O O O HO O HO O(CH2)3NH2 HO O OH O OMe HO O OH HO OH O OH O HO OH 337 OH O OH O O OH O 1) NaOMe, MeOH O HO HO O(CH2)3NH2 Pd/C, H , tBuOH, AcOH, H O 2) 2 2 HO O 330 OH O 65% OMe HO OH 338

Chemoselective TMSOTf-promoted glycosylation between rhamnopyranosyl trichloroacetimidate donor 327 and thiophenyl arabinopyranoside 328 gave the desired α-(1→3)-linked disaccharide donor

329 in 91% yield, with high stereoselectivity, thanks to neighboring group participation from the acetyl ester of 327. This disaccharide was coupled with acceptor 326 using thiophilic promoter

NIS/TfOH, affording the desired β-(1→3)-linked tetrasaccharide 330 in 71% yield. Selective removal of the Lev group, followed by coupling with rhamnopyranoside donor 332 afforded α-(1→2)-linked pentasaccharide 333. After Lev removal, the pentasaccharide was glycosylated with the conformationally-constrained trichloroacetimidate arabinofuranosyl donor 335 affording hexasaccharide 336 in 91% yield and, gratifyingly, exclusively as the β-anomer. Deprotection of both tetrasaccharide 330 and hexasaccharide 336 proceeded smoothly to afford the two target oligosaccharides.

RG-II side chain C

RG II backbone

Figure 22. Side chain C of RG-II.

The C side chain of pectin, consisting of the disaccharide α-L-Rhap-(1→5)-D-KDOp, was first isolated and characterized in 1985.165 The disaccharide is linked to C3-O of a GalpA residue from the RG-II backbone (see Figure 22). Synthesis of this side chain has not yet been reported. However, the rare eight-carbon sugar α-D-KDOp is found in non-plant glycans, and therefore various synthetic procedures have been developed.165–168

RG-II side chain D

RG-II backbone

Figure 23. Side chain D of RG-II.

RG-II side chain D, consisting of disaccharide β-L-Araf-(1→5)-D-Dhap, was first isolated from pectin plant cell wall by Albersheim and co-workers in 1988.169 As for side chain C, this disaccharide is attached to the C-3 position of a galacturonic acid in the backbone (see Figure 23), and has not been synthesized either. However a procedure to synthesize the rare sugar 3-deoxy-D-lyxo-heptulosaric acid (Dhap) was reported in 1995 by Banaszek.170

RG-II side chain E/F

E RG-II backbone

Figure 24. Side chain E of RG-II.

Several independent studies suggest that a single α-L-Araf residue is linked to O-3 of one of the GalpA backbone residues of RG-II (see Figure 24).147,151,171 However, it has not yet been proven, whether this residue is linked to O-3 or O-2 of GalpA and no synthesis of this linkage has been reported.

(Arabino)galactans and arabinans

Arabinogalactans, galactans and arabinans are the three main types of side chain in the pectin RG-I domain. However, arabinogalactans are also present in many plants as a constituent of hemicellulose and in arabinogalactan proteins.172

Type I AGs Galactans

Type II AGs (three examples of possible structures)

Figure 25. Exemplary structures of arabinogalactans (AGs) and galactans.

Arabinogalactans (AGs) are cell wall polysaccharides having high proportions of galactose and arabinose. They are widely present in the plant kingdom and occur as variable complex structures, whose composition is highly dependent on the plant and even cells from which they originate. They are not found to be well conserved. Arabinogalactans are usually divided into two groups: type I and type II, which are schematically represented in Figure 25.131 Type I AGs are characterized by a linear

β-(1→4)-linked D-Galp chain, which may be substituted with α-(1→3)-linked single L-Araf, short α-

(1→3)- or α-(1→5)-linked arabinan oligosaccharides, or with β-(1→6)-linked galactose residues.97

Type I AGs are found in pectic complexes in seeds, bulbs and leaves and are mostly covalently bound to the RG-I backbone.172 Galactans are simply β-(1→4)-linked D-galactan chains. They have been observed in lengths of up to 40 residues.173 Type II AGs are highly branched polysaccharides composed of a β-(1→3) or (1→6)-linked D-galactan backbone frequently substituted at C-3 and/or C-

6 with β-galactan or arabinogalactan branching, and occasionally carrying C-2 substitution.97 This second type of AG exists in various plant tissues, including coniferous woods, mosses as well as gums, saps and exudates of angiosperms.131 Type II AGs make up the carbohydrate part of arabinogalactan proteins (AGPs) and account for over 90% of the molecular mass of these complexes. Although the structure of arabinogalactans appears to be well established, new structural diversity is found in this class of polysaccharides occasionally. Huisman et al., for example, reported the presence of arabinose residues in the galactan backbone as well as a terminal L-Arap residue in the side chains of type I AG from soybean pectin.174 NMR analysis has clearly demonstrated that type I AGs are covalently bound to rhamnose in RG-I and this is the main type of pectic arabinogalactan.172,175 However, Cordeiro and co-workers analyzed pectin from starfruit and found evidence of both type I and II AGs among the

RG-I side chains.176 The next section will give an overview of the AG fragments prepared by chemical synthesis.

6.1 Type I AGs and galactans

β-(1→4)-linked galactan synthesis is challenging, due to the low accessibility and reactivity of the axially disposed C4-OH. There is a requirement for protecting groups at both C-3 and C-6, which increases steric hindrance around the nucleophile. Until 2016, only four groups had reported syntheses of β-(1→4)-D-galacto-oligosaccharides, of which, three only reached the trisaccharide level.177-178

Lately a small library of oligogalactans including linear tetra-, penta-, hexa-, and heptasaccharides have been synthesized by Clausen and co-workers.179 As commented on by Sakamoto and Ishimura in 2013, the low availability of pure type I AG oligosaccharides has hindered the study of galactonolytic enzymes.180 No synthesis of type I AGs had been reported until early 2017, when

Pfrengle and co-workers produced a set of type I AGs and determined the substrate specificities of three β-(1→4)-endogalactanases.181 The galactans and type I AGs that have so far been synthesized are presented in Table 17.

Table 17. Synthetic β-(1→4)-linked galactans including type I AGs. Structure R Year Reference

177 OR Pr 1984 El-Shenawy & Schuerch

Me 1987 Kováč & Taylor182

H 2001 Lichtenthaler et al.183

183 OR PMP 2001 Lichtenthaler et al. 2 181 (CH2)5NH2 2017 Bartetzko et al.

183 OR PMP 2001 Lichtenthaler et al. 3 H 2016 Andersen et al.179

183 OR PMP 2001 Lichtenthaler et al. 4 H 2016 Andersen et al.179

181 (CH2)5NH2 2017 Bartetzko et al.

179 OR H 2016 Andersen et al. 5

181 OR (CH2)5NH2 2017 Bartetzko et al.

179 OR H 2016 Andersen et al. 4

H 2016 Andersen et al.179

H 2016 Andersen et al.179 OR

H 2016 Andersen et al.179

The first reported synthesis of a linear β-(1→4)-D-galactan trisaccharide was achieved by Schuerch

& El-Shenawy177 and is shown in Scheme 35.

Scheme 35. Synthesis of protected β-(1→4)-D-galactotriose 16 by Schuerch & El-Shenawy.177

HO OBn O BnO BzO OAll 342 AcO OBn AcO OBn 1) Rh(PPh3)3Cl 1) Rh(PPh3)3Cl AcO OBn OBn AcO OBn AgOTres i AcO O iPr NEt, EtOH O Pr2NEt, EtOH 1) Ac2O, py CH2Cl2 BnO O OBn 2 O O O BnO O OBn BnO BnO BnO BzO O BzO TiBr O BzO OH 2) 4 BzO 65% BnO ZnCl ZnO OAll 2) HgO, HgCl2 BzO Br 2) 2, BnO EtOAc, CH2Cl2 BzO OH acetone, H2O OAll acetone, H2O BzO 339 340 341 343 344 78% 89% 85%

AcO OBn 342 HO OH 1) PhNCO AcO OBn O OBn N,N'-diphenylurea AgOTres BnO O O O CH Cl 1) KCN in 95% EtOH HO O OH CH2Cl2 BnO O OBn 2 2 BzO O BnO O OBn HO O BzO OH O 85% 2) Pd/C, H2 HO O 2) HCl (g), CH2Cl2 BnO BzO O BnO acetone, H2O HO O BzO HO 80% 345 Cl 346 BzO 347 OAll 70% HO OPr

The key building block 339 carried an allyl group as an exchangeable protecting group at the anomeric center and an acetyl group for temporarily blocking the 4-position. A benzoyl group at the C-2 position ensured neighboring group participation to form the desired β-linkage. Finally, benzyl groups were used to block the remaining hydroxyl groups. Acceptor 342 was coupled to glycosyl bromide 340 in a silver tresylate (2,2,2-trifluoroethanesulfonate, OTres) catalyzed glycosylation reaction providing the disaccharide 343 in 65% yield. This disaccharide was then converted into the corresponding glycosyl chloride 345, over four steps, with an overall yield of 69%. A second glycosylation of acceptor 342 afforded the trisaccharide 346 in 85% yield, and this could easily be deprotected to afford the target 347. Further selective deacetylation of the di- and trisaccharide did not proceed well, and this strategy did not enable the authors to elongate the chain in a convergent way as they had hoped.

100

To improve the efficiency of this galactotriose synthesis, Kovác & Taylor suggested another approach starting from acceptor 348.182 This was converted to the glycosyl chloride 349 in two steps (Scheme

36). Coupling of acceptor 348 and donor 349 promoted by silver trifluoromethanesulfonate gave the desired disaccharide 350 in 62% yield. The bromoacetyl group was selectively removed with thiourea to give acceptor 351, which was once again glycosylated with glycosyl chloride 349. However, in this case, the glycosylation went slower and the trisaccharide 352 could only be obtained in 35% yield.

Further elongation was thus not feasible in this case either.

Scheme 36. Synthesis of β-(1→4)-D-galactriose by Kovác & Taylor.182

BrAcO OBz AgOTf RO OBz 349, AgOTf HO OBz BrAcO OBz , O CH3NO2 toluene O CH2Cl2 BzO O OBz O + O BzO O OBz BzO OMe BzO OBz O 62% OBz O 35% BzO O OBz BzO BzO BzO OMe Cl OBz O OBz OMe BzO = 348 349 thiourea, MeOH 350: R AcBr 352 OBz 351: R = H (85%) NaOMe, MeOH toluene 82% 1) 1,1,3,3-tetramethylurea BrAcBr HO OH 1,2-dimethoxyethane , O 2) ZnCl2 Cl2HCOMe HO O OH OH O 80% HO O OH OH O HO OMe 353 OH

Lichtenthaler and co-workers published the first synthetic route by which longer β-(1→4)-linked galactans, up to a hexasaccharide, could be produced (Scheme 37).183 An iterative block strategy and easy transformation of the PMP acceptors into thioglycoside donors really did make this a convergent synthesis. Furthermore, the protecting group pattern was judiciously designed to minimize steric hindrance around the unreactive galactosyl-4-OH. Hence, allyl and/or benzyl groups were chosen for

C-3 and C-6 in order to enhance the nucleophilicity of the acceptor. In this case, a pivaloyl ester was chosen for C-2 protection. The acetyl group could then be used for blocking C-4, as mild Zemplén

101 conditions will not remove the Piv ester. MeOTf-mediated glycosylation of thioglycoside building block 354 and acceptor 355 afforded the disaccharide 356 in a good yield (79%). This could then be turned both into a new disaccharide donor 357 (by glycosylation with TMSSPh) or a new disaccharide acceptor 358 (by deacetylation). Two rounds of glycosylation then afforded the hexasaccharide 361 on a gram scale. In comparison with the two previous studies, this strategy afforded remarkably high glycosylation yields, considering the low reactivity of the Galp C-4 position. However, the removal of the protecting groups, especially the allyl group, turned out to be challenging, and a five-step procedure was necessary to reach the target 362.

102

Scheme 37. Synthesis of β-(1→4)-D-hexagalactoside 362 by Lichtenthaler and co-workers.183

AcO OBn OBn OBn MeOTf, DTBMP AcO HO O CH2Cl2 OBn O + O AllO O AllO SPh AllO OPMP 79% OPiv O OPiv OPiv AllO OPMP OPiv 354 355 356 PhSSiMe NaOMe . 3 MeOH BF3 OEt2 85% 92%

AcO OBn HO OBn O O AllO O OBn AllO O OBn OPiv O OPiv O AllO SPh AllO OPMP OPiv OPiv 357 358

AcO OBn O AllO O OBn RO OBn OPiv O OBn MeOTf 357, MeOTf AllO O 2,6-di-tert-butyl-4- O 2,6-di-tert-butyl-4- O AllO O OBn OPiv OBn methylpyridine methylpyridine AllO O OPiv O CH Cl OBn CH2Cl2 OPiv O 2 2 AllO O O OBn 357 + 358 AllO OPiv O 76% OBn 78% OPiv O AllO O AllO O OBn OPiv O OPMP OPiv O AllO AllO OPMP = NaOMe,MeOH 359 R Ac OPiv OPiv 86% 360 R = H 361

1) LiOH, MeOH, 81% 2) DIBAL, [NiCl2(dppp)], THF, Et2O Ac O, DMAP, py 3) ,2 4) H2 Pd/C, AcOH 5) NaOMe, MeOH HO OH 54% (4 steps) O HO O OH OH O OH HO O OH O HO O OH OH O HO O OH OH O HO O OH OH O OPMP 362 HO OH

Clausen and co-workers achieved the synthesis of a set of galactooligosaccharides with a β-(1→4)- linked backbone and the potential for (1→6)-branching. The group conducted a convergent block synthesis, using a disaccharide building block to create linear oligosaccharides of varying lenghts.

TMSOTf-mediated coupling of trifluoroacetimidate donor 363 and acceptor 364 afforded the key pentenyl disaccharide 365 in 83% yield (Scheme 38).179

103

Scheme 38. Synthesis of linear β-(1→4)–D-penta- and heptasaccharides by Clausen and co- workers.179

HO OBn O BnO OBn OBn RO OBn ClAcO OBn 366 O ClAcO OBn TMSOTf O NIS,TESOTf BnO O OBn HO OBn BnO O OBn CH2Cl2 MeCN,CH Cl OPiv O O + 2 2 OBn BnO O OPiv O BnO O O CF3 BnO O BnO O OPiv 83% 85% OPiv O NPh OPiv OPiv BnO OBn OBn 363 364 365 NaOMe, MeOH 367: R = ClAc 91% 368: R = H

ClAcO OBn HO OH O O OBn OH 365 BnO O HO O OPiv O Et NOH, MeOH, THF OH O NIS,TESOTf BnO O OBn 1) 4 HO O OH MeCN, CH2Cl2 2) H2,Pd(OH)2/C, THF, MeOH, H2O OPiv O OH O BnO O OBn HO O OH 83% 78% OPiv O OH O BnO O OBn HO O OH OPiv O OH O 369 BnO OBn 370 HO OH OBn OH

NaOMe, MeOH 93%

HO OBn O BnO O OBn OPiv O BnO O OBn OPiv O BnO O OBn OPiv O BnO O OBn OPiv O 371 BnO OBn OBn

365 NIS,TESOTf MeCN,CH2Cl2 ClAcO OBn 80% HO OH O O BnO O OBn HO O OH OPiv O OH O BnO O OBn HO O OH 1) Et4NOH, MeOH, THF OPiv O OH O BnO O OBn 2) H2,Pd(OH)2/C, THF, MeOH, H2O HO O OH OPiv O OH O BnO O OBn 75% HO O OH OPiv O OH O BnO O OBn HO O OH OPiv O OH O BnO O OBn HO O OH OPiv O OH O BnO OBn HO 372 373 OH OBn OH

NIS-TESOTf-mediated glycosylation of the monosaccharide acceptor 366, with disaccharide 365, afforded the trisaccharide 367. Selective deprotection of the chloroacetyl group, followed by coupling with 365, enabled the synthesis both of a linear penta- and heptasaccharide (369 and 372, respectively)

104 in high yields. Finally, de-esterification by treatment with Et4NOH, followed by hydrogenolysis gave the targets 370 and 373 in 78% and 75% yield, respectively.

In 2017, Pfrengle and co-workers reported the synthesis of a small library of type I AG oligosaccharides through an automated solid phase sequence, using the six monosaccharide building blocks 121, 374–378 shown in Scheme 39. Fmoc-protected galactose building blocks 374 and 375 were required for the synthesis of the β-(1→4)-linked backbone. A temporary Lev protecting group was used at the C-3 position of building block 375, allowing for substitution. A third galactose building block (376) with an Fmoc group in the C-3 position was used to prepare a mixed-linkage galactan (383). Both benzyl ethers and benzoyl esters were used as permanent protecting groups for the remaining positions. Finally, the three L-arabinofuranosides 121, 377 and 378 were used for the side chains.

105

Scheme 39. Automated assembly of type I AGs by Pfrengle and co-workers.181

OBn OBn OBn FmocO FmocO BnO O O O BnO O OBu LevO O OBu FmocO O OBu OBz P OBu OBz P OBu OBz P OBu O O O 374 375 376

OBz SEt FmocO SEt OBz SEt O O O BzO BzO FmocO OBz OBz OBz 121 377 378

Automated solid-phase synthesis

Coupling: TMSOTf, CH2Cl2 (374 , 375, 376) , dioxane NIS, TfOH, CH2Cl2 (121, 377, 378) Et N in DMF Deprotection: 20% 3 (Fmoc) N H , H 2 4 py, AcOH, 2O (Lev) DMAP, Capping: BzCl, py hν, Cleavage: CH2Cl2 Deprotection: NaOMe, THF , Pd/C, H2 EtOAc, MeOH, H2O, AcOH

HO OH HO OH HO OH O O O HO O OH HO O OH HO O OH HO O HO O HO O OH O O OH OH HO O OH HO O HO O O HO O HO O HO O OH HO O NH2 HO O OH HO HO O OH OH 3 HO O O NH HO O OH 379: 15% HO 2 380: 6% 381: 2% OH 3 HO O HO O OH HO O HO O NH2 OH 3 HO OH O HO OH HO OH O O OH O OH OH O HO O O HO O HO O OH HO OH HO O OH HO O HO HO OH OH HO O O O OH O O OH HO O O OH HO O HO O HO O HO O O NH HO O OH O O HO 2 O NH HO O NH2 HO 2 HO OH OH 3 HO 3 OH 3 HO 382: 7% 383: 16% 384: 8%

HO OH HO OH O O HO O OH O O NH2 OH HO O O HO 3 HO O OH O HO O OH OH O O OH O OH O HO O OH OH O HO O OH HO O OH HO O HO O NH2 HO OH 385: 38% 386: 5% OH 3

106

6.2 Type II AGs

Various different type II arabinogalactans have been produced by several groups and the synthesized fragments are presented in Table 18 and Table 19. For the most part, arabinogalactans with a β-(1→6)- galactan backbone have been synthesized (Table 19), but two groups reported the synthesis of some fragments containing a β-(1→3)-galactan backbone, which is mostly found in arabinogalactan proteins (AGPs, Table 18).184,185

Table 18. Synthetic type II AG fragments containing a β-(1→3)-linked backbone. Structure R Year Reference

185 OR Me 2002 Yang & Du

184 OR (CH2)5NH2 2015 Bartetzko et al.

184 (CH2)5NH2 2015 Bartetzko et al.

107

184 (CH2)5NH2 2015 Bartetzko et al.

OR 184 (CH2)5NH2 2015 Bartetzko et al.

184 OR (CH2)5NH2 2015 Bartetzko et al.

184 (CH2)5NH2 2015 Bartetzko et al.

108

184 (CH2)5NH2 2015 Bartetzko et al.

The only solution phase synthesis of β-(1→3)-linked type II arabinogalactan was achieved by Yang

& Du in 2002 and is presented in Scheme 40.185 Regioselective glycosylation of diol galactoside 388 with arabinose donor 387 afforded the desired α-linked disaccharide 389 in 76% yield. Standard protecting group manipulations gave acceptor 391, which could be further selectively coupled with trichloroacetimidate donor 392, affording protected trisaccharide 393 in 94% yield. Deprotection gave the desired trisaccharide 394.

Scheme 40. Synthesis of type-II arabinogalactan trisaccharide 394 by Yang & Du.185

O OH BzO OBz O O O BzO HO BzO OMe OBz OC(NH)CCl3 OBz OH BzO BzO HO 388 392 O O NH3 O OBz OC(NH)CCl3 TMSOTf, CH2Cl2 OBz O TMSOTf, CH2Cl2 OBz O MeOH OH O O 3 R O BzO OBz HO HO OH HO 76% 94% BzO O O 74% O OBz 2 O O R O BzO O HO O R1O AcO HO OMe BzO OMe OH OMe

387 389: R1 = H , R2, R3 = C 393 394 Ac O, py (CH3)2 2 1 2 3 390: R = Ac , R , R = (CH3)2C 90% TFA 391: R1 = Ac , R2 = R3 = H

In 2015, Pfrengle and co-workers published the synthesis of a small library of 14 type II AGPs through an automated solid phase approach using the four monosaccharide building blocks 121, 376, 395 and

396 (Scheme 41).184 Both linear and branched oligomers were synthesized as well as arabinogalactan fragments that had either a β-(1→3)- or a β-(1→6)-galactan backbone.

109

Scheme 41. Examples of the type II AG fragments obtained by Pfrengle and co-workers.184

OBn OLev OFmoc OBz SEt BnO BnO BnO O O O O FmocO O OBu FmocO O OBu BnO O OBu BzO OBz OBz P OBu OBz P OBu OBz P OBu O O O 121 376 395 396

Automated solid-phase synthesis

Coupling: TMSOTf, CH2Cl2 (376 , 395, 396) , dioxane NIS, TfOH, CH2Cl2 (121) Et N in DMF Deprotection: 20% 3 (Fmoc) N H , H 2 4 py, AcOH, 2O (Lev) DMAP, Capping: BzCl, py hν, Cleavage: CH2Cl2 Deprotection: NaOMe, THF , Pd/C, H2 EtOAc, MeOH, H2O, AcOH

OH HO OH HO OH HO OH HO O O OH OH O O O OH O O O NH2 HO OH HO HO OH OH O O OH OH HO 3 O O O O O NH OH HO O 2 HO O HO OH OH OH OH HO 3 OH OH OH HO O O O 397:17% 398: 17% OH HO O HO O HO HO O HO OH HO O O O O O O NH HO O NH2 OH O 2 OH OH 3 O OH 3 HO OH 399: 10% 400: 16%

OH HO OH OH HO HO O O O OH OH OH O O OH O HO OH OH OH OH HO O OH OH HO O OH HO OH O O O HO O HO O O O OH O OH O OH HO OH HO HO OH HO O OH HO HO HO OH HO HO HO O HO O O O O O O O O NH2 NH O O O HO HO O O O 2 O NH HO HO O O 2 OH OH 3 OH HO 3 OH OH OH HO 3 401: 16% 402: 11% 403: 9%

Table 19. Synthetic type II AG fragments containing a β-(1→6)-linked backbone. Structure R Year Reference

184 (CH2)5NH2 2015 Bartetzko et al.

110

H 2002 Ning et al.186

Me 2010 Kaji et al.187 OR

188 (CH2)11CO2Me 1998 Timmers et al.

189 OR C12H25 2000 Du et al.

2003 Ning et al.190

188 (CH2)11CO2Me 1998 Timmers et al.

191 H 2002 Csávás et al. OR

190 C12H25 2003 Ning et al.

192 (CH2)3NH2 2009 Fekete et al.

193 2013 Liang et al. OR

184 (CH2)5NH2 2015 Bartetzko et al.

188 (CH2)11CO2Me 1998 Timmers et al.

190 OR C12H25 2003 Ning et al.

H 2002 Csávás et al.191

194 OR PMP 2003 Zeng et al.

192 (CH2)3NH2 2009 Fekete et al.

111

190 C12H25 2003 Ning et al.

PMP 2003 Zeng et al.194

Me 2001 Gu et al.195

H 2002 Ning et al.186

PMP 2005 Li et al.196

197 (CH2)3NH2 2014 Huang et al.

PMP 2004 Li et al.198

112

PMP 2005 Li et al.196

PMP 2004 Li et al.199

PMP 2005 Li et al.196

The first synthesis of a tetrameric arabinogalactan with a β-(1→6)-linked backbone was achieved in

1998 by van Boom and co-workers, based on 1,2-anhydrosugar building blocks, as shown in Scheme

113

42.188 The group synthesized three different tetrasaccharide analogues, differing in the position of the arabinofuranosyl side chain, which was linked either to the 2-, 2’- or 2’’-position of the trisaccharide backbone. To afford the three targets, an oxidative coupling of galactals was undertaken. This methodology involved a two-step procedure starting with stereoselective epoxidation of a suitably- protected galactal 404 with 3,3-dimethyldioxirane, followed by ZnCl2-catalyzed condensation of the resulting 1,2-anhydrosugar with a galactal acceptor. This glycal approach afforded an unprotected 2- hydroxyl function, ready for glycosylation with an arabinosyl donor in order to introduce the branching.

114

Scheme 42. Example of one of the three tetrameric arabinogalactans containing a β-(1→6)-linked backbone prepared by van Boom and co-workers.188

OBz SEt O O HO OMe BzO OBz 10 121 OR 406 BnO 3,3-dimethyldioxirane NIS, TfOH , BnO OTBDPS OTBDPS O BnO OTBDPS CH2Cl2 acetone ZnCl THF BnO ClCH CH Cl, THF O 2, 2 2 BnO O(CH2)11CO2Me O . BnO O O BnO O(CH2)11CO2Me OBz BnO quant 78% 91% O O OH BzO 404 405 407 OBz

408: R = TBDPS TBAF, THF, 80% 409: R = H

BnO OTBDPS BnO OTBDPS BnO OTBDPS O O BnO OR BnO O BnO BnO O O O O OH 405 BnO O 405 BnO , OH , O ZnCl2 THF BnO ZnCl2 THF BnO O O OHBnO 68% BnO O(CH2)11CO2Me 60% O O OBz BnO CO Me O O(CH2)11 2 OBz O BzO OBz O BzO : = OBz TBAF, THF, 76% 410 R TBDPS 412 411: R = H

HO OH O HO O OH HO 1) TBAF, THF t , O 2) BuOK, CH2Cl2 MeOH HO O , t 3) H2 Pd/C, BuOH, H2O OH HO O 64% HO O(CH2)11CO2Me OH O O HO 413 OH

Kong and co-workers later reported several syntheses of well-defined arabinogalactans carrying different lengths of arabinan side chain.189,194–196,198,199 They started by producing a tetrasaccharide in

2000, and by 2005, they had reached a chain length of up to 20 residues. As one example of their work, the synthesis of octaarabinogalactan 429 is presented in Scheme 43.198 This target consists of a

β-(1→6)-linked pentagalactan carrying arabinofuranosyl and diarabinofuranosyl branches at its 2’’’- and 2’-positions, respectively.

115

Scheme 43. Synthesis of octaarabinogalactan by Kong and co-workers.198

BzO OH BzO OH O O OBz OC(NH)CCl3 BzO OPMP BzO OPMP O OAc OBz BzO OBz BzO OBz OBz BzO 414 BzO OBz 417 O O BzO 387 BzO O OBz TMSOTf, CH Cl TMSOTf, CH Cl O BzO 2 2 O 2 2 BzOBzO 4-methoxyphenol BzO O BzOBzO TMSOTf, CH Cl O 4-methoxyphenol 2 2 BzOBzO O O BzO BzO O 81% 80% BzO O 78% BzO O O BzO OC(NH)CCl3 BzO RO BzO 392 OAcOR O OBz O BzO OPMP O BzO OR CAN, H CH CN : R = PMP β OBz OBz 1) 2O, 3 415 ( ) 1% MeCOCl, MeOH 418: R = Ac BzO CCl CN, K CO , CH Cl α OBz 2) 3 2 3 2 2 416: R = C(NH)CCl3 ( ) CH Cl 419: R = H (79%) 79% 2 2 CAN, H CH CN : R = PMP β 1) 2O, 3 420 ( ) K , α 2) CCl3CN, 2CO3 CH2Cl2 421: R = C(NH)CCl3 ( ) 81% BzO OH O BzO OPMP OBz OR 417 BzO OAc BzO TMSOTf, CH Cl O 2 2 O O 4-methoxyphenol BzO BzO AcO BzO AcO 81% O OC(NH)CCl3 BzO OPMP 422 OBz

0.1% MeCOCl 423: R = Ac MeOH, CH2Cl2 424: R = H (82%)

RO OR OBz OC(NH)CCl3 O O OBz RO O BzO O OBz OBz RO RO O O BzO O O BzO RO O BzOBzO OBz O RO 427 O OR O BzO O O RO O TMSOTf, CH2Cl2 O BzO TMSOTf, CH2Cl2 421+ 424 RO RO RO OBz O OR 75% O BzO O 75% O RO O BzO BzO BzO OBz O RO O BzO O OR O O RO OPMP RO BzO OR O O OR OR BzO OPMP O OBz RO OR

1% MeCOCl 425: R = Ac , 428: R = Bz NH3 MeOH MeOH, CH2Cl2 426: R = H (72%) 429: R = H (84%)

Branching at both the C-2 and C-3 positions is a well-known characteristic for AGs. In order to evaluate differences in immunomodulating activity between C-2 and C-3 branched variants, the same group synthesized a series of model AGs in 2005. To illustrate their method, the retrosynthetic analysis of the arabinogalactan 20-mer is shown in Scheme 44.196 The synthesis of the key tetrasaccharide building blocks 431–433 was achieved from the monosaccharides 387, 417, 422, 434 and 435.

116

Through coupling reactions between the tetrasaccharides, the authors reached arabinogalactans of varying lengths, up to the longest (430), with twenty monosaccharides.

Scheme 44. Retrosynthetic analysis of AG 430 synthesized by Kong and co-workers.196

O HO O HO O HO HO O O O OH HO HO O O HO HO O OH HO HO O HO O HO HO O HO O O HO O OH O O HO HO HO HO OH O HO O HO HO O 2 O O 430 OH HO HO O O HO HO OPMP OH HO

BzO OAc BzO OAc BzO OH O O O BzO O O O BzO BzO BzO BzO BzO BzO BzO BzO O O O BzO O O O O O OBzO OBz BzOBzO OBz BzOBzO O O O O O OBz O BzO BzO BzO BzO BzO OPMP OBz OBz BzO BzO BzO BzO OC(NH)CCl3 OC(NH)CCl3 OBz 431 432 433

BzO OAc OH OH BzO OAc BzO BzO OBz OC(NH)CCl3 O O O O O BzO i BzO AllO S Pr BzO OPMP BzO BzO OBz AcO OC(NH)CCl3 OBz OBz OC(NH)CCl3 434 435 417 387 422

117

Separately, the groups of Yang and Zhang have developed one-pot glycosylation strategies for the synthesis of 3,6-branched tetra- and hexaarabinogalactan. Both groups relied on the same galactopyranosyl thioglycoside diol 436 as a key glycosylating agent, as shown in Scheme 45. Zhang and co-workers, achieved the synthesis of the hexasaccharide 441 from the assembly of two disaccharides and two monosaccharides, in a one-pot procedure involving three glycosylation reactions, with an overall yield of 60%.197

OBz BzO BzO OH OH O BzO O O OBz SPh BzO BzO O BzO O O BzO BzO BzO HO SPh O BzO OBz BzO O OBz BzO O(CH2)3N3 BzO OC(NH)CCl3 OBz 436 437 438 439

BzO OBz HO OH O O BzO O HO O BzO BzO HO HO 1) 437, TMSOTf, CH2Cl2 O O BzO O HO O OH 2) 438, NIS, TfOH, CH2Cl2 BzO BzOBzO NaOMe, MeOH, 88% HO HO 3) 439, NIS, TfOH, CH2Cl2 1) O O O HO SPh O O O O 60% over 3 steps OBz 2) Pd/C, H2, MeOH, 81% OH OBz O BzO BzO O HO HO O O BzO O HO O OBz BzO OH HO BzO BzO HO HO 436 440 O 441 O BzO O(CH2)3N3 HO O(CH2)3NH2 OBz OH

Scheme 45. One-pot synthesis to reach a hexaarabinogalactan by Zhang and co-workers.197

In the same manner and starting from the same key monosaccharide building block 436, Yang and co-workers achieved the synthesis of their tetrasaccharide in a one-pot manner, reaching the target in

52% yield over three glycosylation steps.

118

6.3 Arabinans

linear arabinan

branched arabinan Figure 26. Example of linear and branched arabinan structures.

Arabinans consist of α-(1→5)-linked L-Araf residues, with some degree of α-(1→3) and/or α-(1→2) branching, as shown in Figure 26. In many schematic representations of pectin, the arabinan side chains are shown as being attached directly to the rhamnose units of the RG-I backbone. 6,99,200–202

Very limited evidence for this claim has been reported, however. Most arabinan has been observed as being linked to galactose. A great deal of work has been carried out on synthesizing oligomers of D- arabinofuranosides from bacterial cell walls highlighted by a highly efficient assembly of a 92-mer polysaccharide published in 2017 by Xin-Shan and co-workers.203 Significantly fewer syntheses of L- arabinofuranosides have been reported and they are presented in Table 20.

Table 20. Synthetic α-L-arabinofuranosides. Structure R Year Reference

OR Me 1986 Nepogodiev et al.204

OR Me 1995 Kaneko et al.205

119

OR All 1999 Du et al.206,207 6

The first synthesis of a diarabinofuranoside was carried out by Arnd & Graffi in 1976.208 Ten years later, Nepogodiev and co-workers reported the synthesis of a linear α-(1→3)-linked L-arabinotrioside

204 as shown in Scheme 46. TrClO4-mediated coupling of disaccharide 444 with monosaccharide 445 afforded trisaccharide 446 in 85% yield. This was further deesterified to obtain the target triarabinan

447.

Scheme 46. Synthesis of a linear α-(1→3)-linked L-arabinotrioside by Nepogodiev and co-workers.204

OTr OMe O OBz AcO OAc OH BzO OAc AcO HO 445 OH 4Å MS O O OBz O O , MeCN OBz O TrClO4 CH2Cl2 NaOMe, MeOH O Br + OBz O OH O AcO O O O O BzO O 44% 85% OMe 64% OMe OBz CN O O AcO O BzO O HO O O OBz OH CN AcO HO OAc OH 442 443 444 446 447

In 1995, Kusakabe and co-workers synthesized all three α-L-diarabinofuranosides as well as the trisaccharide shown in Table 20.205,209 Koenigs-Knorr glycosylations promoted by AgOTf afforded the four targets. The synthesis of the trisaccharide also involved simultaneous couplings onto the 3- and 5-positions of the reducing end monosaccharide. A year later, Kong and co-workers reported the synthesis of a linear octaarabinan, bearing an allyl group at the anomeric position, as shown in Scheme

47.206 They were able to perform a regioselective glycosylation between the perbenzoylated trichloroacetimidate donor 387 and the unprotected allyl α-L-arabinofuranoside 448 to obtain partially protected disaccharide 449 in 78% yield. The latter was converted both to the diol disaccharide 450 and the trichloroacetimidate disaccharide donor 451, which were further coupled in a selective manner

120 to form arabinotetrasaccharide 452 in 79% yield. This could then be converted into both a new acceptor 453, holding five free hydroxyl groups and a new donor 454. By coupling these, an octasaccharide was obtained, which, after deprotection, afforded the desired target 455 in 71% over the four last steps.

121

Scheme 47. Synthesis of linear allyl octa-α-(1→5)-L-arabinofuranoside by Kong and co-workers.206

OH OAll O HO OH 448 1) BnOC(NH)CCl3 NH OH OAll OBnOAll TMSOTf O TMSOTf O CH2Cl2 2) NaOMe, MeOH OBz O CCl3 O O O OBz OH OH OBn 78% O 70% O BzO OBz BzO HO 387 OBz 449 OH 450

1) BzCl, py PdCl , NaOAc, HOAc 2) 2 , 3) Cl3CCN, K2CO3 CH2Cl2 70% NH

OBz O CCl3 O O OBz OBz O BzO OBz 451

OBnOAll OBnOAll O O 450, TMSOTf 1) BnOC(NH)CCl3 CH2Cl2 O TMSOTf O 451 OH OBn OBn OBn O O 79% 2) NaOMe, MeOH O 67% O OBz OH OH OBn O O O O OBz OBz 452 OH OH 453 O O BzO HO OBz , OH 1) Pd(OH)2/C, H2 EtOAc 2) Ac2O, py PdCl , NaOAc, HOAc 3) 2 , 4) Cl3CCN, K2CO3 CH2Cl2 65%

OAc O CCl3 O O OAc OAc O O OBz OAc O OAll O OH OBz OBz O O 454 O BzO OH OH OBz O O OH OH O O OH OH O 1) 453, TMSOTf, CH 2Cl2 /C, H , EtOAc O 2) Pd(OH)2 2 OH OH 3) BzCl, py O 4) NaOMe, MeOH 454 O OH OH 71% O 455 O OH OH O O OH OH O HO OH

122

Most recently, Yang and co-workers have synthesized a fully-protected pentaarabinan composed of a linear α-(1→5)-linked arabinotrioside substituted with an α-L-arabinose unit at the 3- and 3’-positions through an elegant four-component one-pot synthesis.210

123

Polysaccharides exclusive to algae

This section focuses on the most common polysaccharides found exclusively in algal cell walls: glucuronans, carrageenans, agarans, alginates, fucans and ulvans. β-(1→3)-D-Xylans have also been isolated from the cell wall of green algae.211,212 So far, only one paper by Kong and Chen reports the synthesis of such structures.213 A few other, more complex and highly branched sulfated heteropolysaccharides (xylofucoglucans, xylofucoglucuronans, etc.) have been identified as minor components in polysaccharide isolates from brown algae. However, their structures have not yet been fully elucidated, and they have never been the subject of chemical synthesis, therefore they will not be discussed further in this section.214

7.1 Glucuronans

Figure 27. Glucuronan structure.

Glucuronans are β-(1→4)-D-polyglucuronic acids, i.e. anionic homopolymers as shown in Figure 27.

These polysaccharides have been extracted from the cell wall of green algae such as Ulva and

Enteromorpha compressa.215,216 The biological activity of the glucuronans has only been considered recently, and no efforts to isolate them were reported before the year 2000.217 In 2009, Redouan and co-workers reported a rapid process to extract glucuronans selectively in large amounts from the green algae Ulva lactura, using enzymatic depolymerization. The process could potentially be used on an industrial scale to produce bioactive glucuronic acid oligosaccharides.218 In 1998, Isogai and co-

124 workers prepared some glucuronan oligosaccharides from cellulose using TEMPO-mediated oxidation, work which could open up a new field of cellulose exploitation.219 No chemical synthesis of a glucuronan longer than a disaccharide has been reported to date, and these disaccharides were produced to optimize the TEMPO-oxidation of cellobiose.

7.2 Carrageenans

The carrageenans are a family of highly-sulfated D-galactans found in the cell walls of certain red seaweeds of the Rhodophyte class. This family of polysaccharides has the ability to form thermo- reversible gels or viscous solutions in the presence of salts. They are organized into ten different types of carrabiose repeating units that differ by their degree of sulfation, and the presence or absence of

3,6-anhydrogalactose residues (see Figure 28).

125

γ-carrageenan β-carrageenan

δ-carrageenan α-carrageenan

Enzymes or KOH

µ-carrageenan κ-carrageenan

ν-carrageenan ι-carrageenan

λ-carrageenan θ-carrageenan

Figure 28. Schematic representation of the ten types of idealized disaccharide repeating units of carrageenans.

Natural carrageenans are non-homologous polysaccharides. This diverse family of glycans shares a common galactan backbone (carrabiose) of alternating 3-linked-β-D-galactopyranose and 4-linked-α-

D-galactopyranose. It has been shown that hot alkaline treatment of the carrageenan having a 4-linked-

α-D-galactopyranose residue sulfated at the 6 position leads to cyclization, forming the 3,6-anhydro rings.220,221 During carrageenan biosynthesis, this transformation is catalyzed by enzymes. No synthesis of this type of algal polysaccharide has been reported to date.

126

7.3 Agarans

Enzymes or KOH

Figure 29. Agarobiose backbone units.

The agarans are another family of sulfated galactans found in red seaweeds. As with carrageenans, agarans have linear chains of alternating 3-linked β-galactopyranosyl residues and 4-linked-α- galactopyranosyl residues. They differ from carrageenans in that the latter are L-Galp. As in carrageenans, these residues can have 3,6-anhydro bridges (Figure 29). In 1991, Lahaye reported that at least eleven different agarobiose structures could be identified depending on the type of red alga.222

The common modifications observed were mainly the presence of sulfate substituents, pyruvate, urinate and methoxy groups. Red seaweeds producing agaran were named agarophytes, while the ones producing carrageenans were called carrageenophytes. However, it has recently been shown that some carrageenophytes also synthesize considerable amounts of agaran or DL-hybrid galactans.223 No organic synthesis of agaran has yet been reported.

7.4 Porphyrans

OMe OMe

Enzymes or KOH

Figure 30. Porphyran repeating unit.

127

The porphyrans are polysaccharides found in the red alga Porphyra umbilicalis, which was characterized by Rees in 1961.224 This family of glycans shares the same backbone as agarans, but are substituted with 6-O-methylation of the D-galactose units as shown in Figure 30.225 Similar to carrageenans, Rees and co-workers showed that suitably linked L-galactose 6-sulphate units in porphyrans can be converted quantitatively by alkaline treatment into 3,6-anhydrogalactose.224 This type of oligosaccharide has not received attention from synthetic chemists yet.

7.5 Alginates

The alginates are a family of polysaccharides present in both brown algae and some gram-negative bacteria. They are composed of two uronic acids: D-mannuronic acid (M) and L-guluronic acid (G, the C5 epimer of M). The monomers are connected by (1→4) glycosidic linkages, with a 1,2-cis configuration. The residues are arranged in either homopolymeric (polymannuronate -MM-, or polyguluronate -GG-) or heteropolymeric (-MG-) segments, as shown in Figure 31.226,19

-GG- -MG- -MM-

Figure 31. Block composition of alginates.

These anionic polysaccharides can be isolated from marine brown algae. They have widespread application in the biomaterial and food industries due to their gelling properties. It has been shown that G-block-rich alginates increase gel strength by forming links with calcium, bridging two

128 antiparallel chains, as described earlier for homogalacturonan (see section 5.1).227 Alginates also offer putative medical benefits, so there is considerable interest in generating chemically well-defined alginate fragments. The assembly of mixed alginate sequences -MG- is particularly challenging, as it requires the synthesis of β-D-mannuronic acid and α-L-guluronic acid, both of which have 1,2-cis linkages. Only one synthesis of mixed sequence alginate oligosaccharides has in fact been reported, by Codée and co-workers in 2015.228 Table 21 gives an overview of all the alginate fragments synthesized to date.

Table 21. Synthetic alginates fragments. Structure R Year Reference

229 (CH2)3NH2 2006 van den Bos et al. OR

2008 Codée et al.230

230 (CH2)3NH2 2008 Codée et al. OR 2 Pr 2012 Walvoort et al.231

230 (CH2)3NH2 2008 Codée et al. OR 3

Pr 2015 Tang et al.92 OR 4

Pr 2012 Walvoort et al.231 OR 6

Pr 2012 Walvoort et al.231 OR 10

232 (CH2)3NH2 2008 Dinkelaar et al. OR

129

233 Pr 2009 Chi et al.

233 Pr 2009 Chi et al. OR 2

228 (CH2)2NHAc 2015 Zhang et al. OR n = 1,2,3

228 (CH2)2NHAc 2015 Zhang et al. OR n = 1,2

228 (CH2)2NHAc 2015 Zhang et al. OR

As mentioned earlier (section 4.5), forming the β-mannosidic linkage is considered one of the biggest challenges in carbohydrate chemistry. Lately, syntheses of –MM- fragments up to dodecamers have been published. In 2006, van der Marel and co-workers discovered the use of 1-thiomannuronate ester donors in the stereocontrolled formation of 1,2-cis mannuronic linkages.229 They proved that the β- selectivity of the mannuronate ester comes from the stereoelectronic influence of the C-5 ester and demonstrated the efficiency of this strategy in the generation of longer alginate oligosaccharides (up to a pentasaccharide) using a block coupling strategy (Scheme 48).230 Thioglycosyl donors were pre- activated prior to coupling, and it was shown that judicious choice of promoter was crucial in ensuring high yields as the oligomers grew in length. Excellent β-selectivity was maintained in all glycosylation reactions and was not compromised by the size of the coupling partners.

130

Scheme 48. Assembly of mannuronate ester oligomers (-MM-) by van der Marel and co-workers.230

MeO2C OBn - MeO2C OBn HO O SPh 1) benzyl 2 (hydroxymethylbenzoate) RO O BnO MeO2C OBn O Ph2SO, TTBP, CH2Cl2 BnO 460 MeO C OBn O 2 MeO2C OBn RO O O BnO CO2H RO O , BnO SPh X 2) H2 Pd/C, NH4OAc, EtOAc, MeOH Ph2SO, TTBP, CH2Cl2 BnO

α:β 456: X = β-SPh, R = Lev 458: R = Lev (66%) 461: R = Lev (65%, = 1:10) α α:β 457: X = -SPh, R = Bn 459: R = Bn (65%) 462: R = Bn (65%, = 1:10)

N N MeO C OBn 3 MeO2C OBn MeO C OBn 3 2 O 2 HO O HO O O BnO O BnO BnO O

463 464

a) Ph2SO, TTBP, CH2Cl2 α:β 28% ( = 1:10) H2NNH2 MeO C OBn N OBn N 2 MeO2C OBn MeO C OBn 3 py, HOAc MeO2C MeO2C OBn MeO C OBn 3 or O O 2 O 2 + LevO O O HO O O O 461 463 BnO O O O BnO BnO 78% BnO BnO BnO O b) BSP, Tf2O, TTBP, CH2Cl2 α:β 74% ( = 1:10) 465 466

BSP, Tf TTBP, CH Cl OBn N 2O, 2 2 MeO2C MeO2C OBn MeO C OBn 3 462 + 464 O O 2 BnO O O O 67% α:β = 1:10) BnO BnO O ( 2BnO

467

BSP, Tf2O, TTBP, CH2Cl2 MeO C OBn MeO C OBn N3 2 2 MeO2C OBn 462 + 466 O O BnO O O O 69% α:β = 1:10) BnO BnO BnO O ( 3 468

MeO C OBn OBn N HO C OH OH H N 2 MeO2C MeO C OBn 3 1) KOH, THF, H2O 2 HO2C HO C OBn 2 O O 2 O O 2 RO O O O HO O O O BnO BnO BnO O HO HO HO O H , Pd/C, MeOH, HOAc n 2) 2 n t ( BuOH, H2O) 464: n = 0, R = H 469: n = 0 (59%) 466: n = 1, R = H 470: n = 1 (86%) 467: n = 2, R = Bn 471: n = 2 (65%) 468: n = 3, R = Bn 472: n = 3 (76%)

Mannuronic acid donors were also used by Codée and co-workers in an automated solid-phase synthesis, to build alginates up to a dodecasaccharide. Trifluoroacetimidate donors were used as building blocks, since pre-activation of donors was not possible on their solid phase synthesizer and because the nature of the linker prohibited the use of soft electrophiles required for the activation of

131 thioglycosides. The stereodirecting effect of the C-5 carboxylate again provided the desired (1→2)- cis glycosides with excellent stereoselectivity.

The synthesis of short L-guluronic acid oligomers (-GG- blocks up to a tetrasaccharide) has been achieved by only two groups, Hung and co-workers233 and van der Marel, Codée and co-workers.232

As L-gulose and L-guluronic acid are rare sugars, they are not commercially available and need to be synthesized prior to use, as shown in Scheme 49.

Scheme 49. Synthesis of L-gulose.232,233

OH OH O O

HO OH L-ascorbic acid 473

Hung and co-workers

1) H2SO4 dimethoxypropane OH acetone O OH OH OH 2) DIBAL-H O HO O O 20% TFA in H O O O toluene OH 2 OH

84% quant. OH HO OH O O L-gulonic acid L-gulose γ-lactone 475 476 474 van der Marel, Codée and co-workers 100 g scale

In contrast to D-mannopyranose, L-gulopyranose has an intrinsic preference for the formation of 1,2- cis-glycosidic bonds without the need to incorporate stereodirecting groups. Furthermore, Codée and co-workers demonstrated that L-guluronic acid donors are less stereoselective in glycosylation reactions than L-gulose donors.232 Therefore, a post-glycosylation oxidation strategy has been the preferred approach for the synthesis of -GG- block oligosaccharides.

132

After screening the efficiency of several gulose donors, Codée and co-workers found that a dehydrative glycosylation234 with hemiacetal 479 gave the best yield, and α-selectivity, in the glycosylation step. They achieved the synthesis of a guluronic acid trisaccharide, as shown in Scheme

50.232

Scheme 50. Synthesis of guluronic acid trisaccharide (-GGG-) by Codée and co-workers.232

O OBn O N3 OBn TBSO OH 480 O O OBn OBn N N OH OBn Ph2SO, TTBP, Tf2O O 3 O 3 HO OBn OBn O SPh O R CH2Cl2 TBSO TBAF, THF R2O OH OBn O O O α OH t O 84% :β = 10:1) OBn OBn Bu Si ( O 87% O OBn OBn 477 tBu O R1O 3 tBu Si O OR tBu NIS, TFA, CH2Cl2 478: R = SPh 95% : = 481 TBSCl, imidazole, DMF 482: R1, R2 = R3 = H 479 R OH 78% 483: R1, R2 = TBS, R3 = H

479, Ph2SO, TTBP, Tf2O, CH2Cl2 then 483 42% only α

O O OBn OH O N3 HOOC O NH2 OBn OH R1O O O OBn OH t O HOOC O 1) BuOH, H2O, TEMPO, BAIB OBn OH R1O O O 2) tBuOH, Pd/C, H OBn OH 2 O HOOC O OBn OH R2O 85% OR2 OH

486 484: R1, TBS, R2 = Si(tBu) TBAF, THF 2 83% 485: R1, R2 = H

Hung and co-workers employed 1,6-anhydrogulose synthons to lock the C4-OH in a more accessible environment and increase the reactivity of the nucleophile.233

As mentioned above, only one synthesis of alginate sequences containing both M and G residues has been reported, by Codée and co-workers in 2015.228 They decided on an approach employing GM building blocks, thus featuring a mannuronic acid donor, and pre-glycosylation oxidation in order to

133 minimize functional-group manipulation at a late stage of the synthesis. The preparation of the building blocks required for the assembly of the longer oligosaccharides is presented in Scheme 51.

Glycosylation between the silylene-protected gulose imidate donor 487 and mannuronic acid acceptor

488 or 463 led to GM dimers 489 and 495, respectively, in good yields and with high stereoselectivities. A desilylation/oxidation/methyl ester formation sequence afforded GM acceptor building blocks 490 and 496. The trifluoroacetimidate GM donor 493 was then obtained from 490 in an 83% yield over three steps.

Scheme 51. GM donor and acceptor building blocks for the synthesis of mixed alginate oligosaccharides by Codée and co-workers.228

1) HF·py, THF MeO2C OBn O CO2Me TEMPO, CO Me HO 2) PhI(OAc)2 2 OBn t NPh BnO BuOH, THF, H2O OBn O STol O STol 488 STol OBn MeI, K CO OBn O 3) 2 3 OBn O CF TMSOTf, CH2Cl2 DMF OBn 3 O O O OBn OBn O 91% O MeO C O tBu Si OBn 83% 2 OBn O tBu tBu Si O OR tBu 487 489 LevOH, EDCI, DMAP 490: R = H , CH2Cl2 92% 491: R = Lev

NIS, TFA, CH2Cl2 91%

MeO2C OBn O N3 O OBn O MeO C O OBn BnO 2 OBn MeO C O OR 2 OBn OH OLev 494

F3CC(=NPh)Cl, K2CO3 492: R = H acetone, 98% 493: R = C(=NPh)CF3

1) HF·py, THF N OBn 3 TEMPO, MeO2C 2) PhI(OAc)2 O t THF, H HO O N3 BuOH, 2O NPh BnO MeO2C OBn N3 3) MeI, K2CO3 OBn O 463 O MeO2C OBn O O DMF O O CF3 OBn BnO O O OBn O OBn BnO O TMSOTf OBn MeO C O O O 84% 2 OBn tBu Si CH Cl 2 2 tBu O tBu ο ο Si OH -78 C to -20 C tBu 487 495 496 58%

134

With the GM donors and acceptors in hand, the assembly of larger -GM- oligomers could proceed.

This was successfully achieved for structures ranging from the GMG trisaccharide to a GMGMGMG heptasaccharide (Scheme 52).

135

Scheme 52. Assembly of -GM- oligomers up to heptasaccharide by Codée and co-workers.228

GMG GMGMG

O N3 O N3 OBn OBn MeO2C O MeO2C O OBn OBn TBSOTf OBn 493 MeO2C MeO2C OBn O O CH Cl O O TBSOTf, CH Cl O 2 2 O 2 2 OBn BnO 493 + 494 OBn BnO MeO2C O 2 84% MeO2C O 31% OBn OBn α:β > 20:1 O OR : = , : = , 497 R Lev N2H4 H2O 499 R Lev N2H4 H2O : : 498 R = H AcOH, py, 89% R 500 R = H AcOH, py, 98%

493 TBSOTf, CH2Cl2 34%

, O N O NHAc 1) N2H4 H2O, AcOH, py 3 OH , OBn LiOH, H2O2 H2O, THF MeO C O NaOOC O 2) 2 OH , t OBn NaOOC OH 3) Pd/C, H2 BuOH, THF, H2O MeO2C OBn O O , O O O 4) Ac2O, NaHCO3 THF, H2O O OH HO OBn BnO MeO C NaOOC O 3 2 O 3 OH 60% OBn O 502 O 501

H GMGMGMG Lev GMGMGMG

GMGM GMGMGM

N3 N3 MeO2C OBn MeO2C OBn O O TBSOTf O O 493 O O BnO BnO CH2Cl2 OBn TBSOTf, CH2Cl2 OBn MeO2C O 2 MeO2C O 3 493 + 496 OBn OBn 30% 26% α:β > 20:1 O O

R Lev 505 : = , 503 R Lev N2H4 H2O 504: R = H AcOH, py, 78%

GMGM GMGMGM CO2Me CO2Me OBn OBn O R O R OBn OBn O TBSOTf O 490 OBn OBn MeO C O CH2Cl2 TBSOTf, CH2Cl2 2 493 + 490 MeO2C O OBn OBn MeO2C OBn 91% MeO2C OBn 73% O O α O O α:β > 20:1 O O :β > 20:1 OBn BnO 2 OBn BnO MeO2C O MeO2C O OBn OBn O OLev Lev α NIS, TFA, CH Cl : R = α-STol NIS, TFA, CH Cl 507: R = -STol 1) 2 2 , 505 1) 2 2 , F K CO acetone = F K CO acetone 508: R = 2) 3CC(=NPh)Cl, 2 3 506: R O(C=NPh)CF3 (92%) 2) 3CC(=NPh)Cl, 2 3 O(C=NPh)CF3 (80%)

Surprisingly, the nature of the reducing end anomeric center of the GM acceptor had a tremendous effect on the efficiency of glycosylation by influencing the conformation of the acceptor. The rigid

136

GM acceptor 496, bearing a β-linked anomeric azidopropyl spacer, led to lower glycosylation yields

(26–34%), whereas the conformationally-flexible GM-α-STol acceptor 490 gave much better yields

(73–91%).235 The GMGGMG hexasaccharide was synthesized using a [2+3+1] approach with mannuronic acid donors, as summarized in Error! Reference source not found..

Scheme 53. Convergent assembly of GMGGMG hexasaccharide by Codée and co-workers.228

GM donor: G acceptor: GGM 493 GMGGM 494 reducing end GMGGM donor GMGGMG acceptor 87% 43%

All the oligomers were deprotected by removal of the levulinoyl groups, saponification of the methyl esters, debenzylation plus azide reduction and finally acetylation of the spacer amine group as shown for heptamer 502 in Scheme 52.

7.6 Fucans

In 1913, Kylin reported the first extraction of “fucoidin” from marine brown algae.236 Most of the fucose containing sulfated polysaccharides (FCSPs) found in brown algae are fucoidans, which are heteropolymers of sulfated fucans. These polysaccharides can account for more than 40% of the dry weight of isolated cell walls in brown algae.237 Fucoidans have been found to have two types of homofucose backbone chains as shown in Figure 32. Type I consists of (1→3)-linked α-L- fucopyranoside residues, while type II is made up of alternating α-(1→3)- and α-(1→4)-L- fucopyranoside residues.238 The complexity and heterogeneity of fucoidans is due to the sulfation pattern, which differs depending both on the type of alga as well as between cell types. Sulfation can

137 occur at the O-2, O-3, and/or O-4 position of the fucose residues. Acetylation has also been observed at various positions of the backbone, as has the presence of other sugar residues such as α-D- glucuronopyranosyl or α-L-fucopyranosyl.237

Type I Type II

Figure 32. The two types of homofucose backbone chains found in brown seaweed fucoidans with examples of sulfation patterns.

It has been known for some time that fucoidans act as modulators of coagulation.239 But how their sulfation pattern and conformation influence biological activity has not been studied. Early in the 21st

Century, synthetic chemists began showing interest in these targets, aiming to synthesize certain well- defined oligofucans and help elucidate their structure-activity relationship. An overview of the type I and type II fucan and fucoidan fragments synthesized are shown in Table 22 and Table 23, respectively.

Table 22. Synthetic type I fucan/fucoidan fragments. Structure R Year Reference

OR Pr 2003 Gerbst et al.240

Pr 2011 Krylov et al.241

138

OR Pr 2003 Gerbst et al.240

OR Pr 2016 Vinnitskiy et al.242

OR 243 C8H17 2004 Hua et al.

2015 Kasai et al.244 2

Pr 2006 Ustyuzhanina et al.238

241 2011 Krylov et al.

139

OR Pr 2006 Ustyuzhanina et al.238

OR 245 C8H17 2010 Zong et al.

244 2 2015 Kasai et al.

OR Pr 2006 Ustyuzhanina et al.238

245 2 C8H17 2010 Zong et al.

OR 244 C8H17 2015 Kasai et al.

OR Pr 2011 Krylov et al.241

OR 243 C8H17 2004 Hua et al.

244 2 2015 Kasai et al.

OR 245 C8H17 2010 Zong et al.

140

OR 245 C8H17 2010 Zong et al.

OR Pr 2006 Ustyuzhanina et al.238

2011 Krylov et al.241 4

OR Pr 2006 Ustyuzhanina et al.238

245 4 C8H17 2010 Zong et al.

OR Pr 2006 Ustyuzhanina et al.238

OR 245 C8H17 2010 Zong et al.

OR Pr 2011 Krylov et al.241

141

OR Pr 2006 Ustyuzhanina et al.238

245 6 C8H17 2010 Zong et al.

OR Pr 2011 Krylov et al.241

Table 23. Synthetic type II fucan/fucoidan fragments. Structure R Year Reference

241 OR Pr 2011 Krylov et al.

142

241 OR Pr 2011 Krylov et al.

238 OR Pr 2006 Ustyuzhanina et al.

2011 Krylov et al.241

246 C8H17 2014 Arafuka et al.

244 2015 Kasai et al.

238 OR Pr 2006 Ustyuzhanina et al.

246 OR C8H17 2014 Arafuka et al.

2015 Kasai et al.244

246 OR C8H17 2014 Arafuka et al.

2015 Kasai et al.244

246 OR C8H17 2014 Arafuka et al.

2015 Kasai et al.244

241 OR Pr 2011 Krylov et al.

246 C8H17 2014 Arafuka et al.

2015 Kasai et al.244

238 OR Pr 2006 Ustyuzhanina et al.

241 2 2011 Krylov et al.

238 OR Pr 2006 Ustyuzhanina et al.

143

241 OR Pr 2011 Krylov et al.

In 2003, Nifantiev and co-workers reported the first synthesis of the two type I fucoidan trisaccharides

515 and 516 (Scheme 54).240 Both fucotriosides were prepared by use of the readily accessible disaccharide acceptor 509.247,248 Regio- and stereoselective glycosylation with fucosyl bromide donor

510, promoted by mercury salts afforded the protected α-linked trisaccharide 511 in 65% yield.

Saponification and subsequent catalytic hydrogenolysis provided fucotrioside target 515 in a good yield overall. Protecting group manipulations and sulfation gave trisaccharide 514. The latter was deprotected to afford the sulfated fucotrioside target 516.

Scheme 54. First type I fucan and fucoidan synthesized by Nifantiev and co-workers.240

Br OAll OAll OAll Me O OBn Me O Me O Me O OBz OBn OBn OBn OAll BzO O O 510 HO HO . O NaO3SO , Bu SnO, toluene SO3 py Me O Hg(CN)2 HgBr2 Me O 2 Me O OBn OBn then BzCl OBn DMF Me O CH2Cl2 OBn O O O HO HO HO O 62% NaO3SO 65% 80% Me O OBn Me O Me O OBn OBn Me O OBn OH OR OBz HO RO HO OBz NaO3SO : = 509 NaOMe, MeOH 511 R Bz 513 514 512: R = H (90%) , , H2 Pd/C, MeOH 1) H2 Pd/C, MeOH 2) NaOMe, MeOH 75% 78%

OPr OPr Me O Me O OH OH O O HO NaO3SO Me O Me O OH OH O O HO NaO3SO Me O Me O OH OH OH OH HO NaO3SO 515 516

144

By applying a stepwise elongation strategy with thioglycoside donors, Du and co-workers subsequently reported the synthesis of a fully sulfated tetrasaccharide with a β-configured octyl group at the reducing end.243 In 2006, Nifantiev and co-workers developed a method using [2+2], [2+4] and

[4+4] coupling reactions to produce oligofucoidans of both type I and type II, employing trichloroacetimidates as donors.238 Inspired by these syntheses and in order also to generate type I oligofucoidans of uneven length, Li and co-workers decided to apply a [2+2] coupling strategy to produce the tetrasaccharide, followed by an iterative assembly of the remaining targets, as shown in

Scheme 55.245 They used 4-O-benzoyl protecting groups, claiming this improved the α- stereoselectivity of the glycosylation steps through remote participation, in combination with 3-O- acetyl and 2-O-benzyl groups. This way, they could produce two different sulfation patterns from each oligofucoidan precursor.

Scheme 55. Retrosynthetic analysis of sulfated tetra- to octafucosides by Li and co-workers.245

OC8H17 OC8H17 OC8H17 OC8H17

Me O OR1 Me O OBn Me O OBn Me O OBn O O O O R2O BzO BzO BzO Me O Me O Me O Me O OR1 OBn OBn OBn Me O SEt OBn O O O OH R2O n BzO BzO BzO OAc n 520 BzO iterative 2 + 2 Me O 1 Me O Me O 522 OR OBn strategy OBn assembling + O O O strategy R2O BzO BzO Me O SEt Me O OR1 Me O OBn Me O OBn OBn OR3 OAc OH O R2O 517a-e n = 1-5 BzO 518b-e n = 2-5 BzO 518a BzO Me O OBn + 1 2 3 OAc R = SO3Na, R = R = H BzO 521 OC(NH)CCl3 1 2 3 or R = H, R = R = SO3Na Me O OBn OAc BzO 519

Per-O-sulfation of organic compounds containing many hydroxy groups can require lengthy reaction times (of up to several days) and high temperatures, as well as substantial excess of the sulfating

145 reagent. Even so, one may still produce mixtures of partially sulfated products. This happened to

Nifantiev and co-workers as they tried to synthesize a fully sulfated tetrafucoside. As a result, they developed an improved TfOH-promoted per-O-sulfation protocol with sulfur trioxide/pyridine, enabling the reaction to take place at 0 °C and over a shorter period of time.249 They then adapted this method for use in their previously developed block strategy, synthesizing per-O-sulfated type I and II oligofucoidans from two to sixteen fucose residues. To their surprise, when using TfOH to promote the sulfation, they observed an unusual rearrangement of the reducing pyranose residue into the corresponding furanose. As this pyranoside-into-furanoside rearrangement is of great interest in the synthesis of furanoside containing oligosaccharide targets, the group later explored its mechanism and optimized this reaction.250 To avoid the rearrangement, they exchanged TfOH for TFA and this produced the desired targets without by-products.

In 2014 and 2015, Toshima and co-workers designed a series of type I and type II fucoidan tetrasaccharides with different sulfation patterns.246,251 An orthogonal deprotection strategy utilizing the common key intermediates 528 and 531 (Scheme 56), carrying three types of protecting groups

(Bn, Bz and PMB) enabled an effective synthesis of the eight targets. The synthesis of the tetrafucoside precursor 528 for the type II fragments is presented in Scheme 56.

146

Scheme 56. Key tetrasaccharides precursor 528 and 531 by Toshima and co-workers.246

Me O S OBn OPMB OC(NH)CCl3 HO 1) NIS, Sc(OTf)3 OC H Me O , Me O 8 17 523 , H2O, MeCN OBn 1-octanol OBn Et O S 1) Yb(OTf) 3 Yb(OTf)3 2 Me O 82% OPMB CH Cl 89% OPMB + OBn O 2 2, O 99% OPMB O 2) DBU, CCl3CN Me O thiourea, 2,6-lutidine Me O 2) OBn CH2Cl2 OBn OC(NH)CCl3 DMF, 94% Me O Me O 76% OClAc OH OBn OBn BzO BzO OClAc OClAc BzO BzO 524 525 526 527

H Me O OC8 17 Me O OC8H17 OBn OC(NH)CCl3 OBn 1) 527, TMSOTf, Et2O Me O OPMB OBn O thiourea, 2,6-lutidine O BzO 2) OPMB DMF BzO Me O Me O 529 OBn OBn O 81% (2 steps) O BzO BzO Me O Me O OBn OBn S Me O O OPMB OBn BzO O OH BzO Me O Me O 528 OBn 531 OBn 530 OPMB OH BzO BzO (type II precursor) (type I precursor)

Chemoselective glycosylation of 523 with 524, promoted by Yb(OTf)3, provided the desired α-

(1→4)-linked disaccharide 525 with very good stereoselectivity. The latter could be converted into a trichloroacetimidate donor 526 and a new disaccharide acceptor 527 in two steps. Condensation of

527 with 1-octanol using Yb(OTf)3 afforded the corresponding octyl fucoside with surprisingly high

β-stereoselectivity. TMSOTf-promoted coupling of the two difucosides, followed by deprotection of the ClAc group provided the key intermediate 528 as a single isomer in 81% yield over two steps. For the five type II fucoidan targets, the deprotection and sulfation steps are presented in Scheme 57.

147

Scheme 57. Deprotection and sulfation of 528 to afford five type II tetrafucoidans 532–536.246

529

/C, H 1) Pd(OH)2/C, H2 1) Pd(OH)2/C, H2 1) NaOMe, MeOH 1) NaOMe, MeOH 1) Pd(OH)2 2 MeOH, AcOEt MeOH, AcOEt MeOH, AcOEt 80% 80% . , 88% 88% 98% 2) DDQ, CH2Cl2 2) SO3 NEt3 DMF NaOMe, MeOH . , 2) NaOMe, MeOH SO3 NEt3 DMF phosphate buffer then 3 M aq. NaOH 97% 2) 2) 97% then aq. NaOH 80% quant. . . SO NEt , DMF 97% SO NEt , DMF /C, H 3) 3 3 3) 3 3 3) Pd(OH)2 2 then 3 M aq. NaOH then 3 M aq. NaOH MeOH, H O 2 81% 87% 84%

/C, H 4) Pd(OH)2 2 MeOH, H O 2 99%

Me O OC8H17 Me O OC8H17 Me O OC8H17 Me O OC8H17 Me O OC8H17 OSO3Na OSO3Na OH OH OH OSO Na OSO Na OSO Na OH OH O 3 O 3 O 3 O O Me O Me O Me O Me O Me O OSO3Na OSO3Na OH OH OH O O O O O NaO3SO HO NaO3SO NaO3SO HO Me O Me O Me O Me O Me O OSO3Na OSO3Na OH OH OH OSO Na OSO Na OSO Na OH OH O 3 O 3 O 3 O O

Me O Me O Me O Me O Me O OSO3Na OSO3Na OH OH OH OSO3Na OSO3Na OSO3Na OSO3Na OH NaO3SO HO NaO3SO NaO3SO HO

533 534 535 536 537

A recent discovery showed that the fucoidan from the brown seaweed Chordaria flagelliformis contains an α-(1→3)-linked polyfucopyranoside backbone with one third of the fucopyranosyl residues bearing α-(1→2)-linked D-glucuronic acid substituents, and with half of the uronic acids carrying α-(1→4)-L-fucofuranosides.252 In 2016, Nifantiev and co-workers achieved the synthesis of a selectively O-sulfated branched pentasaccharide of this type (548) as shown in Scheme 58.242

148

Scheme 58. Synthesis of two branched type I fucoidan pentasaccharides by Nifantiev and co- workers.242

Me OBn COOMe OBn O O TBSO OAll OC(NH)CCl3 BnO BnO OC(NH)CCl3 Me O OBz OAll OAll OBn 539 O 542 BzO Me O OBn Me O OBn TFA (90%), CH2Cl2 TMSOTf, CH2Cl2 Me O TMSOTf, CH2Cl2 O O O BzO BzO 96% 87% OClAc 79% Me O Me O BzO O COOMe OPMB OH BnO OR OClAc OClAc BzO BzO OBn α:β = 537 538 540: R = TBS ( 4:1) HF, MeCN, 77% α 541: R = H ( only)

OAll OC(NH)CCl3 OAll OPr Me O Me O OBn OBn Me O OBn Me O OH O OClAc O O BzO ClAcO BzO HO

545 H , Pd/C Me O O Me O O 1) 2 Me O O THF, EtOAc, EtOH OR TMSOTf, CH Cl O NaOH 2 M MeOH O BzO O COOMe 2 2 BzO O COOMe 2) (aq.), HO O COONa BnO O BnO O HO O 82% OBn 86% Me O OBn OBn Me O OBn OH OClAc OH Me OBn ClAcO Me OBn HO Me OH BnO BnO HO O O O 546 547 OBz OBz OH

543: R = ClAc thiourea, MeOH, 80% LiOH 2 M THF 544: R = H 1) (aq.), 2) Bu4NOH, THF SO ·py, DMF 3) , 3 4) H2 Pd/C, THF, EtOAc, EtOH

76%

OPr

Me O OH O NaO3SO Me O O O NaO3SO O COONa HO O Me O OH OH OSO3Na NaO3SO Me OH HO O

548 OSO3Na

The assembly of both pentasaccharide targets 547 and 548 started with glycosylation between difucopyranoside acceptor 538 and glucuronate donor 539, affording the trisaccharide 540 as a 4:1

α:β-mixture. The pure α-isomer 540 could be isolated after removal of the TBS group. This acceptor was then glycosylated with fucofuranosyl donor 542 to give a mixture of α- and β-linked tetrasaccharides in a ratio of 10:1. The desired α-anomer (isolated in 79% yield) was deprotected and

149 subsequently coupled with fucopyranoside donor 545, giving the core pentasaccharide 546 in 86% yield. The target 547 was obtained after removal of all protecting groups. Alternatively, saponification of 546, followed by O-sulfation and hydrogenolysis afforded the partially sulfated target 548.

7.7 Ulvans

Rare sugar structure:

O A = HO Type = L-iduronic acid HO HOOC OH OH

Type B

Figure 33. The two major types of disaccharide repeating units in ulvans.

Ulvans are water-soluble sulfated heteropolysaccharides, which have been extracted from the cell walls of green marine algae. They represent between 8 and 29% of the dry weight of the cell wall.253,254

A review dealing with the structure and functional properties of ulvans was written by Lahaye and

Robic in 2007.254 The determination of the sugar sequences in ulvans represents a big challenge since the polysaccharide composition varies widely and is dependent on algal source and ecophysiological variation.254 However, acid hydrolysis and / or enzymatic degradation of cell walls from different green seaweeds has shown the presence of L-rhamnose, L-xylose, D-glucuronic and L-iduronic acid.

These sugars are mostly distributed in disaccharide repeating units. Among these, the two most frequent are ulvanobiuronic acid 3-sulfate Type A (β-D-GlcpA-(1→4)-α-L-Rhap3S-(1→4)) and Type

B (α-L-IdopA-(1→4)-α-L-Rhap3S-(1→4)), represented in Figure 33. Other less frequently occurring units containing glucuronic acid on O-2 of rhamnose 3-sulfate or partially sulfated xylose replacing

150 the uronic acid have been found.255 No chemical synthesis of this type of oligosaccharide has been undertaken to date.

Conclusion

Plant cell wall polysaccharides include some of the most complex macromolecules found in nature.

For this reason, synthetic chemists have focused their attention on synthesizing well-defined oligosaccharide fragments, as set out in this review. Not only are these targets extremely valuable in studying protein-carbohydrate interactions, but they also help advance the development of new methodologies in carbohydrate chemistry. Interest in this area is increasing, as witnessed by the many contributions over the past twenty years. The more recent progress in automated synthesis holds particular promise, as it is well suited to producing libraries of closely-related fragments of larger polysaccharides. There is clearly no lack of unexplored structures to serve as inspiration for synthetic chemists, especially within the marine polysaccharides. As this review makes clear, there is no consensus as to the types of aglycone chosen for the targets: everything from reducing sugars over simple alkyl groups to linkers predisposed to immobilization and / or conjugation is represented in the review. More homogeneity in this choice, along with the establishment of a central repository for synthetic oligosaccharides, represent areas, where we believe there is room for benefiting future research. Nonetheless, we remain convinced that synthetic glycans will continue to provide breakthroughs in the understanding of plant cell wall biology and physiology.

Author Information

151

Corresponding Author

E-mail: [email protected]

ORCID

Mads H. Clausen: 0000-0001-9649-1729

Notes

The authors declare no competing financial interest.

Biographies

Christine Kinnaert graduated from the Ecole Nationale des Industries Chimiques. Her thesis research was carried out at Glaxo Smith-Kline and encompassed conformational analysis of polysaccharides.

She obtained her Ph. D. from the group of Prof. Mads H. Clausen at the Technical University of

Denmark in 2017. The subject of her thesis research was chemical synthesis of oligosaccharides related to algal cell wall glycans. Christine is currently pursuing postdoctoral research at DTU.

Mathilde Daugaard obtained her M. Sc. (2011) and Ph. D. (2016) from the Technical University of

Denmark under the supervision of Prof. Thomas E. Nielsen and Prof. Mads H. Clausen, respectively.

The Ph. D. thesis research involved carbohydrate synthesis, more specifically preparation of α-L- arabinans from rhamnogalacturonan I. She is currently employed by the Danish biopharmaceutical company Nuevolution.

Faranak Nami obtained her M. Sc. in 2012 from Aarhus University under the supervision of Associate

Professor Henrik H. Jensen. The work centered on the synthesis of ibogaine analogues inducing an

152 inward conformation of the human serotonin transporter. Her Ph. D. was obtained in 2016 from the

Technical University of Denmark under the supervision of Professor Mads H. Clausen. Faranak worked on oligosaccharide synthesis with the aim of carrying out enzymatic studies.

Mads H. Clausen obtained his Ph.D. from the Technical University of Denmark in 2002 under the supervision of Prof. Robert Madsen. He was a postdoctoral fellow at Harvard University from 2002–

2004 in the group of Prof. Andrew G. Myers, where he worked on total synthesis of enediyne antibiotics. In 2004, he started his independent career at the Department of Chemistry, Technical

University of Denmark, where he has been Professor of Chemical Biology since 2014. His research interests center on chemical biology and bioorganic chemistry and include medicinal chemistry, oligosaccharide and lipid synthesis and developing methods for library synthesis.

Acknowledgments

We are highly grateful to Ms. Adeline Berger and Dr. Helen North for providing the image used for

Figure 3. We acknowledge financial support from the Danish Council for Independent Research “A biology-driven approach for understanding enzymatic degradation of complex polysaccharide systems” (Grant Case no.: 107279), the Carlsberg Foundation, the Danish Strategic Research Council

(GlycAct and SET4Future projects), the Villum Foundation (PLANET project) and the Novo Nordisk

Foundation (Biotechnology-based Synthesis and Production Research).

153

References

(1) Ivakov, A.; Persson, S. Plant Cell Walls. In eLS; John Wiley & Sons: Chichester, 2012.

(2) Keegstra, K. Plant Cell Walls. Plant Physiol. 2010, 154, 483–486.

(3) Whitney, S. E. C.; Gothard, M. G. E.; Mitchell, J. T.; Gidley, M. J. Roles of Cellulose and

Xyloglucan in Determining the Mechanical Properties of Primary Plant Cell Walls. Plant

Physiol. 1999, 121, 657–664.

(4) Scheller, H. V.; Ulvskov, P. Hemicelluloses. Annu. Rev. Plant Biol. 2010, 61, 263–289.

(5) Gillis, P. P.; Mark, R. E.; Tang, R. C. Elastic Stiffness of Crystalline Cellulose in the Folded-

Chain Solid State. J. Mater. Sci. 1969, 4, 1003–1007.

(6) Caffall, K. H.; Mohnen, D. The Structure, Function, and Biosynthesis of Plant Cell Wall Pectic

Polysaccharides. Carbohydr. Res. 2009, 344, 1879–1900.

(7) Zykwinska, A.; Thibault, J. F.; Ralet, M. C. Competitive Binding of Pectin and Xyloglucan

with Primary Cell Wall Cellulose. Carbohydr. Polym. 2008, 74, 957–961.

(8) Zykwinska, A. W.; Ralet, M. J.; Garnier, C. D.; Thibault, J. J. Evidence for in Vitro Binding of

Pectin Side Chains to Cellulose. Plant Physiol. 2005, 139, 397–407.

(9) Popper, Z. A.; Fry, S. C. Widespread Occurrence of a Covalent Linkage between Xyloglucan

and Acidic Polysaccharides in Suspension-Cultured Angiosperm Cells. Ann. Bot. 2005, 96, 91–

99.

(10) Ellis, M.; Egelund, J.; Schultz, C. J.; Bacic, A. Arabinogalactan-Proteins (AGPs): Key

Regulators at the Cell Surface. Plant Physiol. 2010, 153, 403–419.

(11) Jamet, E.; Albenne, C.; Boudart, G.; Irshad, M.; Canut, H.; Pont-Lezica, R. Recent Advances

in Plant Cell Wall Proteomics. Proteomics 2008, 8, 893–908.

154

(12) Lee, S. J.; Saravanan, R. S.; Damasceno, C. M. B.; Yamane, H.; Kim, B. D.; Rose, J. K. C.

Digging Deeper into the Plant Cell Wall Proteome. Plant Physiol. Biochem. 2004, 42, 979–

988.

(13) Doblin, M. S.; Pettolino, F.; Bacic, A. Evans Review: Plant Cell Walls: The Skeleton of the

Plant World. Funct. Plant Biol. 2010, 37, 357–381.

(14) Fry, S. C. Plant Cell Walls. In eLS; John Wiley & Sons, Ltd, 2001; pp 1–11.

(15) Burton, R. A.; Gidley, M. J.; Fincher, G. B. Heterogeneity in the Chemistry, Structure and

Function of Plant Cell Walls. Nat. Chem. Biol. 2010, 6, 724–732.

(16) Sørensen, I.; Domozych, D.; Willats, W. G. T. How Have Plant Cell Walls Evolved? Plant

Physiol. 2010, 153, 366–372.

(17) Kenrick, P.; Crane, P. R. The Origin and Early Evolution of Plants on Land. Nature 1997, 389,

33–39.

(18) Becker, B.; Marin, B. Streptophyte Algae and the Origin of Embryophytes. Ann. Bot. 2009,

103, 999–1004.

(19) Michel, G.; Tonon, T.; Scornet, D.; Cock, J. M.; Kloareg, B. The Cell Wall Polysaccharide

Metabolism of the Brown Alga Ectocarpus Siliculosus. Insights into the Evolution of

Extracellular Matrix Polysaccharides in Eukaryotes. New Phytol. 2010, 188, 82–97.

(20) Popper, Z. A.; Tuohy, M. G. Beyond the Green: Understanding the Evolutionary Puzzle of

Plant and Algal Cell Walls. Plant Physiol. 2010, 153, 373–383.

(21) Popper, Z. A. Evolution and Diversity of Green Plant Cell Walls. Curr. Opin. Plant Biol. 2008,

11, 286–292.

(22) Popper, Z. A.; Michel, G.; Herve, C.; Domozych, D. S.; Willats, W. G.; Tuohy, M. G.; Kloareg,

B.; Stengel, D. B. Evolution and Diversity of Plant Cell Walls: From Algae to Flowering Plants.

155

Annu. Rev. Plant Biol. 2011, 62, 567–590.

(23) Popper, Z. A.; Ralet, M. C.; Domozych, D. S. Plant and Algal Cell Walls: Diversity and

Functionality. Ann. Bot. 2014, 114, 1043–1048.

(24) Fangel, J. U.; Ulvskov, P.; Knox, J. P.; Mikkelsen, M. D.; Harholt, J.; Popper, Z. A.; Willats,

W. G. T. Cell Wall Evolution and Diversity. Front. Plant Sci. 2012, 3, 1–8.

(25) Tolonen, L. K.; Juvonen, M.; Niemelä, K.; Mikkelson, A.; Tenkanen, M.; Sixta, H.

Supercritical Water Treatment for Cello-Oligosaccharide Production from Microcrystalline

Cellulose. Carbohydr. Res. 2015, 401, 16–23.

(26) Takeo, K.; Yasato, T.; Kuge, T. Synthesis of α- and β-Cellotriose Hendecaacetates and of

Several 6,6’,6’’-tri-Substituted Derivatives of Methyl β-Cellotrioside. Carbohydr. Res. 1981,

93, 148–156.

(27) Takeo, K.; Okushio, K.; Fukuyama, K.; Kuge, T. Synthesis of Cellobiose, Cellotriose,

Cellotetraose and Lactose. Carbohydr. Res. 1983, 121, 163–173.

(28) Rodriguez, E.; Stick, R. The Synthesis of Active-Site Directed Inhibitors of Some β-Glucan

Hydrolases. Aust. J. Chem. 1990, 43, 665–679.

(29) Xu, J.; Vasella, A. Oligosaccharide Analogues of Polysaccharides. Helv. Chim. Acta 1999, 82,

1728–1752.

(30) Chang, S.; Han, J.; Tseng, S. Y.; Lee, H.; Lin, C.; Lin, Y.; Jeng, W.; Wang, A. H.; Wu, C.;

Wong, C. Glycan Array on Aluminum Oxide-Coated Glass Slides through Phosphonate

Chemistry. J. Am. Chem. Soc. 2010, 132, 13371–13380.

(31) Dallabernardina, P.; Schuhmacher, F.; Seeberger, P. H.; Pfrengle, F. Automated Glycan

Assembly of Xyloglucan Oligosaccharides. Org. Biomol. Chem. 2016, 14, 309–313.

(32) Sugawara, F.; Nakayama, H.; Ogawa, T. Synthetic Studies on Rhynchosporoside:

156

Stereoselective Synthesis of 1-O- and 2-O-Cellotriosyl-3-Deoxy-2(R)-and -2(S)-Glycerol.

Carbohydr. Res. 1983, 123, C25–C28.

(33) Nicolaou, K. C.; Randall, J. L.; Furst, G. T. Stereospecific Synthesis of Rhynchosporosides, a

Family of Fungal Metabolites Causing Scald Disease in Barley and Other Grasses. J. Am.

Chem. Soc. 1985, 107, 5556–5558.

(34) Sugawara, F.; Nakayama, H.; Strobel, G. A.; Ogawa, T. Stereoselective Synthesis of 1- and 2-

O-α-D-Cellotriosyl-3-Deoxy-2(R)- and 2(S)-Glycerols Related to Rhynchosporoside. Agric.

Biol. Chem. 1986, 50, 2261–2271.

(35) Nishimura, T.; Nakatsubo, F. First Synthesis of Cellooctaose by a Convergent Synthetic

Method. Carbohydr. Res. 1996, 294, 53–64.

(36) Koenigs, W.; Knorr, E. Über Einige Derivate Des Traubensuckers Und Der Galactose. Berichte

der Dtsch. Chem. Gesellschaft 1901, 34, 957–981.

(37) Zemplén, G.; Pacsu, E. Über Die Verseifung Acetylierter Zucker Und Verwandter Substanzen.

Berichte der Dtsch. Chem. Gesellschaft 1929, 62, 1613–1614.

(38) Nishimura, T.; Nakatsubo, F.; Murakami, K. Synthetic Studies of Cellulose XI. High Yield

Synthesis of Cellotetraose Derivative by Convergent Synthetic Method from an Acyl Glucose

Derivative. Mouzai Gakkaishi 1994, 40, 44–49.

(39) Schmidt, R. R.; Michel, J. Facile Synthesis of α- and β-O-Glycosyl Imidates; Preparation of

Glycosides and Disaccharides. Angew. Chem. Int. Ed. 1980, 19, 731–732.

(40) Kariyone, K.; Yazawa, H. Oxidative Cleavage of Β, γ-Unsaturated Ether. Tetrahedron Lett.

1970, 33, 2885–2888.

(41) Schultze, E. Zur Kenntniss Der Chemischen Zuzammensetzung Der Pflanzlichen

Zellmembranen. Ber. Dtsch. Chem. Ges. 1891, 24, 2277–2287.

157

(42) Ebringerová, A.; Hromádková, Z.; Heinze, T. Hemicellulose. In Polysaccharides I; Springer-

Verlag: Berlin, 2005; Vol. 186, pp 1–67.

(43) Wotovic, A.; Jacquinet, J. C.; Sinaÿ, P. Synthesis of Oligosaccharins: A Chemical Synthesis of

Propyl O-β-D-Galactopyranosyl-(1,2)-O-α-D-Xylopyranosyl-(1,6)-O-β-D-Glucopyranosyl-

(1,4)-β-D-Glucopyranoside. Carbohydr. Res. 1990, 205, 235–245.

(44) Sakai, K.; Nakahara, Y.; Ogawa, T. Total Synthesis of Nonasaccharide Repeating Unit of Plant

Cell Wall Xyloglucan: An Endogenous Hormone Which Regulates Cell Growth. Tetrahedron

Lett. 1990, 31, 3035–3038.

(45) Watt, D. K.; Brasch, D. J.; Larsen, D. S.; Melton, L. D.; Simpson, J. Oligosaccharides Related

to Xyloglucan: Synthesis and X-Ray Crystal Structure of Methyl α-L- Fucopyranosyl-(1→2)-

β-D-Galactopyranosyl-(1→2)-α-D-Xylopyranoside and the Synthesis of Methyl α-L-

Fucopyranosyl-(1→2)-β-D-Galactopyranosyl-(1→2)-β-D-Xylopyranosid. Carbohydr. Res.

2000, 325, 300–312.

(46) Merrifield, R. B. Solid Phase Peptide Synthesis. I. The Synthesis of a Tetrapeptide. J. Am.

Chem. Soc. 1963, 85, 2149–2154.

(47) Eller, S.; Collot, M.; Yin, J.; Hahm, H. S.; Seeberger, P. H. Automated Solid-Phase Synthesis

of Chondroitin Sulfate Glycosaminoglycans. Angew. Chemie Int. Ed. 2013, 52, 5858–5861.

(48) Plante, O. J.; Andrade, R. B.; Seeberger, P. H. Synthesis and Use of Glycosyl Phosphates as

Glycosyl Donors. Org. Lett. 1999, 1, 211–214.

(49) Manners, D. J.; Masson, A. J.; Patterson, J. C. The Structure of a β-(1→3)-D-Glucan from

Yeast Cell Walls. Biochem. J. 1973, 135, 19–30.

(50) Scherp, P.; Grotha, R.; Kutschera, U. Occurrence and Phylogenetic Significance of

Cytokinesis-Related Callose in Green Algae, Bryophytes, Ferns and Seed Plants. Plant Cell

158

Rep. 2001, 20, 143–149.

(51) Ellinger, D.; Voigt, C. A. Callose Biosynthesis in Arabidopsis with a Focus on Pathogen

Response: What We Have Learned within the Last Decade. Ann. Bot. 2014, 114, 1349–1358.

(52) Tanaka, H.; Kawai, T.; Adachi, Y.; Hanashima, S.; Yamaguchi, Y.; Ohno, N.; Takahashi, T.

Synthesis of β-(1,3)-Oligoglucans Exhibiting a Dectin-1 Binding Affinity and Their Biological

Evaluation. Bioorg. Med. Chem. 2012, 20, 3898–3914.

(53) Takeo, K.; Maki, K.; Wada, Y.; Kitamura, S. Synthesis of the Laminara-Oligosaccharide

Methyl β-Glycosides of Dp 3-8. Carbohydr. Res. 1993, 245, 81–96.

(54) Adamo, R.; Tontini, M.; Brogioni, G.; Romano, M. R.; Costantini, G.; Danieli, E.; Proietti, D.;

Berti, F.; Costantino, P. Synthesis of Laminarin Fragments and Evaluation of a β-(1→3) Glucan

Hexasaccaride-CRM 197 Conjugate as Vaccine Candidate against Candida Albicans. J.

Carbohydr. Chem. 2011, 30, 249–280.

(55) Jamois, F.; Ferrières, V.; Guégan, J. P.; Yvin, J. C.; Plusquellec, D.; Vetvicka, V. Glucan-like

Synthetic Oligosaccharides: Iterative Synthesis of Linear Oligo-β-(1,3)-Glucans and

Immunostimulatory Effects. Glycobiology 2005, 15, 393–407.

(56) de Jong, A. R.; Volbeda, A. G.; Hagen, B.; van den Elst, H.; Overkleeft, H. S.; van der Marel,

G. A.; Codée, J. D. C. A Second-Generation Tandem Ring-Closing Metathesis Cleavable

Linker for Solid-Phase Oligosaccharide Synthesis. Eur. J. Org. Chem. 2013, 29, 6644–6655.

(57) Zeng, Y.; Kong, F. Synthesis of β-D-Glucose Oligosaccharides from Phytophthora Parasitica.

Carbohydr. Res. 2003, 338, 2359–2366.

(58) Huang, G.; Mei, X.; Liu, M. Synthesis of (R)-2,3-epoxypropyl(1→3)-β-D-Pentaglucoside.

Carbohydr. Res. 2005, 340, 603–608.

(59) Yashunsky, D. V.; Tsvetkov, Y. E.; Nifantiev, N. E. Synthesis of 3-Aminopropyl Glycoside of

159

Branched β-(1 → 3)-D-Glucooctaoside. Carbohydr. Res. 2016, 436, 25–30.

(60) Mo, K.; Li, H.; Mague, J. T.; Ensley, H. E. Synthesis of the β-1,3-Glucan, Laminarahexaose:

NMR and Conformational Studies. Carbohydr. Res. 2009, 344, 439–447.

(61) Elsaidi, H. R. H.; Paszkiewicz, E.; Bundle, D. R. Synthesis of a 1,3 β-Glucan Hexasaccharide

Designed to Target Vaccines to the Dendritic Cell Receptor, Dectin-1. Carbohydr. Res. 2015,

408, 96–106.

(62) Liao, G.; Zhou, Z.; Burgula, S.; Liao, J.; Yuan, C.; Wu, Q.; Guo, Z. Synthesis and

Immunological Studies of Linear Oligosaccharides of β-Glucan As Antigens for Antifungal

Vaccine Development. Bioconjug. Chem. 2015, 26, 466–476.

(63) Tanaka, H.; Kawai, T.; Adachi, Y.; Ohno, N.; Takahashi, T. β(1,3) Branched Heptadeca- and

Linear Hexadecasaccharides Possessing an Aminoalkyl Group as a Strong Ligand to Dectin-1.

Chem. Commun. 2010, 46, 8249–8251.

(64) Liao, G.; Burgula, S.; Zhou, Z.; Guo, Z. A Convergent Synthesis of 6- O -Branched β-Glucan

Oligosaccharides. Eur. J. Org. Chem. 2015, 2942–2951.

(65) Weishaupt, M. W.; Hahm, H. S.; Geissner, A.; Seeberger, P. H. Automated Glycan Assembly

of Branched β-(1→3)-Glucans to Identify Antibody Epitopes. Chem. Commun. 2017, 53,

3591–3594.

(66) Weishaupt, M. W.; Matthies, S.; Seeberger, P. H. Automated Solid-Phase Synthesis of a β-

(1,3)-Glucan Dodecasaccharide. Chem. Eur. J. 2013, 19, 12497–12503.

(67) Popper, Z. A.; Fry, S. C. Primary Cell Wall Composition of Bryophytes and Charophytes. Ann.

Bot. 2003, 91, 1–12.

(68) Sørensen, I.; Pettolino, F. A.; Wilson, S. M.; Doblin, M. S.; Johansen, B.; Bacic, A.; Willats,

W. G. T. Mixed-Linkage (1→3),(1→4)-β-D-Glucan Is Not Unique to the Poales and Is an

160

Abundant Component of Equisetum Arvense Cell Walls. Plant J. 2008, 54, 510–521.

(69) Dallabernardina, P.; Schuhmacher, F.; Seeberger, P. H.; Pfrengle, F. Mixed-Linkage Glucan

Oligosaccharides Produced by Automated Glycan Assembly Serve as Tools To Determine the

Substrate Specificity of Lichenase. Chem. A Eur. J. 2017, 23, 1–7.

(70) Simmons, T. J.; Mortimer, J. C.; Bernardinelli, O. D.; Pöppler, A.-C.; Brown, S. P.;

DeAzevedo, E. R.; Dupree, R.; Dupree, P. Folding of Xylan onto Cellulose Fibrils in Plant Cell

Walls Revealed by Solid-State NMR. Nat. Commun. 2016, 7, 1–9.

(71) Rennie, E. A.; Scheller, H. V. Xylan Biosynthesis. Curr. Opin. Biotechnol. 2014, 26, 100–107.

(72) Faik, A. Xylan Biosynthesis: News from the Grass. Plant Physiol. 2010, 153, 396–402.

(73) Balakshin, M.; Capanema, E.; Berlin, A. Isolation and Analysis of Lignin-Carbohydrate

Complexes Preparations with Traditional and Advanced Methods: A Review. In Studies in

Natural Products Chemistry; Atta-ur-Rahman, Ed.; Elsevier B.V., 2014; Vol. 42, pp 83–115.

(74) Hirsch, J.; Kováč, P.; Petráková, E. An Approach to the Systematic Synthesis of (1→4)-β-D-

Xylo-Oligosaccharides. Carbohydr. Res. 1982, 106, 203–216.

(75) Kováč, P.; Hirsch, J. Systematic, Sequential Synthesis of (1→4)-β-D-Xylo-Ligosaccharides

and Their Methyl β-Glycosides. Carbohydr. Res. 1981, 90, C5–C7.

(76) Takeo, K.; Ohguchi, Y.; Hasegawa, R.; Kitamura, S. Synthesis of (1→4)-β-D-Xylo-

Oligosaccharides of Dp 4–10 by a Blockwise Approach. Carbohydr. Res. 1995, 278, 301–313.

(77) Schmidt, D.; Schuhmacher, F.; Geissner, A.; Seeberger, P. H.; Pfrengle, F. Automated

Synthesis of Arabinoxylan-Oligosaccharides Enables Characterization of Antibodies That

Recognize Plant Cell Wall Glycans. Chem. Eur. J. 2015, 21, 5709–5713.

(78) Oscarson, S.; Svahnberg, P. Synthesis of Uronic Acid-Containing Xylans Found in Wood and

Pulp. J. Chem. Soc. Perkin Trans. 1 2001, 100, 873–879.

161

(79) Kováč, P.; Hirsch, J.; Kováčik, V.; Kočiš, P. The Stepwise Synthesis of an Aldopentaouronic

Acid Derivative Related to Branched (4-O-Methylglucurono)xylans. Carbohydr. Res. 1980,

85, 41–49.

(80) Senf, D.; Ruprecht, C.; de Kruijff, G. H. M.; Simonetti, S. O.; Schuhmacher, F.; Seeberger, P.

H.; Pfrengle, F. Active Site Mapping of Xylan-Deconstructing Enzymes with Arabinoxylan

Oligosaccharides Produced by Automated Glycan Assembly. Chem. - A Eur. J. 2017, 23, 3197–

3205.

(81) Helferich, B.; Schmitz-Hillebrecht, E. Eine Neue Methode Zur Synthese von Glykosiden Der

Phenole. Ber. Dtsch. Chem. Ges. 1933, 66, 378–383.

(82) Myhre, D. V.; Smith, F. Synthesis of Xylobiose (4-O-β-D-Xylopyranosyl-D-Xylose) 1. J. Org.

Chem. 1961, 26, 4609–4612.

(83) Kováč, P.; Hirsch, J. Sequential Synthesis and 13C-N.M.R. Spectra of Methyl β-Glycosides of

(1→4)-β-D-Xylo-Oligosaccharides. Carbohydr. Res. 1982, 100, 177–193.

(84) Fürst, A.; Plattner, P. A. 2-α-, 3-α- Und 2-Β, 3-β-Oxido-Chlolestane; Konfiguration Der 2-Oxy-

Cholestane. Helv. Chim. Acta 1949, 32, 275–283.

(85) Moreira, L. R. S.; Filho, E. X. F. An Overview of Mannan Structure and Mannan-Degrading

Enzyme Systems. Appl. Microbiol. Biotechnol. 2008, 79, 165–178.

(86) Gridley, J. J.; Osborn, H. M. I. Recent Advances in the Construction of β-D-Mannose and β-

D-Mannosamine Linkages. J. Chem. Soc. Perkin Trans. 1 2000, 10, 1471–1491.

(87) Lichtenthaler, F. W. 2-Oxoglycosyl (“Ulosyl”) and 2-Oximinoglycosyl Bromides: Versatile

Donors for the Expedient Assembly of Oligosaccharides with β- D-Mannose, β- L-Rhamnose,

N-Acetyl-β- D-Mannosamine, and N-Acetyl-β-D-Mannosaminuronic Acid Units. Chem. Rev.

2011, 111, 5569–5609.

162

(88) Twaddle, G. W. J.; Yashunsky, D. V; Nikolaev, A. V. The Chemical Synthesis of β-(1→4)-

Linked D-Mannobiose and D-Mannotriose. Org. Biomol. Chem 2003, 1, 623–628.

(89) Kim, K. S.; Fulse, D. B.; Baek, J. Y.; Lee, B.-Y.; Jeon, H. B. Stereoselective Direct

Glycosylation with Anomeric Hydroxy Sugars by Activation with Phthalic Anhydride and

Trifluoromethanesulfonic Anhydride Involving Glycosyl Phthalate Intermediates. J. Am.

Chem. Soc. 2008, 130, 8537–8547.

(90) Crich, D.; Banerjee, A.; Yao, Q. Direct Chemical Synthesis of the β-D-Mannans: The β-(1→2)

and β-(1→4) Series. J. Am. Chem. Soc. 2004, 126, 14930–14934.

(91) Ohara, K.; Lin, C.-C.; Yang, P.-J.; Hung, W.-T.; Yang, W.-B.; Cheng, T. J. R.; Fang, J.-M.;

Wong, C.-H. Synthesis and Bioactivity of β-(1→4)-Linked Oligomannoses and Partially

Acetylated Derivatives. J. Org. Chem. 2013, 78, 6390–6411.

(92) Tang, S.-L.; Pohl, N. L. B. Automated Solution-Phase Synthesis of β-1,4-Mannuronate and β-

1,4-Mannan. Org. Lett. 2015, 17, 2642–2645.

(93) Ekholm, F. S.; Ardá, A.; Eklund, P.; André, S.; Gabius, H.-J.; Jiménez-Barbero, J.; Leino, R.

Studies Related to Norway Spruce Galactoglucomannans: Chemical Synthesis, Conformation

Analysis, NMR Spectroscopic Characterization, and Molecular Recognition of Model

Compounds. Chem. Eur. J. 2012, 18, 14392–14405.

(94) Imamura, A.; Ando, H.; Korogi, S.; Tanabe, G.; Muraoka, O.; Ishida, H.; Kiso, M. Di-Tert-

Butylsilylene (DTBS) Group-Directed α-Selective Galactosylation Unaffected by C-2

Participating Functionalities. Tetrahedron Lett. 2003, 44, 6725–6728.

(95) Vauquelin, M. Analyse Du Tamarin. Ann. Chim. 1790, 5, 92–106.

(96) Braconnot, H. Recherche Sur Un Nouvel Acid Universellement Répandu Dans Tous Les

Végétaux. Ann. Chim. Phys. 1825, 2, 173–178.

163

(97) Mohnen, D. Pectin Structure and Biosynthesis. Curr. Opin. Plant Biol. 2008, 11, 266–277.

(98) Ridley, B. L.; O’Neill, M. A.; Mohnen, D. Pectins: Structure, Biosynthesis, and

Oligogalacturonide-Related Signaling. Phytochemistry 2001, 57, 929–967.

(99) Yapo, B. M. Pectic Substances: From Simple Pectic Polysaccharides to Complex Pectins-A

New Hypothetical Model. Carbohydr. Polym. 2011, 86, 373–385.

(100) Neill, M. A. O.; Ishii, T.; Albersheim, P.; Darvill, A. G. Rhamnogalacturonan II : Structure and

Function of a Borate Cross-Linked Cell. Annu. Rev. Plant Biol. 2004, 55, 109–139.

(101) Nepogodiev, S. A.; Field, R. A.; Damager, I. Chapter 3:Approaches to Chemical Synthesis of

Pectic Oligosaccharides. In Annual Plant Reviews, Volume 41: Plant Polysaccharides,

Biosynthesis and Bioengineering; Ulvskov, P., Ed.; Blackwell Publishing Ltd: West Sussex,

United Kingdom, 2011; Vol. 41, pp 65–92.

(102) Yapo, B. M.; Lerouge, P.; Thibault, J. F.; Ralet, M. C. Pectins from Citrus Peel Cell Walls

Contain Homogalacturonans Homogenous with Respect to Molar Mass, Rhamnogalacturonan

I and Rhamnogalacturonan II. Carbohydr. Polym. 2007, 69, 426–435.

(103) Willats, W. G. T.; McCartney, L.; Mackie, W.; Knox, J. P. Pectin: Cell Biology and Prospects

for Functional Analysis. Plant Mol. Biol. 2001, 47, 9–27.

(104) Petersen, B. O.; Meier, S.; Duus, J. Ø.; Clausen, M. H. Structural Characterization of

Homogalacturonan by NMR Spectroscopy-Assignment of Reference Compounds. Carbohydr.

Res. 2008, 343, 2830–2833.

(105) Nakahara, Y.; Ogawa, T. Synthesis of (1 ->4)-Linked Galacturonic Acid Trisaccharides, a

Proposed Plant Wound-Hormone and a Stereoisomer. Carbohydr. Res. 1990, 200, 363–375.

(106) Pourcelot, M.; Barbier, M.; Landoni, M.; Couto, A.; Grand, E.; Kovensky, J. Synthesis of

Galacturonate Containing Disaccharides and Protein Conjugates. Curr. Org. Chem. 2011, 15,

164

3523–3534.

(107) Kester, H. C. M.; Magaud, D.; Roy, C.; Anker, D.; Doutheau, A.; Shevchik, V.; Hugouvieux-

Cotte-Pattat, N.; Benen, J. A. E.; Visser, J. Performance of Selected Microbial Pectinase on

Sythetic Monomethyl-Esterified Di- and Trigalacturonates. J Biol Chem 1999, 274, 37053–

37059.

(108) Clausen, M. H.; Jørgensen, M. R.; Thorsen, J.; Madsen, R. A Strategy for Chemical Synthesis

of Selectively Methyl-Esterified Oligomers of Galacturonic Acid. J. Chem. Soc. Perkin Trans.

1 2001, 543–551.

(109) Clausen, M. H.; Madsen, R. Synthesis of Hexasaccharide Fragments of Pectin. Chem. Eur. J.

2003, 9, 3821–3832.

(110) Nakahara, Y.; Ogawa, T. Stereocontrolled Synthesis of the Propyl Glycoside of a

Decagalacturonic Acid, a Model Compound for the Endogenous Phytoalexin Elicitor-Active

Oligogalacturonic. Carbohydr. Res. 1989, 194, 95–114.

(111) Nakahara, Y.; Ogawa, T. Stereoselective Total Synthesis of Dodecagalacturonic Acid, a

Phytoalexin Elicitor of Soybean. Carbohydr. Res. 1990, 205, 147–159.

(112) van den Bos, L. J.; Codée, J. D. C.; Litjens, R. E. J. N.; Dinkelaar, J.; Overkleeft, H. S.; van der

Marel, G. A. Uronic Acids in Oligosaccharide Synthesis. Eur. J. Org. Chem. 2007, 3963–3976.

(113) Magaud, D.; Grandjean, C.; Doutheau, A.; Anker, D.; Shevchik, V.; Cotte-Pattat, N.; Robert-

Baudouy, J. An Efficient and Highly Stereoselective α(1→4) Glycosylation between Two D-

Galacturonic Acid Ester Derivatives. Tetrahedron Lett. 1997, 38, 241–244.

(114) Kramer, S.; Nolting, B.; Ott, A.; Vogel, C. Synthesis of Homogalacturonan Fragments. J.

Carbohydr. Chem. 2000, 19, 891–921.

(115) Omura, K.; Swern, D. Oxidation of Alcohols by “activated” Dimethyl Sulfoxide. A Preparative,

165

Steric and Mechanistic Study. Tetrahedron 1978, 34, 1651–1660.

(116) Bal, B. S.; Childers, W. E.; Pinnick, H. W. Oxidation of Α,β-Un Saturated Aldehydes.

Tetrahedron 1981, 37, 2091–2096.

(117) Kraus, G. A.; Roth, B. Synthetic Studies toward Verrucarol. 2. Synthesis of the AB Ring

System. J. Org. Chem. 1980, 45, 4825–4830.

(118) Lindgren, B. O.; Nilsson, T.; Husebye, S.; Mikalsen, Ø.; Leander, K.; Swahn, C.-G. Preparation

of Carboxylic Acids from Aldehydes (Including Hydroxylated Benzaldehydes) by Oxidation

with Chlorite. Acta Chem. Scand. 1973, 27, 888–890.

(119) Clausen, M. H.; Madsen, R. Synthesis of Hexasaccharide Fragments of Pectin. Chemistry 2003,

9, 3821–3832.

(120) Dess, D. B.; Martin, J. C. Readily Accessible 12-I-5 Oxidant for the Conversion of Primary

and Secondary Alcohols to Aldehydes and Ketones. J. Org. Chem. 1983, 48, 4155–4156.

(121) Schols, H. A.; Bakx, E. J.; Schipper, D.; Voragen, A. G. J. A Xylogalacturonan Subunit Present

in the Modified Hairy Regions of Apple Pectin. Carbohydr. Res. 1995, 279, 265–279.

(122) Hart, D. A.; Kindel, P. K. Isolation and Partial Characterization of Apiogalacturonans from the

Cell Wall of Lemna Minor. Biochem. J. 1970, 116, 569–579.

(123) Buffet, M. a J.; Rich, J. R.; McGavin, R. S.; Reimer, K. B. Synthesis of a Disaccharide

Fragment of Rhamnogalacturonan II. Carbohydr. Res. 2004, 339, 2507–2513.

(124) Nepogodiev, S. A.; Fais, M.; Hughes, D. L.; Field, R. a. Synthesis of Apiose-Containing

Oligosaccharide Fragments of the Plant Cell Wall: Fragments of Rhamnogalacturonan-II Side

Chains A and B, and Apiogalacturonan. Org. Biomol. Chem. 2011, 9, 6670–6684.

(125) Koós, M.; Mosher, H. S. A Convenient Synthesis of Apiose. Carbohydr. Res. 1986, 146, 335–

341.

166

(126) Nachman, R. J.; Hönel, M.; Williams, T. M.; Halaska, R. C.; Mosher, H. S. Methyl 3-Formyl -

2,3-O-Isopropylidene-D-Erythrofuranoside (D-Apiose Aldal) and Derivatives. J. Org. Chem.

1986, 51, 4802–4806.

(127) Ho, P.-T. Branched-Chain Sugars. Reaction of Furanoses with Formaldehyde: A Simple

Synthesis of D- and L-Apiose. Can. J. Chem. 1979, 57, 381–383.

(128) Williams, D. T.; Jones, J. K. N. The Chemistry of Apiose. Part I. Can. J. Chem. 1964, 42, 69–

72.

(129) Scanlan, E. M.; Mackeen, M. M.; Wormald, M. R.; Davis, B. G. Synthesis and Solution-Phase

Conformation of the RG-I Fragment of the Plant Polysaccharide Pectin Reveals a Modification-

Modulated Assembly Mechanism. J. Am. Chem. Soc. 2010, 132, 7238–7239.

(130) Mossine, V. V; Glinsky, V. V; Mawhinney, T. P. Food-Related Carbohydrate Ligands for

Galectins. In Food and Feed Chemistry; Columbia, USA., 2008; pp 235–270.

(131) Clarke, A. E.; Anderson, R. L.; Stone, B. A. Form and Function of Arabinogalactans and

Arabinogalactan-Proteins. Phytochemistry 1979, 18, 521–540.

(132) Ma, Y.; Cao, X.; Yu, B. Synthesis of Oligosaccharide Fragments of the Rhamnogalacturonan

of Nerium Indicum. Carbohydr. Res. 2013, 377, 63–74.

(133) Maruyama, M.; Takeda, T.; Shimizu, N.; Hada, N.; Yamada, H. Synthesis of a Model

Compound Related to an Anti-Ulcer Pectic Polysaccharide. Carbohydr. Res. 2000, 325, 83–

92.

(134) Nemati, N.; Karapetyan, G.; Nolting, B.; Endress, H.-U.; Vogel, C. Synthesis of

Rhamnogalacturonan I Fragments by a Modular Design Principle. Carbohydr. Res. 2008, 343,

1730–1742.

(135) Reiffarth, D.; Reimer, K. B. Synthesis of Two Repeat Units Corresponding to the Backbone of

167

the Pectic Polysaccharide Rhamnogalacturonan I. Carbohydr. Res. 2008, 343, 179–188.

(136) Zakharova, A. N.; Madsen, R.; Clausen, M. H. Synthesis of a Backbone Hexasaccharide

Fragment of the Pectic Polysaccharide Rhamnogalacturonan I. Org. Lett. 2013, 15, 1826–1829.

(137) Ndeh, D.; Rogowski, A.; Cartmell, A.; Luis, A. S.; Baslé, A.; Gray, J.; Venditto, I.; Briggs, J.;

Zhang, X.; Labourel, A.; Terrapon, N.; Buffetto, F.; Nepogodiev, S.; Xiao, Y.; Field, R. A.;

Zhu, Y.; O’Neill, M. A.; Urbanowicz, B. R.; York, W. S.; Davies, G. J.; Abbott, D. W.; Ralet,

M-C.; Martens, E. C.; Henrissat, B.; Gilbert, H. J. Complex Pectin Metabolism by Gut Bacteria

Reveals Novel Catalytic Functions. Nature 2017, 544, 65–70.

(138) Pellerin, P.; Neill, M. A.; Pierre, C.; Cabanis, M.-T.; Darvill, A. G.; Albersheim, P.; Moutounet,

M. Leat Complexation in Wines with the Dimers of the Grape Pectic Polysaccharide

Rhamnogalacturonan II. Int. J. Sci. Vigne Vin 1997, 31, 33–41.

(139) Pellerin, P.; Doco, T.; Vidal, S.; Williams, P.; Brillouet, J. M.; O’Neill, M. A. Structural

Characterization of Red Wine Rhamnogalacturonan II. Carbohydr. Res. 1996, 290, 183–197.

(140) Whitcombe, A. J.; O’Neill, M. A.; Steffan, W.; Albersheim, P.; Darvill, A. G. Structural

Characterization of the Pectic Polysaccharide, Rhamnogalacturonan-II. Carbohydr. Res. 1995,

271, 15–29.

(141) Pabst, M.; Fischl, R. M.; Brecker, L.; Morelle, W.; Fauland, A.; Köfeler, H.; Altmann, F.;

Léonard, R. Rhamnogalacturonan II Structure Shows Variation in the Side Chains

Monosaccharide Composition and Methylation Status within and across Different Plant

Species. Plant J. 2013, 76, 61–72.

(142) Scheller, H. V.; Jensen, J. K.; Sørensen, S. O.; Harholt, J.; Geshi, N. Biosynthesis of Pectin.

Physiol. Plant. 2006, 129, 283–295.

(143) Glushka, J. N.; Terrell, M.; York, W. S.; Neill, M. A. O.; Gucwa, A.; Darvill, A. G.;

168

Albersheim, P.; Prestegard, J. H. Primary Structure of the 2-O-Methyl-a- L -Fucose-Containing

Side Chain of the Pectic Polysaccharide , Rhamnogalacturonan II. Carbohydr. Res. 2003, 338,

341–352.

(144) Penhoat, H. C.; Gey, C.; Pellerin, P.; Perez, S. An NMR Solution Study of the Mega-

Oligosaccharide, Rhamnogalacturonan II. J. Biomol. NMR 1999, 14, 253–271.

(145) Pérez, S.; Rodríguez-Carvajal, M. A.; Doco, T. A Complex Plant Cell Wall Polysaccharide:

Rhamnogalacturonan II. A Structure in Quest of a Function. Biochimie 2003, 85, 109–121.

(146) Vidal, S.; Doco, T.; Williams, P.; Pellerin, P.; York, W. S.; O’Neill, M. A.; Glushka, J.; Darvill,

A. G.; Albersheim, P. Structural Characterization of the Pectic Polysaccharide

Rhamnogalacturonan II: Evidence for the Backbone Location of the Aceric Acid-Containing

Oligoglycosyl Side Chain. Carbohydr. Res. 2000, 326, 277–294.

(147) Rodrı́guez-Carvajal, M. A.; Hervé du Penhoat, C.; Mazeau, K.; Doco, T.; Pérez, S. The Three-

Dimensional Structure of the Mega-Oligosaccharide Rhamnogalacturonan II Monomer: A

Combined Molecular Modeling and NMR Investigation. Carbohydr. Res. 2003, 338, 651–671.

(148) Ishii, T.; Matsunaga, T. Isolation and Characterization of a Boron-Rhamnogalacturonan-II

Complex from Cell Walls of Sugar Beet Pulp. Carbohydr. Res. 1996, 284, 1–9.

(149) Funakawa, H.; Miwa, K. Synthesis of Borate Cross-Linked Rhamnogalacturonan II. Front.

Plant Sci. 2015, 6, 1–8.

(150) Ishii, T.; Matsunaga, T.; Hayashi, N. Formation of Rhamnogalacturonan II-Borate Dimer in

Pectin Determines Cell Wall Thickness of Pumpkin Tissue. Plant Physiol. 2001, 126, 1698–

1705.

(151) Melton, L. D.; McNeil, M.; Darvill, A. G.; Albersheim, P. Structural Characterization of

Oligosaccharides Isolated from the Pectic Polysaccharide RG-II. Carbohydr. Res. 1986, 146,

169

279–305.

(152) Spellman, M. W.; McNeil, M.; Darvill, A. G.; Albersheim, P.; Dell, A. Characterization of a

Structurally Complex Heptasaccharide Isolated from the Pectic Polysaccharide

Rhamnogalacturonan II. Carbohydr. Res. 1983, 122, 131–153.

(153) Backman, I.; Jansson, P.; Kenne, L. Synthesis, NMR and Conformational Studies of Some 1,

4-Linked Disaccharides. J. Chem. Soc., Perkin Trans. 1 1990, 50, 1383–1388.

(154) Chauvin, A.-L.; Nepogodiev, S. A.; Field, R. A. Synthesis of a 2,3,4-Triglycosylated

Rhamnoside Fragment of Rhamnogalacturonan-II Side Chain A Using a Late Stage Oxidation

Approach. J. Org. Chem. 2005, 70, 960–966.

(155) Chauvin, A.-L.; Nepogodiev, S. A.; Field, R. A. Synthesis of an Apiose-Containing

Disaccharide Fragment of Rhamnogalacturonan-II and Some Analogues. Carbohydr. Res.

2004, 339, 21–27.

(156) Crich, D.; Picione, J. Direct Synthesis of the β-L-Rhamnopyranosides. Org. Lett. 2003, 5, 781–

784.

(157) Hodosi, G.; Kovac, P. A Fundamentally New, Simple, Stereospecific Synthesis of

Oligosaccharides Containing the β-Mannopyranosyl and β-Rhamnopyranosyl Linkage. J. Am.

Chem. Soc. 1997, 119, 2335–2336.

(158) Yasomanee, J. P.; Demchenko, A. V. Effect of Remote Picolinyl and Picoloyl Substituents on

the Stereoselectivity of Chemical Glycosylation. J. Am. Chem. Soc. 2012, 134, 20097–20102.

(159) Mukherjee, A.; Palcic, M. M.; Hindsgaul, O. Synthesis and Enzymatic Evaluation of Modified

Acceptors of Recombinant Blood Group A and B Glycosyltransferases. Carbohydr. Res. 2000,

326, 1–21.

(160) Amer, H.; Hofinger, A.; Puchberger, M.; Kosma, P. Synthesis of O-Methylated Disaccharides

170

Related to Excretory/secretory Antigens of Toxocara Larvae. J. Carbohydr. Chem. 2001, 20,

719–731.

(161) Rao, Y.; Buskas, T.; Albert, A.; O’Neill, M. a; Hahn, M. G.; Boons, G.-J. Synthesis and

Immunological Properties of a Tetrasaccharide Portion of the B Side Chain of

Rhamnogalacturonan II (RG-II). Chembiochem 2008, 9, 381–388.

(162) Rao, Y.; Boons, G.-J. A Highly Convergent Chemical Synthesis of Conformational Epitopes

of Rhamnogalacturonan II. Angew. Chem. Int. Ed. Engl. 2007, 46, 6148–6151.

(163) Jones, N. A.; Nepogodiev, S. A.; MacDonal, C. J.; Hughes, D. L.; Field, R. A. Synthesis of the

Branched-Chain Sugar Aceric Acid: A Unique Component of the Pectic Polysaccharide

Rhamnogalacturonan-II. J. Org. Chem. Notes 2005, 70, 8556–8559.

(164) Timmer, M.; Stocker, B. L.; Seeberger, P. H. De Novo Synthesis of Aceric Acid and an Aceric

Acid Building Block. J. Org. Chem. Notes 2006, 71, 8294–8297.

(165) York, W. S.; Darvill, A. G.; McNeil, M.; Albersheim, P. 3-Deoxy-D-Manno-2-Octulosonic

Acid (KDO) Is a Component of Rhamnogalacturonan II, a Pectic Polysaccharide in the Primary

Cell Walls of Plants. Carbohydr. Res. 1985, 138, 109–126.

(166) Unger, F. M.; Stix, D.; Schulz, G. The Anomeric Configurations of the Two Ammonium

(Methyl 3-Deoxy-D-Manno-2-Octulopyranosid)onate Salts (Methyl α- and β-Ketopyranosides

of Kdo). Carbohydr. Res. 1980, 80, 191–195.

(167) van der Klein, P. A. M.; Boons, G. J. P. H.; Veeneman, G. H.; van der Marel, G. A.; van Boom,

J. H. An Efficient Route to 3-Deoxy-D-Manno-2-Octulosonic Acid (KDO) Derivatives via a

1,4-Cyclic Sulfate Approach. Tetrahedron Lett. 1989, 30, 5477–5480.

(168) Feng, Y.; Dong, J.; Xu, F.; Liu, A.; Wang, L.; Zhang, Q.; Chai, Y. Efficient Large Scale

Syntheses of 3-Deoxy-D-Manno-2-Octulosonic Acid (Kdo) and Its Derivatives. Org. Lett.

171

2015, 17, 2388–2391.

(169) Stevenson, T. T.; Darvill, A. G.; Albersheim, P. 3-Deoxy-D-Lyxo-2-Heptulosaric Acid, a

Component of the Plant Cell-Wall Polysaccharide Rhamnogalacturonan-II. Carbohydr. Res.

1988, 179, 269–288.

(170) Banaszek, A. The First Synthesis of Acid ( DHA ) Derivatives. Tetrahedron 1995, 51, 4231–

4238.

(171) Thomas, J. R.; Darvill, A. G.; Albersheim, P. Isolation and Structural Characterization of the

Pectic Polysaccharide Rhamnogalacturonan II from Walls of Suspension-Cultured Rice Cells.

Carbohydr. Res. 1989, 185, 261–277.

(172) Heredia, A.; Jiménez, A.; Guillén, R. Composition of Plant Cell Walls. Z. Lebensm. Unters.

Forsch. 1995, 200, 24–31.

(173) Nakamura, A.; Furuta, H.; Maeda, H.; Takao, T.; Negamatsu, Y. Structural Studies by Stepwise

Enzymatic Degradation of the Main Backbone of Soybean Soluble Polysaccharides Consisting

of Galacturonan and Rhamnogalacturonan. Biosci. Biotechnol. Biochem. 2002, 66, 1301–1313.

(174) Huisman, M. M. H.; Brüll, L. P.; Thomas-Oates, J. E.; Haverkamp, J.; Schols, H. A.; Voragen,

A. G. J. The Occurrence of Internal (1→5)-Linked Arabinofuranose and Arabinopyranose

Residues in Arabinogalactan Side Chains from Soybean Pectic Substances. Carbohydr. Res.

2001, 330, 103–114.

(175) Hinz, S. W. A.; Verhoef, R.; Schols, H. A.; Vincken, J. P.; Voragen, A. G. J. Type I

Arabinogalactan Contains β-D-Galp-(1→3)-β-D-Galp Structural Elements. Carbohydr. Res.

2005, 340, 2135–2143.

(176) Leivas, C. L.; Iacomini, M.; Cordeiro, L. M. C. Pectic Type II Arabinogalactans from Starfruit

(Averrhoa Carambola L.). Food Chem. 2016, 199, 252–257.

172

(177) El-Shenawy, H.; Schuerch, C. Synthesis and Characterization of Propyl O-β-D-

Galactopyranosyl-(1→4)-O-β-D-Galactopyranosyl-(1→4)-α-D-Galactopyranoside.

Carbohydr. Res. 1984, 131, 239–246.

(178) Oberthur, M.; Peters, S.; Kumar Saibal, D.; Lichtenthaler, F. W. Synthesis of Linear β-(1→4)-

Galacto Hexa- and Heptasaccharides and Studies Directed towards Cyclogalactans. Carbohydr.

Res. 2002, 337, 2171–2180.

(179) Andersen, M. C. F.; Kračun, S. K.; Rydahl, M. G.; Willats, W. G. T.; Clausen, M. H. Synthesis

of β-1,4-Linked Galactan Side-Chains of Rhamnogalacturonan I. Chem. - Eur. J. 2016, 22,

11543–11548.

(180) Sakamoto, T.; Ishimaru, M. Peculiarities and Applications of Galactanolytic Enzymes That Act

on Type I and II Arabinogalactans. Appl. Microbiol. Biotechnol. 2013, 97, 5201–5213.

(181) Bartetzko, M. P.; Schuhmacher, F.; Seeberger, P. H.; Pfrengle, F. Determining Substrate

Specificities of β-1,4-Endogalactanases Using Plant Arabinogalactan Oligosaccharides

Synthesized by Automated Glycan Assembly. J. Org. Chem. 2017, 82, 1842–1850.

(182) Kovác, P.; Taylor, R. B. Alternative Syntheses and Related NMR Studies of Precursors for

Internal β-D-Galactopyranosyl Residues in Oligosaccharides, Allowing Chain Extension at O-

4. Carbohydr. Res. 1987, 167, 153–173.

(183) Lichtenthaler, F. W.; Oberthur, M.; Peters, S. Directed and Efficient Syntheses of β-(1->4)-

Linked Galacto-Oligosaccharides. Eur. J. Org. Chem. 2001, 3849–3869.

(184) Bartetzko, M. P.; Schuhmacher, F.; Hahm, H. S.; Seeberger, P. H.; Pfrengle, F. Automated

Glycan Assembly of Oligosaccharides Related to Arabinogalactan Proteins. Org. Lett. 2015,

17, 4344–4347.

(185) Yang, F.; Du, Y. Synthesis of Oligosaccharide Derivatives Related to Those from Sanqi, a

173

Chinese Herbal Medicine from Panax Notoginseng. Carbohydr. Res. 2002, 337, 485–491.

(186) Ning, J.; Wang, H.; Yi, Y. A Simple Approach to 3,6-Branched Galacto-Oligosaccharides and

Its Application to the Syntheses of a Tetrasaccharide and a Hexasaccharide Related to the

Arabinogalactans (AGs). Tetrahedron Lett. 2002, 43, 7349–7352.

(187) Kaji, E.; Nishino, T.; Ishige, K.; Ohya, Y.; Shirai, Y. Regioselective Glycosylation of Fully

Unprotected Methyl Hexopyranosides by Means of Transient Masking of Hydroxy Groups

with Arylboronic Acids. Tetrahedron Lett. 2010, 51, 1570–1573.

(188) Timmers, C. M.; Wigchert, S. C. M.; Leeuwenburgh, M. A.; van der Marel, G. A.; van Boom,

J. H. Synthesis of Tetrameric Arabinogalactans Based on 1,2-Anhydrosugars. European J. Org.

Chem. 1998, 1, 91–97.

(189) Du, Y.; Pan, Q.; Kong, F. Synthesis of a Tetrasaccharide Representing a Minimal Epitope of

an Arabinogalactan. Carbohydr. Res. 2000, 323, 28–35.

(190) Ning, J.; Yi, Y.; Yao, Z. An Efficient Method for the Synthesis of 2,6-Branched Galacto-

Oligosaccharides and Its Applications to the Synthesis of Three Tetrasaccharides and a

Hexasaccharide Related to the Arabinogalactans (AGs). Synlett 2003, 14, 2208–2212.

(191) Csávás, M.; Borbás, A.; Szilágyi, L.; Lipták, A. Successful Combination of

(Methoxydimethyl)methyl (MIP) and (2-Naphthyl)methyl (NAP) Ethers for the Synthesis of

Arabinogalactan-Type Oligosaccharides. Synlett 2002, 6, 887–890.

(192) Fekete, A.; Borbás, A.; Antus, S.; Lipták, A. Synthesis of 3,6-Branched Arabinogalactan-Type

Tetra- and Hexasaccharides for Characterization of Monoclonal Antibodies. Carbohydr. Res.

2009, 344, 1434–1441.

(193) Liang, X.-Y.; Liu, Q.-W.; Bin, H.-C.; Yang, J.-S. One-Pot Synthesis of Branched

Oligosaccharides by Use of Galacto- and Mannopyranosyl Thioglycoside Diols as Key

174

Glycosylating Agents. Org. Biomol. Chem. 2013, 11, 3903–3917.

(194) Zeng, Y.; Li, A.; Kong, F. A Concise Synthesis of Arabinogalactan with β-(1→6)

Galactopyranose Backbone and α-(1→2) Arabinofuranose Side Chains. Tetrahedron Lett.

2003, 44, 8325–8329.

(195) Gu, G.; Yang, F.; Du, Y.; Kong, F. Synthesis of a Hexasaccharide That Relates to the

Arabinogalactan Epitope. Carbohydr. Res. 2001, 336, 99–106.

(196) Li, A.; Kong, F. Concise Syntheses of Arabinogalactans with β-(1→6)-Linked

Galactopyranose Backbones and α-(1→3)- and α-(1→2)-Linked Arabinofuranose Side Chains.

Bioorg. Med. Chem. 2005, 13, 839–853.

(197) Huang, H.; Han, L.; Lan, Y.-M.; Zhang, L.-L. One-Pot Synthesis of a 3,6-Branched

Hexaarabinogalactan Using Galactopyranosyl Thioglycoside Diol as a Key Glycosylating

Agent. J. Asian Nat. Prod. Res. 2014, 16, 640–647.

(198) Li, A.; Zeng, Y.; Kong, F. An Effective Synthesis of an Arabinogalactan with a β-(1→6)-

Linked Galactopyranose Backbone and α-(1→2) Arabinofuranose Side Chains. Carbohydr.

Res. 2004, 339, 673–681.

(199) Li, A.; Kong, F. Syntheses of Arabinogalactans Consisting of β-(1→6)-Linked D-

Galactopyranosyl Backbone and α-(1→3)-Linked L-Arabinofuranosyl Side Chains.

Carbohydr. Res. 2004, 339, 1847–1856.

(200) Vincken, J.-P.; Schols, H. A.; Oomen, R. J. F. J.; Mccann, M. C.; Ulvskov, P.; Voragen, A. G.

J.; Visser, R. G. F. If Homogalacturonan Were a Side Chain of Rhamnogalacturonan I .

Implications for Cell Wall Architecture. Plant Physiol. 2003, 132, 1781–1789.

(201) Sakamoto, T.; Sakai, T. Analysis of Structure of Sugar-Beet Pectin by Enzymatic Methods.

Phytochemistry 1995, 39, 821–823.

175

(202) Øbro, J.; Harholt, J.; Scheller, H. V.; Orfila, C. Rhamnogalacturonan I in Solanum Tuberosum

Tubers Contains Complex Arabinogalactan Structures. Phytochemistry 2004, 65, 1429–1438.

(203) Wu, Y.; Xiong, D.-C.; Chen, S.-C.; Wang, Y.-S.; Ye, X.-S. Total Synthesis of Mycobacterial

Arabinogalactan Containing 92 Monosaccharide Units. Nat. Commun. 2017, 8, 14851.

(204) Nepogodiev, S. A.; Backinowsky, L. V.; Kochetkov, N. K. No Title. Bioorg. Khim. 1986, 12,

940–946.

(205) Kaneko, S.; Kawabata, Y.; Ishii, T.; Gama, Y.; Kusakabe, I. The Core Trisaccharide of a-L-

Arabinofuranan : Synthesis of Methyl L-Arabinofuranoside. Carbohydr. Res. 1995, 268, 307–

311.

(206) Du, Y.; Pan, Q.; Kong, F. Regio- and Stereoselective Synthesis of 1-5-Linked a-L-

Arabinofuranosyl Oligosaccharides. Synlett 1999, 10, 1648–1650.

(207) Du, Y.; Pan, Q.; Kong, F. An Efficient and Concise Regioselective Synthesis of α-(1→5)-

Linked L-Arabinofuranosyl Oligosaccharides. Carbohydr. Res. 2000, 329, 17–24.

(208) Arndt, D.; Graffi, A. The Synthesis of P-Nitrophenyl 5-O-α-L-Arabinofuranosyl-α-L-

Arabinofuranoside. Carbohydr. Res. 1976, 48, 128–131.

(209) Kawabata, Y.; Kaneko, S.; Kusakabe, I.; Gama, Y. Synthesis of Regioisomeric Methyl a-L-

Arabinofuranobiosides. Carbohydr. Res. 1995, 267, 39–47.

(210) Deng, L.; Liu, X.; Liang, X.; Yang, J. Regioselective Glycosylation Method Using Partially

Protected Arabino- and Galactofuranosyl Thioglycosides as Key Glycosylating Substrates and

Its Application to One-Pot Synthesis of Oligofuranoses. J. Org. Chem. 2012, 77, 3025–3037.

(211) Yamagaki, T.; Maeda, M.; Kanazawa, K.; Ishizuka, Y.; Nakanishi, H. Structures of Caulerpa

Cell Wall Microfibril Xylan with Detection of β -1,3-Xylooligosaccharides as Revealed by

Matrix-Assisted Laser Desorption Ionization/Time of Flight/Mass Spectrometry. Biosci.

176

Biotechnol. Biochem. 1996, 60, 1222–1228.

(212) Umemoto, Y.; Shibata, T.; Araki, T. D-Xylose Isomerase from a Marine Bacterium, Vibrio Sp.

Strain XY-214, and D-Xylulose Production from β-1,3-Xylan. Mar. Biotechnol. 2012, 14, 10–

20.

(213) Chen, L.; Kong, F. An Efficient and Practical Synthesis of β-(1→3)-Linked

Xylooligosaccharides. Carbohydr. Res. 2002, 337, 2335–2341.

(214) Synytsya, A.; Copiková, J.; Kim, W. J.; Park, Y. I. Cell Wall Polysaccharides of Marine Algae.

In Springer Handbook of Marine Biotechnology; Se-Kwon, K., Ed.; 2015; pp 543–590.

(215) Ray, B. Polysaccharides from Enteromorpha Compressa: Isolation, Purification and Structural

Features. Carbohydr. Polym. 2006, 66, 408–416.

(216) Domozych, D. S.; Ciancia, M.; Fangel, J. U.; Mikkelsen, M. D.; Ulvskov, P.; Willats, W. G. T.

The Cell Walls of Green Algae: A Journey through Evolution and Diversity. Front. Plant Sci.

2012, 3, 1–7.

(217) Abouraïcha, E. F.; El Alaoui-Talibi, Z.; Tadlaoui-Ouafi, A.; El Boutachfaiti, R.; Petit, E.;

Douira, A.; Courtois, B.; Courtois, J.; El Modafar, C. Glucuronan and Oligoglucuronans

Isolated from Green Algae Activate Natural Defense Responses in Apple Fruit and Reduce

Postharvest Blue and Gray Mold Decay. J. Appl. Phycol. 2017, 29, 471–480.

(218) Redouan, E.; Cedric, D.; Emmanuel, P.; Mohamed, E. G.; Bernard, C.; Philippe, M.;

Cherkaoui, E. M.; Josiane, C. Improved Isolation of Glucuronan from Algae and the Production

of Glucuronic Acid Oligosaccharides Using a Glucuronan Lyase. Carbohydr. Res. 2009, 344,

1670–1675.

(219) Isogai, A.; Kato, Y. Preparation of Polyuronic Acid from Cellulose by TEMPO-Mediated

Oxidation. Cellulose 1998, 5, 153–164.

177

(220) Distantina, S.; Wiratni; Fahrurrozi, M.; Rochmadi. Carrageenan Properties Extracted From. Int.

J. Chem. Mol. Nucl. Mater. Metall. Eng. 2011, 5, 487–491.

(221) Hoffmann, R. A.; Gidley, M. J.; Cooke, D.; Frith, W. J. Effect of Isolation Procedures on the

Molecular Composition and Physical Properties of Eucheuma Cottonii Carrageenan. Food

Hydrocoll. 1995, 9, 281–289.

(222) Lahaye, M.; Rochas, C. Chemical Structure and Physico-Chemical Properties of Agar.

Hydrobiologia 1991, 221, 137–148.

(223) Perez Recalde, M.; Canelón, D. J.; Compagnone, R. S.; Matulewicz, M. C.; Cerezo, A. S.;

Ciancia, M. Carrageenan and Agaran Structures from the Red Seaweed Gymnogongrus Tenuis.

Carbohydr. Polym. 2016, 136, 1370–1378.

(224) Rees, D. A. Estimation of the Relative Amounts of Isomeric Sulphate Esters in Some Sulphated

Polysaccharides. J. Chem. Soc. 1961, 5168–5171.

(225) Rees, D. A.; Conway, E. The Structure and Biosynthesis of Porphyran: A Comparison of Some

Samples. Biochem. J. 1962, 84, 411–416.

(226) Haug, A.; Larsen, B.; Smidsrød, O. Uronic Acid Sequence in Alginate from Different Sources.

Carbohydr. Res. 1974, 32, 217–225.

(227) Draget, K. I.; Smidsrød, O.; Skjåk-Broek, G. Alginates from Algae. In Polysaccharides and

Polyamides in the Food Industry; Steinbuchel, A., Rhee, S. K., Eds.; 2005.

(228) Zhang, Q.; van Rijssel, E. R.; Walvoort, M. T. C.; Overkleeft, H. S.; van der Marel, G. A.;

Codée, J. D. C. Acceptor Reactivity in the Total Synthesis of Alginate Fragments Containing

α-L-Guluronic Acid and β-D-Mannuronic Acid. Angew. Chem. Int. Ed. 2015, 54, 7670–7673.

(229) van den Bos, L. J.; Dinkelaar, J.; Overkleeft, H. S.; van der Marel, G. a. Stereocontrolled

Synthesis of β-D-Mannuronic Acid Esters: Synthesis of an Alginate Trisaccharide. J. Am.

178

Chem. Soc. 2006, 128, 13066–13067.

(230) Codée, J. D. C.; van den Bos, L. J.; de Jong, A.-R.; Dinkelaar, J.; Lodder, G.; Overkleeft, H.

S.; van der Marel, G. A. The Stereodirecting Effect of the Glycosyl C5-Carboxylate Ester:

Stereoselective Synthesis of β-Mannuronic Acid Alginates. J. Org. Chem. 2009, 74, 38–47.

(231) Walvoort, M. T. C.; van den Elst, H.; Plante, O. J.; Kröck, L.; Seeberger, P. H.; Overkleeft, H.

S.; van der Marel, G. A.; Codée, J. D. C. Automated Solid-Phase Synthesis of β-Mannuronic

Acid Alginates. Angew. Chem. Int. Ed. 2012, 51, 4393–4396.

(232) Dinkelaar, J.; Van Den Bos, L. J.; Hogendorf, W. F. J.; Lodder, G.; Overkleeft, H. S.; Codée,

J. D. C.; Van Der Marel, G. A. Stereoselective Synthesis of L-Guluronic Acid Alginates. Chem.

Eur. J. 2008, 14, 9400–9411.

(233) Chi, F. C.; Kulkarni, S. S.; Zulueta, M. M. L.; Hung, S. C. Synthesis of Alginate

Oligosaccharides Containing L-Guluronic Acids. Chem. Asian J. 2009, 4, 386–390.

(234) Garcia, B. A.; Poole, J. L.; Gin, D. Y. Direct Glycosylations with 1-Hydroxy Glycosyl Donors

Using Trifluoromethanesulfonic Anhydride and Diphenyl Sulfoxide. J. Am. Chem. Soc. 1997,

119, 7597–7598.

(235) Zhang, Q.; van Rijssel, E. R.; Overkleeft, H. S.; van der Marel, G. A.; Codée, J. D. C. On the

Reactivity of Gulose and Guluronic Acid Building Blocks in the Context of Alginate Assembly.

European J. Org. Chem. 2016, 14, 2393–2397.

(236) Kylin, H. Zur Biochemie Der Meeresalgen. Physiol. Chemie 1913, 83, 171–197.

(237) Berteau, O.; Mulloy, B. Sulfated Fucans, Fresh Perspectives: Structures, Functions, and

Biological Properties of Sulfated Fucans and an Overview of Enzymes Active toward This

Class of Polysaccharide. Glycobiology 2003, 13, 29–40.

(238) Ustyuzhanina, N.; Krylov, V.; Grachev, A.; Gerbst, A.; Nifantiev, N. Synthesis, NMR and

179

Conformational Studies of Fucoidan Fragments, 8: Convergent Synthesis of Branched and

Linear Oligosaccharides. Synthesis (Stuttg). 2006, 23, 4017–4031.

(239) Chargaff, E.; Bancroft, F. W.; Stanley-brown, M. Studies on the Chemistry of Blood

Coagulation. J. Biol. 1936, 4, 155–161.

(240) Gerbst, A. G.; Ustuzhanina, N. E.; Grachev, A. A.; Khatuntseva, E. A.; Tsvetkov, D. E.;

Shashkov, A. S.; Usov, A. I.; Preobrazhenskaya, M. E.; Ushakova, N. A.; Nifantiev, N. E.

Synthesis, NMR and Conformational Studies of Fucoidan Fragments. V. Linear 4,4′,4″‐Tri‐O‐

Sulfated and Parent Non‐sulfated (1→3)‐Fucotrioside Fragments. J. Carbohydr. Chem. 2003,

22, 109–122.

(241) Krylov, V. B.; Kaskova, Z. M.; Vinnitskiy, D. Z.; Ustyuzhanina, N. E.; Grachev, A. A.;

Chizhov, A. O.; Nifantiev, N. E. Acid-Promoted Synthesis of per-O-Sulfated

Fucooligosaccharides Related to Fucoidan Fragments. Carbohydr. Res. 2011, 346, 540–550.

(242) Vinnitskiy, D. Z.; Krylov, V. B.; Ustyuzhanina, N. E.; Dmitrenok, A. S.; Nifantiev, N. E. The

Synthesis of Heterosaccharides Related to the Fucoidan from Chordaria Flagelliformis Bearing

an α-L-Fucofuranosyl Unit. Org. Biomol. Chem. 2016, 14, 598–611.

(243) Hua, Y.; Gu, G.; Du, Y. Synthesis and Biological Activities of Octyl 2,3,4-Tri-O-Sulfo-α-L-

Fucopyranosyl-(1→3)-2,4-Di-O-Sulfo-α-L-Fucopyranosyl-(1→3)-2,4-Di-O-Sulfo-α-L-

Fucopyranosyl-(1→3)-2,4-Di-O-Sulfo-β-L-Fucopyranoside. Carbohydr. Res. 2004, 339, 867–

872.

(244) Kasai, A.; Arafuka, S.; Koshiba, N.; Takahashi, D.; Toshima, K. Systematic Synthesis of Low-

Molecular Weight Fucoidan Derivatives and Their Effect on Cancer Cells. Org. Biomol. Chem.

2015, 13, 10556–10568.

180

(245) Zong, C.; Li, Z.; Sun, T.; Wang, P.; Ding, N.; Li, Y. Convenient Synthesis of Sulfated

Oligofucosides. Carbohydr. Res. 2010, 345, 1522–1532.

(246) Arafuka, S.; Koshiba, N.; Takahashi, D.; Toshima, K. Systematic Synthesis of Sulfated

Oligofucosides and Their Effect on Breast Cancer MCF-7 Cells. Chem. Commun. 2014, 50,

9831–9834.

(247) Gerbst, A. G.; Ustuzhanina, N. E.; Grachev, A. A.; Zlotina, N. S.; Khatuntseva, E. A.; Tsvetkov,

D. E.; Shashkov, A. S.; Usov, A. I.; Nifantiev, N. E. Synthesis, NMR, and Conformational

Studies of Fucoidan Fragments 4: 4-Mono-and 4,4’-disulfated (1→3)-α-L-Fucobioside and 4-

Sulfated Fucoside Fragments. J. Carbohydr. Chem. 2002, 21, 313–324.

(248) Gerbst, A. G.; Ustuzhanina, N. E.; Grachev, A. A.; Khatuntseva, E. A.; Tsvetkov, D. E.;

Whitfield, D. M.; Berces, A.; Nifantiev, N. E. Synthesis, NMR, and Conformational Studies of

Fucoidan Fragments. III. Effect of Benzoyl Group At O-3 on Stereoselectivity of Glycosylation

by 3-O- and 3,4-Di-O-Benzoylated 2-O-Benzylfucosyl Bromides. J. Carbohydr. Chem. 2001,

20, 821–831.

(249) Krylov, V. B.; Ustyuzhanina, N. E.; Grachev, A. A.; Nifantiev, N. E. Efficient Acid-Promoted

per-O-Sulfation of Organic Polyols. Tetrahedron Lett. 2008, 49, 5877–5879.

(250) Krylov, V. B.; Argunov, D. A.; Vinnitskiy, D. Z.; Verkhnyatskaya, S. A.; Gerbst, A. G.;

Ustyuzhanina, N. E.; Dmitrenok, A. S.; Huebner, J.; Holst, O.; Siebert, H.-C.; et al. Pyranoside-

into-Furanoside Rearrangement: New Reaction in Carbohydrate Chemistry and Its Application

in Oligosaccharide Synthesis. Chem. - A Eur. J. 2014, 20, 16516–16522.

(251) Ustyuzhanina, N. E.; Fomitskaya, P. A.; Gerbst, A. G.; Dmitrenok, A. S.; Nifantiev, N. E.

Synthesis of the Oligosaccharides Related to Branching Sites of Fucosylated Chondroitin

Sulfates from Sea Cucumbers. Mar. Drugs 2015, 13, 770–787.

181

(252) Bilan, M. I.; Vinogradova, E. V.; Tsvetkova, E. A.; Grachev, A. A.; Shashkov, A. S.; Nifantiev,

N. E.; Usov, A. I. A Sulfated Glucuronofucan Containing Both Fucofuranose and

Fucopyranose Residues from the Brown Alga Chordaria Flagelliformis. Carbohydr. Res. 2008,

343, 2605–2612.

(253) Stadnik, M. J.; De Freitas, M. B. Algal Polysaccharides as Source of Plant Resistance Inducers.

Trop. Plant Pathol. 2014, 39, 111–118.

(254) Lahaye, M.; Robic, A. Structure and Function Properties of Ulvan, a Polysaccharide from

Green Seaweeds. Biomacromolecules 2007, 8, 1765–1774.

(255) Robic, A.; Gaillard, C.; Sassi, J. F.; Leral, Y.; Lahaye, M. Ultrastructure of Ulvan: A

Polysaccharide from Green Seaweeds. Biopolymers 2009, 91, 652–664.

182

TOC Graphic

OH OH O HO O OH HO O OH HO HO OH HO OH OH OH

OH HO OH Me O Chemical OH O synthesis OH HO HO OH OH

S X F G

183