Exploring the effect of sialic acid orientation on ligand-receptor interactions

Article abstract of DOI:10.1039/c2cc32587j

Here, we present the synthesis of two sialo-micelles to validate the significance of sialic acid orientation during specific carbohydrate-protein and carbohydrate-carbohydrate interactions. Our data clearly suggest that orientation of carboxylic acid and

Full text of DOI:10.1039/c2cc32587j

View Online / Journal Homepage
Dynamic Article Links
ChemComm
Cite this: DOI: 10.1039/c2cc32587j
www.rsc.org/chemcomm
COMMUNICATION
Exploring the effect of sialic acid orientation on ligand–receptor
interactionsw
Rohan Yadav and Raghavendra Kikkeri*
Received 11th April 2012, Accepted 16th May 2012
DOI: 10.1039/c2cc32587j
Here, we present the synthesis of two sialo-micelles to validate the
significance of sialic acid orientation during specific carbohydrate–
protein and carbohydrate–carbohydrate interactions. Our data
clearly suggest that orientation of carboxylic acid and glycerol
side chains of sialic acid moieties exert fine tuning of ligand–
receptor interactions.
To obtain sialo-micelles that could serve as multivalent
9
probes, we conjugated amphiphilic groups at C-2 and C-9
positions of sia respectively. Upon dissolution of amphiphiles
in water, self-assembled highly regular micelles were obtained.
The bioavailability of resultant sialo-micelles with plant and
human sialic acid binding protein (SBP) was evaluated by
surface plasma resonance (SPR) and in vitro assay. Sambucus
1
0
Sialic acids (sias) occur as the terminal part of glycan chains on
glycoproteins/glycolipids expressed by the deuterostome line-
1
age of animals and on certain bacterial species. Sias appear to
nigra agglutinin (SNA), Limax flavus agglutinin (LFA), P,
E-selectin and CD22 (Siglec-2) were selected as sialic acid
binding proteins (SBP), where SNA and LFA bind to all
1
0,11
dictate a wide variety of biological functions such as the
binding of hormones, toxins and viruses, maintenance of
surface negative charge, and contribution to the viscosity of
common sialic acid residues,
1
human CD22-Fc recognizes
2
siaa(2-6)-linked sias, and E and P-selectin, which belong to
the subgroup of the C-type lectins that mediate leukocyte
X
trafficking, are specific to sialyl Lewis and sialyl Lewis
2
a
mucins, etc. Further, a significant difference exists between
1
3
sialylation patterns in normal cells and their malignant
3
counterparts. Therefore, sia-analogs are crucial in order to
glycans respectively.
The syntheses of compounds 3 and 4 are depicted in
Scheme 1. The synthesis of O-sialoside 3 was carried out with
fine-tune their biological behaviour. Over fifty different natu-
rally occurring sia species have so far been identified. In
mammals, the most abundant sias are N-acetylneuraminic
acid (Neu5Ac) and N-glycolylneuraminic acid (NeuGc) which
exist either as O-acetate, lactate, sulfate or phosphate ester, or
methyl ether forms at 4, 7, 8, and 9 positions giving rise to a
1
4
b-thioglycoside donor 6
n-undecanol, followed by de-acetylation. 9-Amidosialic acid
which was glycosylated with
4 was synthesized starting from mono-tosylation of sialic acid
6
b
7, followed by azidation, hydrogenation of 8 and coupling
with dodecanoic acid (Scheme 1). The structures of 1 and 2
could be unambiguously confirmed by mass spectrometry and
1a,4
great variety of compounds and isomers.
These substitu-
tions alter the biological properties of sias compared to their
parental Neu5Ac ligand. For example, C-9 and C-5 O-acety-
lated Neu5Ac are expressed by human tissues such as brain,
colon, salivary and gastric mucins and peripheral blood cells
5
to offer resistance against bacterial or viral neuramidases.
Similarly, sialic acid–CD22 interaction was modulated by
introducing an aromatic substituent at the C-9 position of
6
sias. Furthermore, it has been shown that lactamized-sialyl
X
-sulfo Lewis , but not conventional sialyl Lewis , serves as
X
6
7
the major ligand for L-selectin. Overall, a recent survey along
with experimental evidence clearly indicates that modified sias
8
are more widely distributed than the parental Neu5Ac ligand.
In this communication, we describe that sialic acid locked in
two opposite orientations on micelles effectively tunes sia
dependent ligand–receptor interactions in their native context.
Indian Institute of Science Education and Research,
Sai Trinity Building, Pashan, Pune 411021, India.
E-mail: rkikkeri@iiserpune.ac.in; Fax: +91-20-25899790;
Tel: +91-20-25908207
Scheme 1 Synthesis of sialic acid based amphiphiles (1 and 2);
+
a) amberlite-H /MeOH; Ac
(
(
(
2
O/pyridine; (b) PhSH/BF
˚
c) undecanoic acid/NIS/MS 4 A/TfOH, NaOMe/MeOH; (d) H
3
ÁEt
2
O;
O;
2
3 2
e) TsCl/pyridine; (f) NaN /acetone/water, Pd–C/H ; (g) dodecanoic
w Electronic supplementary information (ESI) available. See DOI:
10.1039/c2cc32587j
acid/DIC/NHS/dioxane/water.
This journal is c The Royal Society of Chemistry 2012
Chem. Commun.

View Online
Fig. 1 AFM topographical images of (a) compound 1, (b) compound
and its corresponding topology. The image scale is in micrometers.
2
1
13
NMR spectrometry in MeOD. Both H and C NMR spectra
displayed characteristic signals of aliphatic chains and sialic
acid units.
À1
Upon dissolution of 10 mg ml of 1 and 2 in water, self-
assembled behaviour of sia based micelles was characterized
by means of dynamic light scattering (DLS) and atomic force
microscopy (AFM). DLS patterns of 1 and 2 were almost
H
identical and the corresponding hydrodynamic radius (R )
was between 100 and 150 nm (Fig. S4, ESIw). Evidence for the
formation of circular aggregates was provided by AFM
experiments (Fig. 1).
After synthesizing sia micelles, the kinetics and mechanism of
SBP binding were investigated by using SPR. Kinetic analysis
1
5
was performed on a 1 : 1 interaction model. SBP of 25 mg
concentration was covalently bound to a polycarboxylated
CM5 sensor chip. Of the five SBP, four bind to micelle 2
Fig. 2 SPR sensorgrams of different sialic acid binding protein (SBP)
(
Fig. 2 and Table 1) to a varying extent and no binding has
binding to captured micelles containing two opposite orientations of
sialic acid. Sensorgrams a and b show different concentrations of 1 and
2 binding to the CM5 surface expressing SNA lectin; sensorgrams c
and d show interaction of 1 and 2 with LFA lectin; sensorgrams e and f
show interaction of 1 and 2 binding with E-selectin lectin; sensorgrams
g and h show interaction of 1 and 2 binding with P-selectin lectin;
sensorgrams i and j show interaction of 1 and 2 with CD22-Fc;
concentrations of 1 and 2: 0 mM (black line), 5 mM (red line), 10 mM
been observed with CD22-Fc. 1 served as a positive control.
For SNA lectin interaction at 25 1C, there is a marginal
decrease in the binding of 2 compared to 1 (0.88-fold decrease
in affinity versus compound 1). With the LFA lectin module,
there is 0.9-fold decrease in affinity to compound 2, which also
disassociates from LFA much slower than compound 1. Over-
all, the degree of plant lectin binding induced by micelle 2 was
more or less similar to that of 1. This may be due to the non-
specific recognition of sias including Neu5Ac, Neu5Gc and
(
dark blue line), 20 mM (green line), 30 mM (dark red line) 40 mM (blue
line) and 50 mM (orange line) respectively.
1
1
their respective glycan residues by plant lectins.
For E and P-selectin–micelle interactions, K
Table 1 Equilibrium constant, K
D
, of 1 and 2
(mM)
D
for 2 was
K
D
almost similar to that for 1. This may be due to the
16
a-hydroxycarboxylic acid residue of 2 which coordinates with
calcium ions in a manner akin to sialyl Lewis . To confirm
Lectins
Comp 1
Comp 2
X
Sambucus nigra agglutinin (SNA)
Limax flavus agglutinin (LFA)
E-Selectin
P-Selectin
CD22-Fc
0.146
0.149
0.32
0.35
0.43
0.129
0.134
0.31
0.36
2
+
this hypothesis, we soaked compound 2 with Ca
ions.
Interestingly, DLS measurements showed aggregates of size
B1000–1500 nm (Fig. S6, ESIw) which was further supported
by AFM imaging (Fig. S5, ESIw). Taken together, lectin bind-
ing assays indicate that sialic acid orientation has little impact
No binding
X
and 2 may be a small potential mimic of the sialyl Lewis
for interaction between 2 and CD22-Fc was comparable to
results obtained using only buffer as a negative control. These
observations were consistent with those of Kelm and Oetke
et al., who reported the requirement of hydroxyl groups at C-9
ligand. In addition, the terminal carboxylic acid group facil-
2+
itates Ca mediated carbohydrate–carbohydrate interactions.
With the CD22-Fc (Siglec-2) module, there was a significant
difference between 1 and 2. Data derived from SPR assay
demonstrate no binding activity with 2, whereas binding of 1
displays a KD value of 0.43 mM. The binding curves obtained
1
7
for sialic acid recognition by CD22.
To further support our assessment by SPR, a CD22 trans-
fected CHO cell line was used to study sialic acid orientation
Chem. Commun.
This journal is c The Royal Society of Chemistry 2012

View Online
P. Stanley, C. R. Bertozzi, G. W. Hart and M. E. Etzler, Essentials
of Glycobiology, Cold Spring Harbor Press, 2009, 2nd edn,
pp. 199–218; (c) X. Chen and A. Varki, ACS Chem. Biol., 2010,
5
, 163–176.
2
(a) L.-M. Chen, O. Blixt, J. Stevens, A. S. Lipatov, C. T. Davis,
B. E. Collins, N. J. Cox, J. C. Paulson and R. O. Donis, Virology,
2
012, 424, 154; (b) M. A. Langereis, Q. Zeng, B. Heesters,
E. G. Huizinga and R. J. deGroot, PLoS Pathog., 2012,
, e1002492; (c) J. C. C. Lai, J.-M. Garcia, J. C. Dyason,
8
R. Bohm, P. D. Madge, F. J. Rose, J. M. Nicholls, J. S. Malik
Peiris, T. Haselhorst and M. von Itzstein, Angew. Chem., Int. Ed.,
2
8
012, 51, 2221; (d) H. Ogura, Proc. Jpn. Acad., Ser. B., 2011,
7, 328.
3
4
(a) A. Takeshita, N. Yamakage, K. Shinjo, t. Ono, I. Hirano,
S. Nakamura, K. Shigeno, T. Tobita, M. Maekawa, H. Kiyoi,
T. Naoe, K. Ohnishi, Y. Sugimoto and R. Ohno, Leukemia, 2009,
23, 1372; (b) D. H. Nguyen, P. Tangvoranuntakul and A. Varki,
J. Immunol., 2005, 175, 228.
R. Schauer, Sialic acids: Chemistry, Metabolism and Function, Cell
Biology Monographs, Springer-Verlag, New York, 1982, vol. 10.
R. Schauer, Adv. Carbohydr. Chem. Biochem., 1982, 40, 131.
(a) B. E. Collins, O. Blixt, A. R. DeSieno, N. Bovin, J. D. Marth
and J. C. Paulson, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 6104;
5
6
Fig. 3 Flow cytometry graph of CHO-CD22 with 1 and 2: CHO-CD22
cells were left untreated (tinted area), incubated with 50 mM of compound
1
(red line area, positive control), compound 2 (black line area).
(
b) S. Han, B. E. Collins, P. Bengtson and J. C. Paulson, Nat.
Chem. Biol., 2005, 1, 93; (c) B. Blixt, S. Han, L. Liao, Y. Zeng,
J. Hoffmann, S. Futakawa and J. C. Paulson, J. Am. Chem. Soc.,
dependent lectin recognition and binding. Fluorescently
labeled sialic acid micelles were prepared by mixing 5 mg of
fluorescein dye with equal amounts of ligands 3 and 4 in water,
followed by 10 K cutoff microcon filtration to obtain fluorescein
hosted sialic acid micelles. The sialic acid–CD22 interactions
on the cell surfaces were then measured by flow cytometry. As
shown in Fig. 3, compound 2 did not bind effectively to CD22-
transfected CHO-cells, indicating the requirement of native
sialic acid orientation for the recognition. On the other hand,
the CHO-K1 cell line served as a negative control and showed
no major uptake of either 1 or 2 (data not shown). These
findings indicate that compound 2 is a potential sialic acid
moiety for fine-tuning sialic acid based ligand–receptor inter-
action in mammalian SBP.
2
008, 28, 6680; (d) M. K. O’Reilly, B. E. Collins, S. Han, L. Liao,
C. Rillahan, P. I. Kitaov, D. R. Bundle and J. C. Paulson, J. Am.
Chem. Soc., 2008, 130, 7736.
M. Yamaguchi, H. Ishida, A. Kanamori, R. Kannagi and M. Kiso,
Carbohydr. Res., 2003, 338, 2793.
(a) N. K. Sauter, J. E. Hanson, G. D. Glick, J. H. Brown, R. L.
Crowther, S.-J. Park, J. J. Skehel and D. C. Wiley, Biochemistry,
7
8
1
2
992, 31, 9609; (b) S. J. Gamblin and J. J. Skehel, J. Biol. Chem.,
010, 285, 28403; (c) C. Mitsuoka, K. Ohmori, N. Kimura,
A. Kanamori, S. Komba, H. Ishida, M. Kiso and R. Kannagi,
Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 1597; (d) A. Cariappa,
H. Takematsu, H. Liu, S. Diaz, K. Haider, C. Boboila, G. Kalloo,
M. Connole, H. N. Shi, N. Varki, A. Varki and S. Pillai, J. Exp.
Med., 2009, 206, 125.
(a) P. V. Santacroce and A. Basu, Angew. Chem., Int. Ed., 2003,
42, 95; (b) Y. Nagasaki, K. Yasugi, Y. Yamamoto, A. Harada and
K. Kataoka, Biomacromolecules, 2001, 2, 1067; (c) X. Qu,
V. V. Khutoryanskiy, A. Stewart, S. Rahman, B. Papahadjopoulos-
Sternberg, C. Dufes, D. McCarthy, C. G. Wilson, R. Lyons,
K. C. Carter, A. Schatzlein and I. F. Uchegbu, Biomacromolecules,
2006, 7, 3452; (d) R. A. Bader, A. L. Silvers and N. Zhang,
Biomacromolecules, 2011, 12, 314.
9
We have shown that structural rearrangement of sialic acid
moieties around a multivalent system could modulate the
ligand–receptor interactions. Surface plasmon resonance
(
SPR) and in vitro binding assays show that out of five
1
0 R. L. Miller and J. D. Cannon Jr, Prog. Clin. Biol. Res., 1984,
57, 31.
potential sialic acid binding plant and human lectins (SBP),
only CD22 (Siglec-2) abolishes binding with an amphiphile at
the C-9 position of sialic acid under biophysical conditions,
while O-sialoside was used as a positive control. These differences
in the relative binding affinities of two human lectins provide a
novel lead in the development of inhibitors and biomarkers
for specific human SBP. Such studies are currently being
considered. Furthermore, stereo-specificity of the carboxylic
1
11 (a) M. J. Linman, J. D. Taylor, H. Yu, X. Chen and Q. Chen, Anal.
Chem., 2008, 80, 4007; (b) R. Schauer, S. Kelm, G. Reuter,
P. Roggentin and L. Shaw, in Biology of the Sialic Acid, ed.
A. Rosenberg, New York, 1995, p. 7; (c) J. A. Wasylnka,
M. L. Simmer and M. M. Moore, Microbiology, 2001, 147, 869;
(d) J. M. Wolosin and Y. Wang, Invest. Ophthalmol. Visual Sci.,
1
995, 36, 2277.
1
2 (a) C. R. Bertozzi, Chem. Biol., 1995, 2, 703; (b) T. A. Springer,
Cell, 1994, 76, 301; (c) S. D. Rosen and C. R. Bertozzi, Curr. Opin.
Cell Biol., 1994, 6, 663.
2
+
acid residue of sialic acids also influences Ca
mediated
carbohydrate–carbohydrate interactions. These observations
provide a novel lead in the development of inhibitors and
biomarkers for specific sialic acid binding proteins. Such
studies are currently being considered.
13 (a) W. C. Chen, G. C. Completo, D. S. Sigal, P. R. Crocker,
A. Saven and J. C. Paulson, Blood, 2010, 115, 4778–4786;
(
9
b) N. Razi and A. Varki, Proc. Natl. Acad. Sci. U. S. A., 1998,
5, 7469–7474.
14 (a) R. Roy, J. Carbohydr. Chem., 2002, 21, 769; (b) A. Hasegawa,
J. Nakamura and M. Kiso, J. Carbohydr. Chem., 1986, 5, 11.
R. K. and R. Y. thank IISER, Pune, Indo-German (DST-MPG)
program, and CSIR, India, for financial support. We thank
Prof. Ajit Varki for providing the CHO-CD22, CHO-K1 cell
lines and CD22-Fc.
1
1
5 K. D. Hardman and C. F. Ainswort, Biochemistry, 1972, 11, 4910.
6 (a) R. Codd, Chem. Commun., 2004, 2653; (b) J.-P. Behr and
J.-M. Lehn, FEBS Lett., 1973, 31, 297; (c) J.-P. Behn and
J.-M. Lehn, FEBS Lett., 1972, 22, 178.
1
7 (a) S. Kelm, R. Brossmer, R. Isecke, H. J. Gross, K. Strenge and
R. Schauer, Eur. J. Biochem., 1998, 263, 18507; (b) C. Oetke,
R. Brosmer, L. R. Mantey, S. Hinderlich, R. Isecke,
W. Reutter, O. T. Keppler and M. Pawlite, J. Biol. Chem., 2002,
277, 6688.
Notes and references
1
(a) T. Angata and A. Varki, Chem. Rev., 2002, 102, 439;
b) A. Varki, R. D. Cummings, J. D. Esko, H. H. Freeze,
(
This journal is c The Royal Society of Chemistry 2012
Chem. Commun.