Synthesis of a backbone hexasaccharide fragment of the pectic polysaccharide rhamnogalacturonan i

Article abstract of DOI:10.1021/ol400430p

Synthesis of the fully unprotected hexasaccharide backbone of the pectic polysaccharide rhamnogalacturonan I is described. The strategy relies on iterative coupling of a common pentenyl disaccharide glycosyl donor followed by a late-stage oxidation of the C-6 positions of the galactose residues. The disaccharide donor is prepared by an efficient chemoselective armed-disarmed coupling of a thiophenyl rhamnoside donor with a pentenyl galactoside acceptor bearing the strongly electron-withdrawing pentafluorobenzoyl ester (PFBz) protective group.

Full text of DOI:10.1021/ol400430p

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Synthesis of a Backbone Hexasaccharide
Fragment of the Pectic Polysaccharide
Rhamnogalacturonan I
Alexandra N. Zakharova,‡ Robert Madsen,‡ and Mads H. Clausen*,†,‡
Center for Nanomedicine and Theranostics and Department of Chemistry,
Technical University of Denmark, Kemitorvet, Building 207, DK-2800 Kgs.
Lyngby, Denmark
mhc@kemi.dtu.dk
Received February 14, 2013
ABSTRACT
Synthesis of the fully unprotected hexasaccharide backbone of the pectic polysaccharide rhamnogalacturonan I is described. The strategy relies
on iterative coupling of a common pentenyl disaccharide glycosyl donor followed by a late-stage oxidation of the C-6 positions of the galactose
residues. The disaccharide donor is prepared by an efficient chemoselective armed-disarmed coupling of a thiophenyl rhamnoside donor with a
pentenyl galactoside acceptor bearing the strongly electron-withdrawing pentafluorobenzoyl ester (PFBz) protective group.
Pectins are highly heterogeneous polysaccharides of plant
origin. They are found in the primary cell wall and con-
tribute to various cell functions, including support, defense,
signaling, and cell adhesion.1 Pectins also play important
roles as food additives, serving as stabilizing and thickening
agents in products such as jams, yogurts, and jellies.2
Rhamnogalacturonan I (RG-I) is one of the structural
classes of pectic polysaccharides, along with homogalactur-
onan, rhamnogalacturonan II, and xylogalacturonan.3
The chemical structure of RG-I is complex having a
backbone consisting of alternating R-linked L-rhamnose
and D-galacturonic acid units with numerous branches of
arabinans, galactans, or arabinogalactans positioned at
C-4 of the rhamnose residues.
The structural complexity of pectin together with the
wide range of its practical applications and desire to
understand its structure and functions in details have
inspired many researchers to pursue chemical syntheses
of pectic oligosaccharides. Herein, we report the synthesis
of a hexasaccharide fragment of the RG-I rhamnogalac-
turonan backbone (1, Figure 1).
Synthesis of the fully unprotected hexasaccharide frag-
ment of RG-I has not been previously reported. However,
smaller fully and partially unprotected RG-I oligosaccha-
rides, as well as fully protected oligosaccharides up to
‡ Department of Chemistry.
† Center for Nanomedicine and Theranostics.
(1) Mohnen, D. Curr. Opin. Plant Biol. 2008, 11, 266–277. Harholt,
J.; Suttangkakul, A.; Scheller, H. V. Plant Physiol. 2010, 153, 384–395.
(2) Willats, W. G. T.; Knox, J. P.; Mikkelsen, J. D. Trends Food Sci.
Technol. 2006, 17, 97–104.
(3) Caffall, K. H.; Mohnen, D. Carbohydr. Res. 2009, 344, 1879–
1900.
r
10.1021/ol400430p
XXXX American Chemical Society

the fully unprotected RG-I tetrasaccharide and its di-
methyl ester. Interestingly, the initial attempt to couple a
galactorhamnosyl disaccharide donor to the galactose of a
disaccharide acceptor failed due to a lack of reactivity,
forcing the authors to change the strategyand assemble the
RG-I tetrasaccharide through galactosylation instead of
rhamnosylation. The potential of this methodology for
iterative elongation of the oligosaccharide chain was de-
monstrated by preparation of a fully protected analog of
the native hexasaccharide, containing both galactose and
galacturonic acid residues.
Figure 1. Structure of the hexasaccharide fragment of RG-I.
hexamers, have been prepared by different approaches.
Someof thestrategiesused galacturonic acidasthestarting
material, while others favored the oxidation of galactose to
galacturonic acid at a late stage, i.e., pre- and postglyco-
sylationꢀoxidation strategies, respectively. Reimer and
co-workers4 reported the synthesis of a protected tetra-
saccharide containing galactose instead of galacturonic
acid as an intermediate for the preparation of RG-I
fragments. The protective group pattern was designed to
allow for further chain elongation and introduction of
branching. It was envisioned that the global deprotection
and oxidation of the primary hydroxyl groups of the
galactose units would furnish the native oligosaccharides.
In later work, the group synthesized the fully unprotected
methyl glycoside of an RG-I tetrasaccharide, both in the
methyl ester and the free carboxylic acid forms.5 In this
case, a similar protective group pattern was used, but
galacturonic acid was employed from the early stages. This
lowered the overall number of synthetic steps by avoiding
the late stage oxidation. Unfortunately, the key glycosyla-
tion reaction proved to be problematic and only low yields
of the protected tetrasaccharide product could be ob-
tained. Vogel and co-workers6 prepared a partially depro-
tected RG-I trisaccharide bearing a benzoyl group at C-4
of the rhamnose residue where galacturonic acid was used
as a starting material. Later, the same group reported the
synthesis of the fully unprotected propyl glycoside of an
RG-I tetrasaccharide, as well as synthesis of its protected
hexasaccharide fragment and protected tri- and tetrasac-
charides suitable for the assembly of the branched RG-I
fragments.7 The synthesis was based on a modular design
principle and used galacturonic acid as the starting materi-
al. Takeda and co-workers prepared8 the unprotected
propyl glycoside of an RG-I tetrasaccharide using a late-
stage oxidation approach. All the mentioned work em-
ployed the generation of glycosyl donors before each
glycosylation step. In a recent report by Davis and co-
workers,9 a latent-active approach was utilized and com-
bined with the late-stage oxidation strategy to synthesize
Retrosynthetic analysis of the target RG-I hexasaccha-
ride 1 is depicted in Figure 2.
Figure 2. Retrosynthesis of the RG-I hexasaccharide 1.
Choosing between the two possible approaches10 for
synthesis of oligosaccharides containing uronic acids (that
is, oxidation prior to or after glycosylation), we adopted
the postglycosylation strategy, which we had previously
successfully employed11 in the synthesis of homogalactur-
onans. Although it requires additional protective group
manipulations, the nonoxidized carbohydrates are gener-
ally more reactive glycosyl donors than their oxidized
counterparts,12 where the reactivity is decreased by the
presence of the electron-withdrawing carboxyl groups.
Moreover, introduction of the carboxylic acid functional-
ity ata latestage of the synthesisreducesthe riskof possible
side reactions, such as epimerization to L-altruronic acid
and β-elimination leading to the formation of 4-deoxy-L-
threo-hex-4-enopyranuronic acid. According to this rea-
soning, we envisioned that the target hexasaccharide 1
could be obtained from the partially deprotected hexasac-
charide 2 by oxidation of the primary hydroxyl groups to
the carboxylic acids followed by a global deprotection.
ꢀ
(10) van den Bos, L. J.; Codee, J. D. C.; Litjens, R. E. J. N.;
Dinkelaar, J.; Overkleeft, H. S.; van der Marel, G. A. Eur. J. Org. Chem.
2007, 3963–3976.
(11) (a) Clausen, M. H.; Jørgensen, M. R.; Thorsen, J.; Madsen, R.
J. Chem. Soc., Perkin Trans. 1 2001, 543–551. (b) Clausen, M. H.;
Madsen, R. Chem.;Eur. J. 2003, 9, 3821–3832.
(4) Rich, J. R.; Mcgavin, R. S.; Gardner, R.; Reimer, K. B. Tetra-
hedron: Asymmetry 1999, 10, 17–20.
(5) Reiffarth, D.; Reimer, K. B. Carbohydr. Res. 2008, 343, 179–188.
(6) Nolting, B.; Boye, H.; Vogel, C. J. Carbohydr. Chem. 2000, 19,
923–938.
ꢀ
(12) Codee, J. D. C.; Christina, A. E.; Walvoort, M. T. C.; Overkleeft,
(7) Nemati, N.; Karapetyan, G.; Nolting, B.; Endress, H.-U.; Vogel,
C. Carbohydr. Res. 2008, 343, 1730–1742.
(8) Maruyama, M.; Takeda, T.; Shimizu, N.; Hada, N.; Yamada, H.
H. S.; van der Marel, G. a. Top. Curr. Chem. 2011, 301, 253–289.
Walvoort, M. T. C.; de Witte, W.; van Dijk, J.; Dinkelaar, J.; Lodder,
ꢀ
G.; Overkleeft, H. S.; Codee, J. D. C.; van der Marel, G. A. Org. Lett.
Carbohydr. Res. 2000, 325, 83–92.
(9) Scanlan, E. M.; Mackeen, M. M.; Wormald, M. R.; Davis, B. G.
J. Am. Chem. Soc. 2010, 132, 7238–7239.
2011, 13, 4360–4363. de Jong, A.-R.; Hagen, B.; van der Ark, V.;
ꢀ
Overkleeft, H. S.; Codee, J. D. C.; Van der Marel, G. A. J. Org. Chem.
2012, 77, 108–25.
B
Org. Lett., Vol. XX, No. XX, XXXX

Hexasaccharide 2 in turn was planned to be assembled by
two iterative glycosylations using disaccharide 3. Employ-
ing the common disaccharide 3 would minimize the num-
ber of monosaccharide building blocks required for the
synthesis. Donor 3 was designed to possess a nonpartici-
pating benzyl (Bn) ether atC-2 promotingthe formation of
the R-glycosidic linkage and was intended to be produced
through a chemoselective coupling between rhamnose
donor 4 and galactose acceptor 5. Donor 4 was designed
to carry a nonparticipating 2-naphthylmethyl (NAP) ether
at C-2 ensuring the formation of the R-glycosidic linkage
and later on allowing for selective deprotection and elon-
gation of the oligosaccharide chain at this position. The
C-6 moiety in acceptor 5 was capped with a pentafluoro-
benzoyl ester (PFBz) that could be selectively removed to
release this position for oxidation. Apart from functioning
as a temporary protective group, the PFBz ester was also
envisioned to tune the reactivity of 5. It is known that
electron-withdrawing protective groups decrease the reac-
tivity of glycosyl donors,13 and the donors protected with
electron-donating (ether) groups can be selectively activated
in a glycosylation reaction over the donors protected with
electron-withdrawing (ester) groups. This phenomenon,
first formulated by Fraser-Reid,14 is known as the “armed-
disarmed effect”. In the present synthesis, the “armed”
rhamnose donor 4 fully protected with ether groups was
planned to be selectively activated over the “disarmed”
galactose acceptor 5 bearing an electron-withdrawing PFBz
group. In addition to the electronic effects of the protective
groups, rhamnose was expected to have a higher reactivity,
because it is a deoxy sugar and lacks the electron-with-
drawing group at C-6 compared to galactose.13
acceptor 5b with the identical rhamnose donor 4 produced
the desired disaccharide 3b as the sole product, and we
isolated the R-anomer in 78% yield after flash chromato-
graphy. As an alternative to the armedꢀdisarmed approach
that we describe here, we also explored selective activation of
the thiophenyl glycoside with other promotors: MeOTf17
resulted in a low yielding glycosylation with many byprod-
ucts while activation with NIS/Yb(OTf)318 mainly led to the
formation of the C-glycoside 6 (Scheme 1). Attempts to
preactivate the glycosyl donor with diphenyl sulfoxide and
triflic anhydride19 also resulted in the formation of 6 as the
major product. This could be circumvented by replacing the
O-2 NAP protective group with chloroacetyl and with that
thioglycoside donor the preactivation conditions gave a
coupling yield of 45%. Nonetheless, the armedꢀdisarmed
coupling of 4 and 5b resulted in the highest yield; the
reactivity difference between the thiophenyl glycoside and
the corresponding pentenyl glycosides was somewhat sur-
prising, and we are currently investigating whether this is a
general trend.
Scheme 1. Synthesis of the Disaccharide Building Blocks
Protected monosaccharide building blocks 4,15 5a,16 and
5b11b were synthesized from L-rhamnose and D-galactose
(see the Supporting Information). Initially, the use of galac-
tose thiophenyl acceptor 5a in a chemoselective coupling
with rhamnose thiophenyl donor 4 was explored. When N-
iodosuccinimide (NIS) in the presence of a catalytic amount
of triethylsilyl trifluoromethanesulfonate (TESOTf) was
applied as the promoter it was possible to obtain the target
disaccharide 3a but, unfortunately, only as an inseparable
mixture in almost equal amounts with a trisaccharide by-
product derived from a reaction of 3a with 4. Attempts to
conduct this glycosylation under different reaction condi-
tions (applying I2 as the promoter, converting 4 into the
corresponding glycosyl bromide and subsequent activation
with silver triflate, or applying in situ anomerisation
conditions) did not improve the reaction outcome. In some
cases, most of donor 4 was converted into a C-glycoside
through an intramolecular reaction (vide infra). Given the
lack of success in synthesizing thiophenyl disaccharide 3a,
we turned to pentenyl glycosides as an alternative (Scheme 1).
The NIS/TESOTf-mediated coupling of galactose pentenyl
To assemble the hexasaccharide from the disaccharide
3b, it was first converted to the glycosyl bromide and then,
by glycosylation of benzyl alcohol, to benzyl glycoside 7.
This two-step sequence ensured the formation of the
R-glycoside, where direct activation with NIS/TESOTf
resulted in an R/β-mixture. This was followed by removal
of the NAP-group at C-20 by oxidation with DDQ in
CH2Cl2 in the presence of water furnishing disaccharide
acceptor 8. Pentenyl disaccharide 3b was used as the key
disaccharide donor in the further iterative assembly of
hexasaccharide 2 (Scheme 2). Glycosylation of 8 with 3b
using the aforementioned conditions led to the formation
of tetrasaccharide 9 as a single R-isomer in 71% yield.
Tetrasaccharide 9 was subjected to the same procedure for
removal of the NAP-group with DDQ to furnish tetra-
saccharide acceptor 10, which was coupled again with
donor 3b and the crude product was directly subjected to
ꢀ
the Zemplen conditions, and after the selective removal of
(17) Demchenko, A. V.; Meo, C. D. Tetrahedron Lett. 2002, 43,
8819–8822.
(13) Zhang, Z.; Ollmann, I. R.; Ye, X.-S.; Wischnat, R.; Baasov, T.;
Wong, C.-H. J. Am. Chem. Soc. 1999, 121, 734–753.
(14) Fraser-Reid, B.; Wu, Z.; Udodong, U. E.; Ottosson, H. J. Org.
(18) Jayaprakash, K. N.; Fraser-Reid, B. Org. Lett. 2004, 6, 4211–
4214. Jayaprakash, K. N.; Chaudhuri, S. R.; Murty, C. V. S. R.;
Fraser-Reid, B. J. Org. Chem. 2007, 72, 5534–5545.
Chem. 1990, 55, 6068–6070.
ꢀ
(19) Codee, J. D. C.; Litjens, R. E. J. N.; den Heeten, R.; Overkleeft,
(15) Crich, D.; Vinogradova, O. J. Org. Chem. 2007, 72, 3581–3584.
(16) Schmidt, T. H.; Madsen, R. Eur. J. Org. Chem. 2007, 3935–3941.
H. S.; van Boom, J. H.; van der Marel, G. A. Org. Lett. 2003, 5, 1519–
1522.
Org. Lett., Vol. XX, No. XX, XXXX
C

Scheme 2. Assembly of the Hexasaccharide
the PFBz-groups at C-6 of galactose, the hexasaccharide 2
was isolated in a pure form in 40% yield over two steps. The
liberated primary alcohols were oxidized with DessꢀMartin
periodinane to aldehydes and then with NaClO2 to car-
boxylic acids. The carboxylic acid functionalities were
protected as benzyl esters by reaction with PhCHN2 to
facilitate purification. Finally, treatment of 11 under stan-
dard conditions for catalytic hydrogenolysis allowed re-
moval of all the benzyl groups as well as the NAP group,
furnishing the fully unprotected hexasaccharide 1.
with galactan and arabinan, which will be the focus of future
efforts, in addition to using hexasaccharide 1 in character-
ization of enzymes acting on RG-I.
Acknowledgment. This work was partially financed by
the EU 7th Framework Programme via the Marie Curie
Initial Training Network, LeanGreenFood. We thank
Mr. Clive Phipps Walter, Novozymes, and Dr. Kok-Phen
Yan, University of Southern Denmark, for MALDI-MS
analysis.
In summary, we have presented the first successful synth-
esis of a fully unprotected hexasaccharide RG-I fragment
employing a highly modular synthesis that takes advantage
of the armedꢀdisarmed effect to generate the key disaccha-
ride donor in a chemoselective fashion. We envision that this
flexible strategy allows for easy introduction of side chains
Supporting Information Available. Experimental pro-
cedures and analytical and spectral data, including copies
of the NMR spectra. This material is available free of
charge via the Internet at http://pubs.acs.org.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX