Synthesis of structure-defined branched hyaluronan tetrasaccharide glycoclusters

Article abstract of DOI:10.1016/j.tetlet.2017.06.032

Hyaluronan is a glycosaminoglycan with a large number of biological activity. Hyaluronan of different molecular weight often shows different biological activity, sometimes even completely opposite, but the mechanism is not clear. Herein, the hyaluronan tetrasaccharide glycoclusters using hyaluronan tetrasaccharide obtained by enzymolysis of natural hyaluronan were firstly synthesized in high yield. The structurally determined and diverse glycoclusters were of wide molecular weight range and might be used for mimicking the biological activity of natural hyaluronan and facilitating the mechanism study.

Full text of DOI:10.1016/j.tetlet.2017.06.032

Accepted Manuscript
Synthesis of structure-defined branched hyaluronan tetrasaccharide glycoclus-
ters
Meng Sha, Wang Yao, Xiao Zhang, Zhongjun Li
PII:
S0040-4039(17)30757-8
DOI:
http://dx.doi.org/10.1016/j.tetlet.2017.06.032
TETL 49024
Reference:
To appear in:
Tetrahedron Letters
Received Date:
Revised Date:
Accepted Date:
26 April 2017
10 June 2017
12 June 2017
Please cite this article as: Sha, M., Yao, W., Zhang, X., Li, Z., Synthesis of structure-defined branched hyaluronan
tetrasaccharide glycoclusters, Tetrahedron Letters (2017), doi: http://dx.doi.org/10.1016/j.tetlet.2017.06.032
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Synthesis of structure-defined branched
hyaluronan tetrasaccharide glycoclusters
Leave this area blank for abstract info.
Meng Sha, Wang Yao, Xiao Zhang, and Zhong-jun Li

1
Tetrahedron Letters
Synthesis of structure-defined branched hyaluronan tetrasaccharide glycoclusters
a
a
a
a
Meng Sha , Wang Yao , Xiao Zhang and Zhongjun Li 
a
State eyꢀaboratoryoꢁꢂaturalaꢃndiomimeticꢄrugs;ꢄepartmeꢃtoꢁꢅhemicaldiology,SchooloꢁꢆharmaceuticalScieꢃces,ꢆekiꢃgꢇꢃiversity,deijiꢃg
1
00191,ꢅhiꢃa.
ARTICLE INFO
ABSTRACT
Articlehistory:
Received
Received in revised form
Accepted
Hyaluronan is a glycosaminoglycan with a large number of biological activity. Hyaluronan of
different molecular weight often shows different biological activity, sometimes even completely
opposite, but the mechanism is not clear. Herein, the hyaluronan tetrasaccharide glycoclusters
using hyaluronan tetrasaccharide obtained by enzymolysis of natural hyaluronan were firstly
synthesized in high yield. The structurally determined and diverse glycoclusters were of wide
molecular weight range and might be used for mimicking the biological activity of natural
hyaluronan and facilitating the mechanism study.
Available online
eyworns:
Hyaluronan
Hyaluronidase
2
009 Elsevier Ltd. All rights reserved.
Glycocluster
Glycomimetic
Large-scale enzymatic production
The hyaluronan (HA) is a linear and unbranched polymer
weight hyaluronan can bind to different numbers of receptors,
7
constituted by disaccharide units of
D-glucuronic acid (GlcA)
which is called the multivalent interactions , and can yield
8, 9,10
linked to ꢂ-acetyl- -glucosamine (GlcNAc) with a β (1→3)
D
different processes and biological responses
. However, there
linkage between GlcA and GlcNAc and a bond β (1→4) between
GlcNAc and GlcA, as showed in Fig 1. Hyaluronan is
synthesized by hyaluronan synthases iꢃv ivo and could be
is no clear and precise relationship between the bioactivity of
hyaluronan and its molecular weight, and the reason was
reviewed by Robert Stern . Especially for hyaluronan with
5
1
degraded by several kinds of hyaluronidases (Hyals) . As a result,
moderate molecular weight (several tens to hundreds of
the natural hyaluronan has a wide range of molecular weight
distribution. The hyaluronan with the molecular weight more
than 100 kDa is called high molecular weight hyaluronan
disaccharides units), the literature often reports completely
opposite results . A major underlying reason is the difficulty of
5
gaining the HA materials that are homogenous in molecular
weight, which is limited by the extraction and purification
techniques. Recently, glycoclusters have been artificially
synthesized to mimic the multivalent interactions of natural
(HMW-HA), about tens kDa is called low molecular weight
hyaluronan (LMW-HA), and a few kDa is called hyaluronan
2
oligosaccharides (o-HA) .
11
carbohydrates and proteins . Thus, we envisioned that o-HA
based glycoclusters was quite a suitable molecules to solve the
challenges above. Though synthetic glycoconjugates generally
would not reproduce exactly the valency and topology of the
natural ligands, the binding properties obtained in most cases are
Fig 1 The structure of hyaluronan.
7
Hyaluronan could interacts with many proteins, which are
termed hyaladherins (often called HA binding proteins, HABPs)
and most of them are cell-surface receptors .The hyaladherins
contain lots of proteins, mainly are CD44, Receptor for
Hyaluronan Mediated Motility (RHAMM, also known as
CD168), Hyaluronan Receptor for Endocytosis (HARE),
Lymphatic Vessel Endothelial hyaluronan receptor 1 (LYVE 1,
in line with possible applications iꢃvivoand as therapeutics .
3
Herein, we present the synthesis of branched hyaluronan
tetrasaccharide (HA4) glycoclusters based on copper-catalyzed
Huisgen 1, 3 cycloaddition (CuAAC) reaction. The products
obtained have a wide spectrum of molecular weight ranging from
3
kDa to 30 kDa.
also known as Cell Surface Retention Sequence Binding Protein-
In order to obtain the structure of the determination and
4
1, CRSBP1) and Toll Like Receptors (TLRs) . Hyaluronan
diversity of branched o-HA clusters, three aspects need to be
considered. Firstly was the technic to afford a large amount of o-
regulates an array of biological and pathological processes. Also,
hyaluronan have extraordinarily wide-ranging and sometimes
12
HA to assemble the glycoclusters . Secondly, branched
molecules with multiple identical modifiable sites should be
chosen as the “scaffold” of the glycoclusters. Also chemical
5, 6
opposing biological functions related with molecular weight
.
The possible explanation for those is that different molecular
—
——

Corresponding author. E-mail: zjli@bjmu.edu.cn (Z-J. Li).

2
Tetrahedron Letters
Scheme 1. Synthesis of HA4 building blocks 4-6. (a) BTH, 37 °C, pH=5.0; (b) AcCl, MeOH, 4 °C; (c) Ac
for a-c. (d) DMAPA, dry THF, 0 °C →rt, ; (e) CNCCl , DBU, CH Cl , 0 °C →rt, 84.9% for d-e; (f) TMSOTf, CH
1.5%. (g) propynol, CuCl , CHCl , reflux, 88.1% . (h) LiOH(aq.) / H , THF, 0 °C→rt, then 4M NaOH, MeOH 81.4% for 5
, CHCl , reflux, 88.1%; (j)TsCl, Et N, CH Cl , 0 °C →rt; (k) NaN , DMF, 65 °C, 64.1% for two
2
O, Pyridine, rt, 42%
3
2
2
2
2
Cl , 0 °C →rt,
8
2
3
2 2
O
and 92.1 % for 6; (i) 7, CuCl
2
3
3
2
2
3
steps.
reactions with high efficiency was needed to connect o-HA to the
scaffolds.
hydroxyl was then converted to trichloroacetimidate 2. After
2
treating with catalytic amount of TMSOTf in cold CH Cl ,
2
oxazoline 3 was obtained in 85% from 1. As shown in Scheme 1,
the successfully glycosylation occurred when oxazoline 3 was
refluxed with linker 7 in chloroform under the catalytic of copper
chloride that yielded the desired building blocks 4 in 88%. In
order to obtain a alkynyl linked tetrasaccharide, 3 was treated
Several approaches towards o-HA such as chemical synthesis
13-16
17
strategy
and enzymatic synthesis strategy have been
reported. Also, the direct digestion of HMW-HA by the
1
8, 19
hyaluronidases was widely used
. In this paper, bovine
19
testicular hyaluronidases (BTH) was used to gain o-HA . As HA
disaccharide and its derivatives exhibit little biological
activities , we focused on HA4, which was the smallest unit with
with propargyl alcohol under the same CuCl
2
catalyzed
20
glycosylation reaction, and then saponification, which yield
compound 5 in 81% yield.
5
biological activity . It was also the major degradation products of
BTH. The degradation process was carried out in a NaOAc/NaCl
buffer (pH=5.0). The degree of polymerization was quickly
decreased after the enzyme was added and o-HA was appeared in
the system. When the amount of HA4 was gradually increased,
the degradation rate decreased . The enzymatic hydrolysis
process was continued for 2 weeks to afford the completed
degradation products when the HMW-HA starting material was
To examine whether the building blocks 4, 5 and 6 was
suiltable for the glycoclusters assembling, a variety of scaffolds
was designed, which containing propargylated polyols 8a and 8b,
propargylated poly (amidoamine) (PAMAM) dendrimers 11c
and 11d, and GATG (gallic acid-triethylene glycol) dendrimers
13e, 13f and 13g.
23
2
1
50 g. After the hydrolysis process, hyaluronidase was
successfully removed from the reaction mixture by heating and
filtering procedure (see SI). To avoid the time-consuming anion
exchange or size exclusion chromatography that was usually
18
utilized for the o-HA purification , the mixture of degradation
2
0
products were directly protected and the fully-protected HA4 1
was obtained in 42% for 3 steps from the HA polysaccharide in
23 grams scales.
As the hyaluronan containing carboxyl group which is its
active group, amide condensation reaction should be avoided in
the next linkages. In this work, the linker was designed to
connect onto the reducing end of HA4 via O-glycosidic bond and
connected to the scaffold by copper-catalyzed azide-alkyne
22
Huisgen 1, 3-cycloadditions (CuAAC) . Therefore, the linker
molecules should contain both a hydroxyl group to connect with
o-HA and an azido or alkynyl group that could be coupled to core
scaffolds. To achieve these goals, azido-alcohol 7 was firstly
synthesized from triethylene glycol in two steps. Challenges
came from the O-glycosylation at the reducing end of the
tetrasaccharide 1. Since the low-reactivity of an acetyl protected
N-acetyl saccharide donor, the direct glycosylation at C-1 has
proven to be a great difficulty. The exchange of the NHAc group
Scheme 2. Synthesis of polyol derivatives HA glycoclusters
1
(
9
0a-b. (a) propargyl bromide, KOH, DMF, 0 →50 °C, 80.2 %
8a), 89.0 % (8b). (b) 4, CuSO , Na ascorbate, MeOH/H O, rt,
7.9% (9a), 91.0% (9b); (c) 1M LiOH(aq.)/H , THF, 0
4
2
2 2
O
°
C→rt, then 4 M NaOH, MeOH, 99% (10a), 99% (10b).
2
to NHPhth, NHTca, NHTroc N(Ac) or NHTFA was always
A dendrimer is a polymeric molecule composed of multiple
utilized to eliminate the formation of the oxazoline, which was
sluggish to transform into the desired glycosylated product under
the classical condition. However, in our synthetic approach, the
transformation of already existed NHAc groups required multiple
steps, thereby, a novel approach was developed. Firstly, the
acetyl protect group at the reducing end was removed by 3-
dimethyl aminopropyl amine (DMAPA) in THF. The exposed
perfectly branched monomers that emanate radially from a
central core, reminiscent of a tree, whence dendrimers derive
their name (Greek, dendra). The dendrimers are defined by
generation (G1, G2, G3…) and dendrimers of higher generations
are larger, more branched and have more end groups at their
24
periphery than dendrimers of lower generations . The first series
of dendrimers we chose was poly (amidoamine) (PAMAM)

3
25, 26
which was used frequently to assemble glycoclusters
and
offered the significant advantage of ease of preparation. Also, the
PAMAM dendrimers of varying size are commercially available.
In this research, G1-PAMAM (8 amino groups) and G2-
PAMAM (16 amino groups) were used as scaffold. Since the end
of PAMAM is an amino group, modification is required to
perform click chemistry. Thus 1- hexynoic acid was connected to
PAMAM with classic amide condensation.
Scheme 4. Synthesis of G1-G3-GATG-HA4 dendrimers (14e-
4 2
g). (a) 5, CuSO , Na ascorbate, MeOH / H O, rt, 86% (14e),
9
1% (14f), 90% (14g).
Scheme 3. Synthesis of (G1-G2)-PAMAM-HA4 dendrimers
The last step was connect the HA4 with the linker to the
scaffold. CuAAC was the perfect reaction due to its high
1
8
2c-d. (a) 5-Hexynoicacid, EDC, HOBt, DMF, 0 °C→rt,
8.4% (11c), 91.1% (11d); (b) 6, CuI, MeOH/H O, rt, 91.0%
2
22, 29-31
efficiency, no side reactions and mild conditions
. There
(12c), 87.0% (12d).
were two strategies for assembling glycoclusters, strategy I was
using the fully-protected HA4 to synthesis the clusters and then
the protective groups were removed while strategy II was
removing the protective groups firstly and then assemble
glycoclusters. Procedure for global deprotection was followed
the method provided by J-C Jacquinet to eliminate the formation
We also chose GATG dendrimers as scaffold because
presence of terminal azides in them could directly perform click
chemistry. GATG dendrimers which are composed of a repeating
unit carrying a gallic acid core and hydrophilic triethylene glycol
arms with terminal azide groups, was firstly synthesized by Roy
32
27
of the Δ4,5 GlcA . Strategy I was used for the synthesis of
compound 9a,b and the results were satisfactory (Scheme 2),
while compound 6 did not react with 9a and 9b at the same
condition (data was not shown). But the strategy I was not
suitable to all situations because the PAMAM series scaffolds
were not tolerated under H O condition. Thus, an alternative
and his group . The (G1-G3) GATG (13e-g) was obtained by
constant catalytic hydrogenation and amide condensation as
28
described in ref. .
Table 1. The synthesized hyaluronan tetrasaccharide
glycoclusters.
2
2
strategy (II) was utilized that yielded 12c and 12d in 91% and
7%, respectively (Scheme 3). The strategy II could be also
Compound
Name
M.W.
Calc.)
MALDI-
TOF
SEC-MALS
Mw (g/mol)
N/A
8
(
28
applied to the GATG series of scaffolds as reported and 14e-g
was synthesized with high yield (Scheme 4). The glycoclusters
were characterized by Matrix-Assisted Laser Desorption /
Ionization Time of Flight Mass Spectrometry (MALDI-TOF
MS), however, as molecular weight increases, glycoclusters fly
with more difficulty in MALDI-TOF MS and fragmentation
9
9
1
1
a
Pent-HA4-4
Gly-HA4-3
4023.6
3007.7
9653.3
22403.0
4042.8
+
[
M+Na]
b
3028.7
N/A
N/A
+
[M+Na]
2c
2d
G1-PAMAM-
HA4-8
9656.3+
33
appears . As a supplement, size exclusion chromatography
[M+H]
coupled to multiangle light scattering (SEC-MALS) method was
used to confirm the molecular weights of 12d, 14f and 14g. For
14f the Mw tested by SEC-MALS was twice of the calculated
value and for 14g was 20 times, maybe polymerization occurs
due to the GATG scaffolds. The results were shown in Table 1.
4
G2-PAMAM-
HA4-16
4448.12
[M+5Na]
2.168×10
5
+
(±4.135%)
1
1
4e
4f
G1-GATG-
HA4-3
3120.8
9807.8
3144.4 +
N/A
[M+Na]
4
G2-GATG-
HA4-9
-
-
1.933×10
In summary, hyaluronan tetrasaccharide was obtained by
enzymatic hydrolysis of high molecular weight hyaluronan with
large scale and high yield, and a new method of direct reaction
from the degradation products to the building blocks and then
separation and purification is the first report. The HA4 building
blocks was used for assembling glycoclusters in this research and
could be used for some other applications. The structurally
determined and diverse HA4 glycoclusters with wide molecular
(±2.956%)
5
1
4g
G3-GATG-
HA4-27
29967.3
6.018×10
(±2.160%)

4
Tetrahedron
weight range (3 kDa to 30 kDa) were synthesized, which were
the first reported structurally determined HA4 glycoclusters and
also large molecular weight rare tetrasaccharide glycoclusters.
These compounds can be used to study the mechanism of
hyaluronan biological properties and could be applied in the field
of biomedicine. Also this research could provide methods for the
synthesis of hyaluronan and other glycosaminoglycan
glycoclusters.
Acknowledgments
This work was financially supported by the National Natural
Science Foundation of China (No. 21472007).
References and notes
1
2
.
.
Stern, R.; Jedrzejas, M. J. Chem. Rev., 2006, 106, 818-839.
Karbownik, M. S.; Nowak, J. Z. Pharmacol. Rep., 2013, 65, 1056-
1
074.
3
4
.
.
Day, A. J.; Prestwich, G. D. J. Biol. Chem., 2002, 277, 4585-4588.
Vigetti, D.; Karousou, E.; Viola, M.; Deleonibus, S.; De Luca, G.;
Passi, A. Bba-Gen. Subjects., 2014, 1840, 2452-2459.
5
.
Stern, R.; Asari, A. A.; Sugahara, K. N. Eur. J. Cell. Biol., 2006, 85,
6
99-715.
6
7
8
.
.
.
Monslow, J.; Govindaraju, P.; PurÃ, E. Front. Immuno., 2015, 6.
Cecioni, S.; Imberty, A.; Vidal, S. Chem. Rev., 2015, 115, 525-561.
Kiessling, L. L.; Gestwicki, J. E.; Strong, L. E. Curr. Opin. Chem.
Biol., 2000, 4, 696-703.
9
.
Mammen, M.; Choi, S. K.; Whitesides, G. M. Angew. Chem. Int. Ed..
1
998, 37, 2754-2794.
1
1
1
0. Weigel, P. H.; Baggenstoss, B. A. Glycobiology 2017, 1-10.
1. Lundquist, J. J.; Toone, E. J. Chem. Rev., 2002, 102, 555-578.
2. Rele, S. M.; Iyer, S. S.; Chaikof, E. L. Tetrahedron. Lett., 2007, 48,
5
055-5060.
3. Slaghek, T.; Nakahara, Y.; Ogawa, T. Tetrahedron. Lett., 1992, 33,
971-4974.
4. Slaghek, T. M.; Hyppönen, T. K.; Ogawa, T.; Kamerling, J. P.;
Vliegenthart, J. F. G. Tetrahedron. Lett., 1993, 34, 7939-7942.
5. Adamski-Werner, S. L.; Yeung, B. K. S.; Miller-Deist, L. A.; Petillo,
P. A. Carbohydr. Res., 2004, 339, 1255-1262.
1
1
1
1
4
6. Blundell, C. D.; Reed, M. A. C.; Almond, A. Carbohydr. Res., 2006,
3
41, 2803-2815.
1
1
7. Li, C.; Wang, L. X. Adv Carbohydr Chem Biochem 2016, 73, 73-116.
8. Tawada, A.; Masa, T.; Oonuki, Y.; Watanabe, A.; Matsuzaki, Y.;
Asari, A. Glycobiology, 2002, 12, 421-426.
1
9. Yuan, P.; Lv, M.; Jin, P.; Wang, M.; Du, G.; Chen, J.; Kang, Z.
Carbohydr Polym 2015, 129, 194-200.
2
2
0. Lu, X.; Huang, X. Glycoconjugate J 2015, 32, 549-556.
1. Deschrevel, B.; Tranchepain, F.; Vincent, J. Matrix Biol 2008, 27,
4
75-486.
2
2. Wu, P.; Feldman, A. K.; Nugent, A. K.; Hawker, C. J.; Scheel, A.;
Voit, B.; Pyun, J.; Fréchet, J. M. J.; Sharpless, K. B.; Fokin, V. V.
Angew Chem Int Ed., 2004, 43, 3928-3932.
2
2
2
3. Bosman, A. W.; Janssen, H. M.; Meijer, E. W. Chem. Rev., 1999, 99,
1
665-1688.
4. Lee, C. C.; MacKay, J. A.; Frechet, J. M.; Szoka, F. C. Nat. Biotech.,
005, 23, 1517-1526.
5. Tomalia, D. A.; Naylor, A. M.; Goddard, W. A. Angew Chem Int Ed.,
990, 29, 138-175.
2
1
2
2
6. Toyokuni, T.; Singhal, A. K. Chem Soc Rev 1995, 24, 231-242.
7. Roy, R.; Park, W. K. C.; Wu, Q.; Wang, S. N. Tetrahedron Lett 1995,
3
6, 4377-4380.
2
2
8. Fernandez-Megia, E.; Correa, J.; Rodríguez-Meizoso, I.; Riguera, R.
Macromolecules 2006, 39, 2113-2120.
9. Wu, P.; Malkoch, M.; Hunt, J. N.; Vestberg, R.; Kaltgrad, E.; Finn,
M. G.; Fokin, V. V.; Sharpless, K. B.; Hawker, C. J. Chem. Comm.,
2
005, 5775-5777.
0. Ledin, P. A.; Friscourt, F.; Guo, J.; Boons, G. J. Chemistry, 2011, 17,
39-846.
1. Kikkeri, R.; Liu, X.; Adibekian, A.; Tsai, Y. H.; Seeberger, P. H.
Chem. Comm. 2010, 46, 2197-2199.
3
3
3
3
8
2. Jacquinet, J-C.; Lopin-Bon, C.; Vibert, A. Chem. Eur. J., 2009, 15,
9
579-9595.
3. Fernandez-Megia, E.; Correa, J.; Riguera, R. Biomacromolecules
006, 7, 3104-3111.
2

5
Highlights
1
2
3
. An efficient method of obtaining HA4 building
block has been developed.
. HA4 glycoclusters with wide molecular weight
range (3k to 30 k) were synthesized.
. Strategy for synthesis of glycosaminoglycan
glycoclusters has been established.