Anti-GM1/GD1a complex antibodies in GBS sera specifically recognize the hybrid dimer GM1-GD1a

Article abstract of DOI:10.1093/glycob/cwr139

It is now emerging the new concept that the antibodies from some patients with Guillain-Barre syndrome (GBS) recognize an antigenic epitope formed by two different gangliosides, a ganglioside complex (GSC). We prepared the dimeric GM1-GD1a hybrid ganglioside derivative that contains two structurally different oligosaccharide chains to mimic the GSC. We use this compound to analyze sera from GBS patients by high-performance thin-layer chromatography immunostaining and enzyme-linked immunosorbent assay. We also synthesized the dimeric GM1-GM1 and GD1a-GD1a compounds that were used in control experiments together with natural gangliosides. The hybrid dimeric GM1-GD1a was specifically recognized by human sera from GBS patients that developed anti-oligosaccharide antibodies specific for grouped complex oligosaccharides, confirming the information that GBS patients developed antibodies against a GSC. High-resolution ¹H-¹³C heteronuclear single-quantum coherence-nuclear overhauser effect spectroscopy nuclear magnetic resonance experiments showed an interaction between the IV Gal-H1 of GM1 and the IV Gal-H2 of GD1a suggesting that the two oligosaccharide chains of the dimeric ganglioside form a single epitope recognized by a single-antibody domain. The availability of a method capable to prepare several hybrid gangliosides, and the availability of simple analytical approaches, opens new perspectives for the understanding and the therapy of several neuropathies. The Author 2011. Published by Oxford University Press. All rights reserved.

Full text of DOI:10.1093/glycob/cwr139

Glycobiology vol. 22 no. 3 pp. 352–360, 2012
doi:10.1093/glycob/cwr139
Advance Access publication on September 14, 2011
Anti-GM1/GD1a complex antibodies in GBS sera
specifically recognize the hybrid dimer GM1–GD1a
Laura Mauri2, Riccardo Casellato2, Maria G Ciampa2,
Yoshinori Uekusa3,6, Koichi Kato3,6, Ken-ichi Kaida4,
Mayumi Motoyama5, Susumu Kusunoki5, and
Sandro Sonnino1,2
Keywords: antibodies / ganglioside complexes / GBS / lipid
rafts / neuropathies
2Department of Medical Chemistry, Biochemistry and Biotechnology,
University of Milan, Via Fratelli Cervi 93, 20090 Segrate, Italy; 3Okazaki
Institute for Integrative Bioscience and Institute for Molecular Science,
National Institutes of Natural Sciences, 5-1 Higashiyama, Myodaiji, Okazaki
444-8787, Japan; 4Division of Neurology, Internal Medicine, National
Defense Medical College, Saitama 359-8513, Japan; 5Department of
Neurology, Kinki University School of Medicine, 377-2 Ohno-Higashi,
Osaka-Sayama, Osaka 589-8511, Japan; and Graduate School of
Pharmaceutical Sciences, Nagoya City University, Tanabe-dori 3-1, Mizuho-
ku, Nagoya 467-8603, Japan
Introduction
High titers of antibodies that recognize carbohydrate epitopes
shared by mammalian and bacterial cell membranes can be
found in human sera following bacterial infection and in the
course of such neuropathies, as Guillain–Barré syndrome
(GBS; Kusunoki et al. 2008), promoted by them, by targeting
the nodes of Ranvier or motor nerve terminals (Willison and
Yuki 2002; Van Doorn et al. 2008).
6
Received on July 29, 2011; revised on September 7, 2011; accepted on
September 7, 2011
Gangliosides are widely used in thin-layer chromatography
(TLC) immunostaining and in enzyme-linked immunosorbent
assays (ELISAs) to recognize serum anti-carbohydrate anti-
bodies. The specificity of anti-oligosaccharide antibodies is
variable. Some sera recognize several oligosaccharides, linked
to lipids and/or proteins, with different complexity, whereas
others seem highly specific (Kaida, Kamakura, et al. 2008).
This not necessarily represents the antibody–antigen inter-
action properties in vivo as proved by the non-constant capa-
bility of anti-oligosaccharide antibodies to develop
neuropathies when injected in animals (Paparounas et al.
1999). Glycolipids, and particularly gangliosides, are not ran-
domly distributed within the membrane but rather form
domains (Sonnino et al. 2006) that participate and modulate
signal transduction processes. Oligosaccharides of membrane
glycoconjugates, and particularly the oligosaccharides of gly-
colipids, are cryptic structures. Glycoproteins protrude from
the plasma membrane into the extracellular environment, but
neighboring proteins can prevent external interactions. The
oligosaccharide chain of glycolipids protrude from the plasma
membrane to a maximum of only 20–25 Å (Sonnino et al.
1994) and from many years are considered very cryptic for
targeting by antibodies, lectins and enzymes (Prince 1992;
Greenshields et al. 2009).
It is now emerging the new concept that the antibodies
from some patients with Guillain–Barré syndrome (GBS)
recognize an antigenic epitope formed by two different
gangliosides, a ganglioside complex (GSC). We prepared
the dimeric GM1–GD1a hybrid ganglioside derivative that
contains two structurally different oligosaccharide chains
to mimic the GSC. We use this compound to analyze sera
from GBS patients by high-performance thin-layer chrom-
atography immunostaining and enzyme-linked immunosor-
bent assay. We also synthesized the dimeric GM1–GM1
and GD1a–GD1a compounds that were used in control
experiments together with natural gangliosides. The hybrid
dimeric GM1–GD1a was specifically recognized by human
sera from GBS patients that developed anti-oligosaccharide
antibodies specific for grouped complex oligosaccharides,
confirming the information that GBS patients developed
1
antibodies against a GSC. High-resolution H-13C hetero-
nuclear single-quantum coherence-nuclear overhauser
effect spectroscopy nuclear magnetic resonance exper-
iments showed an interaction between the IV Gal-H1 of
GM1 and the IV Gal-H2 of GD1a suggesting that the two
oligosaccharide chains of the dimeric ganglioside form a
single epitope recognized by a single-antibody domain. The
availability of a method capable to prepare several hybrid
gangliosides, and the availability of simple analytical
approaches, opens new perspectives for the understanding
and the therapy of several neuropathies.
A new concept that is emerging in recent times suggests
that a specific combination of oligosaccharide structure, of
their position and clustering can form a distinct epitope for
serum anti-oligosaccharide antibodies that not necessarily
show binding properties for the individual oligosaccharides or
part of them (Kaida et al. 2004; Kaida and Kusunoki 2010).
This needs the availability at the same time of specific anti-
oligosaccharide antibodies and of a specific plasma membrane
1To whom correspondence should be addressed: Tel: +39-0250330360;
Fax: +39-0250330365; e-mail: sandro.sonnino@unimi.it
organization, for the developing of
a
neuropathology.
© The Author 2011. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
352

Hybrid dimer ganglioside
The complexity of the natural membranes makes difficult to
proof this using simple experimental models.
GD1a and ꢀ50% for the hybrid dimer GM1–GD1a. TLC of
the dimers (Figure 3) shows homogeneity over 98% for the
three compounds that due to the high sugar content displayed
a retention time lower than that of natural gangliosides. As
expected, the chromatographic behavior of the hybrid dimer
GM1–GD1a was intermediate between those of dimers GM1–
GM1 and GD1a–GD1a.
To overcome this, we prepared the dimeric GM1–GD1a
hybrid ganglioside that contains two structural different oligo-
saccharide chains. We found that the serum IgG reactive with
the GM1/GD1a complex actually recognize the GM1–GD1a
hybrid dimer.
The three dimers were characterized by mass spectrometry
(MS) and nuclear magnetic resonance (NMR). By NMR, we
also performed experiments aimed to have information on the
three-dimensional structure of the hybrid dimer GM1–GD1a.
The synthesis of the three dimers started from lysoganglio-
sides prepared from natural compounds extracted from calf
brains. These gangliosides are a mixture of C18- and
C20-sphingosine. Combining the lysoderivatives, each com-
pound homogeneous in the oligosaccharide chains results het-
erogeneous in the lipid moiety, displaying the three
combinations of sphingosine C18/C18, C18/C20 and C20/
C20. MS confirmed the calculated molecular mass for the
three compounds. The electro-spray ionization (ESI)-MS
spectra show the main charged ions [M-2H]2− at 1334, 1348
and 1362, [M-3H]3− at 986, 995 and 1004 and [M-4H]4− at
Results
The preparation of ganglioside dimers GM1–GM1, GD1a–
GD1a and GM1–GD1a was carried out starting from lysogan-
gliosides. Lysoganglioside preparations were described in
great details (Neuenhofer et al. 1985; Sonnino et al. 1992;
Mauri et al. 2004, Valiente et al. 2001) in the past and they
are currently available in our laboratory.
The lysogangliosides are connected with adipic acid to
form the dimer compounds. Figures 1 and 2 show schemes
for the preparation of ganglioside dimers and for the hybrid
dimer GM1–GD1a. The yield of preparation was good and
determined to be 70% for dimers GM1–GM1 and GD1a–
Fig. 1. Scheme of the preparation of ganglioside dimers GM1–GM1 and GD1a–GD1a.
353

L Mauri et al.
Fig. 2. Scheme of the preparation of ganglioside hybrid dimer GM1–GD1a.
812, 819 and 826 for GM1–GM1, GM1–GD1a and GD1a–
GD1a, respectively. Figure 4 shows the mass spectra of the
hybrid dimer GM1–GD1a. Less abundant signals correspond-
ing to differently charged ions were also in the spectra.
GD1a has an additional sialic acid with respect to GM1
and it is very well separated by high-performance TLC
(HPTLC). Figure 3 shows the results of TLC immunostaining
with GBS serum. Both gangliosides were not recognized by
the serum from
a
patient with GBS in the
The hybrid dimer GM1–GD1a was characterized by high-
resolution NMR experiments. Table I reports the proton NMR
HPTLC-immunostaining procedure. Nevertheless, if GD1a is
applied on the plate above GM1 to have a less separation
from GM1, an immunostained spot with chromatographic
behavior intermediate between GM1 and GD1a was observed.
We know that the chromatographic resolution is not absolute
and that each ganglioside is represented on the TLC plate by
a large spot due to heterogeneity of the lipid moiety. Between
the two spots of GM1 and GD1a, we expect to have a
mixture of the more hydrophilic species of GM1 (short-chain
fatty acids) and less hydrophilic species of GD1a (long-chain
fatty acids). Of course, the serum antibodies recognized the
GM1/GD1a mixture in a dot-spot assay.
1
assignments and Figure 5 a portion of the H-13C HSQC (het-
eronuclear single-quantum coherence)-NOESY (nuclear over-
hauser effect spectroscopy) spectrum. Amphiphilic
compounds, like gangliosides and ganglioside/phospholipid
mixtures, form big aggregates not suitable for NMR analyses
in water solution. To mimic a membrane organization, exper-
iments in water were carried out inserting the hybrid dimer
GM1–GD1a into DPC aggregates. These aggregates have
been shown to be small and to display high mobility in sol-
ution; this allowed to obtain high-resolution spectra (Poppe
et al. 1994).
The three dimers GM1–GM1, GD1a–GD1a and GM1–
GD1a are separated each other by HPTLC. The dimeric
hybrid GM1–GD1a was very well recognized by the GBS
serum (Figure 3). As a control, we used dimeric GM1 and
dimeric GD1a. No reactivity was observed with the patient
Figure 5 shows a correlation between the IV Gal-H1 of the
monosialyl oligosaccharide chain and the IV Gal-H2 of the
disialyl oligosaccharide chain. This suggests a possibility that
the two oligosaccharide chains should be locked in a quite
rigid conformation forming a single epitope.
354

Hybrid dimer ganglioside
Fig. 3. HPTLC and HPTLC immunostaining of gangliosides and ganglioside dimers. Two micrograms of GM1, GD1a, a mixture of GM1 and GD1a,
GM1–GM1 dimer, GD1a–GD1a dimer and GM1–GD1a hybrid dimer was applied to a TLC plate, developed with a solvent system chloroform:methanol:0.2%
CaCl2-2H2O (50:45:10, v/v). The left panel was stained by the orcinol reagent and the right panel was immunostained by a patient serum IgG (diluted 1:100)
specifically reactive with a mixture of GM1 and GD1a. Anti-human IgG Fc, diluted to 1:200 with 1% BSA in PBS, was used as a secondary antibody. Each
horizontal line indicates the start line for GM1 and GD1a. The start line for each dimer is same as that for GM1.
serum, excluding any possible artifact deriving by the pres-
ence of two general oligosaccharide chains. This strongly
suggests that both the GM1 and GD1a chains are necessary
for a strong interaction and to maintain the stability of the
antibody–antigen complex.
followed by a process of serious neuropathy. Sometimes, this
interaction, in in vitro assays, requires an epitope obtained
mixing
two
structurally
different
gangliosides.
Anti-ganglioside complex (GSC) antibodies, in vivo, may
recognize a conformational epitope formed by two different
gangliosides on the lipid rafts.
Twelve sera that had been confirmed to be IgG positive for
a mixture of GM1 and GD1a according to the criteria reported
by Kaida et al. (2004) were used for the ELISA using the
three dimers as an antigen (Table II). The 12 sera showed no
reactivity against GM1 and no or minor reactivity against
GD1a, suggesting that both the oligosaccharide chain must be
present for antibody–antigen complex formation. Among the
sera, 10 showed strong antibody activity to the GM1–GD1a
hybrid dimer, whereas the two sera #3 and 11 did not. Sera
#3 and 11, from patients that had ophthalmoplegia, showed a
strong cross-reactivity against GD1b, GQ1b and GT1a.
Within the sera that recognize the hybrid dimer GM1–GD1a,
sera #4, 5 and 6 reacted with GD1b, but not with GQ1b or
GT1a.
Antibodies against a GM1/GD1a complex were detected in
39 of 234 GBS sera (Kaida et al. 2007). A subset of
anti-GM1/GD1a-positive patients also had the antibodies
against GD1a/GD1b and/or GD1b/GT1b complexes, which
are significantly associated with severe GBS (Kaida et al.
2007). The antibodies against the GM1/GalNAc-GD1a
complex were associated with pure motor GBS or acute motor
conduction block neuropathy (Kaida, Sonoo, et al. 2008). The
antibodies against GSCs containing GQ1b or GT1a are
present in the Fisher syndrome and related disorders (Kanzaki
et al. 2008). In contrast, the binding activities of the anti-
bodies highly specific to GD1b, associated with sensory
ataxic neuropathy, are decreased by the addition of such
gangliosides as GD1a to GD1b antigen due to the interaction
between the two gangliosides and resultant conformational
change (Kaida, Kamakura, et al. 2008). A similar inhibitory
effect of neighboring gangliosides on the antibody binding
has also been reported by the other research groups
(Greenshields et al. 2009; Nobile-Orazio et al. 2010).
Therefore, the analyses of the clustered gangliosides on the
plasma membrane are more and more important for clarifying
the pathogenetic mechanisms of autoimmune neuropathies.
We successfully prepared the hybrid dimer GM1–GD1a. In
addition, we also prepared the GM1–GM1 and GD1a–GD1a
dimers that were very useful in biological assays as controls.
The synthesis was carried out from lysogangliosides and did
Discussion
Several data (Sonnino et al. 2006; Prinetti et al. 2009; Sonnino
and Prinetti 2010; van Zanten et al. 2010) have been reported
on the segregation of gangliosides in membranes and on the
role of gangliosides in stabilizing lipid membrane domains
known as lipid rafts. The existence of lipid rafts subdomains
has been also reported (Lingwood and Simons 2010). This is
probably one of the reasons that make the oligosaccharide
chain cryptic reducing the possibility of an oligosaccharide–
serum antibody interaction. Sometimes, this occurs and it is
355

L Mauri et al.
Fig. 4. Negative-ion ESI/MS spectra of ganglioside hybrid dimer GM1–GD1a. (A) MS1 spectrum of the double-charged ions and triple-charged ions, resulting
from the ionization of the carboxylic group of sialic acid, from the three different possible combinations between lysogangliosides containing 18:1 and 20:1
sphingosine. (B) MS2 spectrum from ion at 995; the loss of an Neu5Ac residue originated bi-charged ions. (C) MS2 spectrum from ion at 1504.
not present specific difficulties and the final yields were from
50 to 70%. The availability of procedures for the preparation
of several lysogangliosides, such as lyso-GM3, lyso-GM1,
lyso-GD1a and lyso-GD1b, opens the possibility to prepare
several ganglioside hybrid dimers.
Most of the antibodies specifically reactive with a mixture
of GM1 and GD1a recognized the GM1–GD1a hybrid dimer,
suggesting that, at least in some cases, the hybrid dimer actu-
ally mimics the GM1/GD1a complex, a novel antigen formed
by the carbohydrate structures of GM1 and GD1a
gangliosides.
However, the two sera #3 and 11 (Table I) did not recog-
nize the GM1–GD1a dimer. Both of the patients 3 and 11 had
ophthalmoplegia and anti-GQ1b and anti-GT1a IgG antibody
activities in the sera. It indicates that a difference exists
between the antigenic epitope of the GM1/GD1a complex and
that of the GM1–GD1a dimer and that some of the IgG
antibodies reactive with both GQ1b and GT1a could cross-
react to the GM1/GD1a complex but not to the GM1–GD1a
dimer. These results suggest that the position and the features
of protein domain(s) interacting with the epitope(s) are differ-
ent, also if all our sera recognized the GM1/GD1a complex.
Many further NMR experiments will be necessary to have
the complete secondary structure of the two chains of the
hybrid dimer GM1–GD1a. This is due to the complexity of
the spectra, to some pick overlapping and to the fact that
some interactions occur via water bridges (Brocca et al.
1998). Nevertheless, while chemical-shift assignments suggest
that the conformation reported for the chains in the natural
gangliosides should be maintained in the hybrid dimer GM1–
GD1a, a defined interaction between the two external galac-
toses could be observed (Figure 5). Thus, the covalent lin-
kages of the two glucoses with the lipid moiety and the
hydrogen bond interactions at the end of the two chains
356

Hybrid dimer ganglioside
Table I. Chemical shifts of hybrid dimer GM1–GD1a in DPC aggregates in D2O at 30°C
GM1 part
GD1a part
Ceramide part
1H (ppm)
1H (ppm)
13C (ppm)
1H (ppm)
13C (ppm)
13C (ppm)
Glc (I)
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
7
8
1
2
3
4
5
6
1
2
3a
3e
4
5
6
4.40
3.29
3.60
3.56
3.54
3.75, 3.93
4.48
3.32
4.10
4.08
3.71
3.7–3.8
4.74
4.00
3.76
4.12
3.68
3.7–3.8
103.01
72.79
74.33
79.05
74.92
60.31
102.84
70.09
74.47
77.36
Glc (I)
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
7
8
1
2
3
4
5
6
1
2
3a
3e
4
5
6
4.40
3.29
3.60
3.56
3.54
3.75, 3.93
4.48
3.32
4.10
4.08
3.71
3.7–3.8
4.74
4.00
3.76
4.12
3.68
3.7–3.8
103.01
72.79
74.33
79.05
74.92
60.31
102.84
70.09
74.47
77.36
74.32
60.5–61.3
102.66
51.17
80.65
67.98
74.50
60.5–61.3
174.73
22.75
104.66
69.26
75.63
67.59
74.77
60.5–61.3
180.60
101.52
37.13
37.14
68.86
51.75
73.23
68.24
72.36
62.97
62.97
175.07
22.21
174.17
99.86
39.74
39.78
68.53
51.82
72.96
68.24
71.95
62.74
62.73
175.07
22.21
1a
1b
2
3
4
5
6
7
8
3.69
4.13
3.87
4.01
5.37
5.66
1.93
69.55
69.61
53.45
71.10
130.23
133.25
32.76
Gal (II)
Gal (II)
9
74.32
10
11
12
13
14
15
16
17
CH3
60.5–61.3
102.66
51.17
80.65
67.98
1.2–1.3
29.0–30.3
GalNAc (III)
GalNAc (III)
74.50
60.5–61.3
1.20
0.77
22.50
13.52
1.96
4.49
3.48
3.58
3.87
3.64
3.7–3.8
22.75
104.90
70.80
72.73
68.75
1.96
4.56
3.49
4.04
3.91
Gal (IV)
Gal (IV)
Linker part
1H (ppm)
13C (ppm)
74.93
60.5–61.3
3.63
3.7–3.8
1
2
2.11, 2.21
1.49, 1.54
35.65
25.20
NeuAc (A)
NeuAc (A)
1.89
2.62
3.74
3.77
3.45
3.55
3.71
3.59
3.83
37.13
37.14
68.52
51.75
73.23
68.24
72.36
62.97
62.97
1.89
2.62
3.74
3.77
3.45
3.55
3.71
3.59
3.83
7
8
9a
9b
10
11
7
8
9a
9b
10
11
1
1.99
22.21
1.99
NeuAc (B)
2
3a
3e
4
5
6
7
8
9a
9b
10
11
1.76
2.71
3.64
3.79
3.58
3.55
3.85
3.59
3.83
1.99
should arrange the eight neutral sugars in a quite fixed and
rigid conformation. This suggests that the GM1–GD1a oligo-
saccharide system behave as a single epitope recognized by a
single-antibody domain. This domain is present in 10 of the
12 sera analyzed. The remaining two sera #3 and 11 should
have two distinct domains, one for the GM1 oligosaccharide
and one for the GD1a oligosaccharide. Thus, these two sera
are capable to recognize a mixture of GM1 and GD1a but not
the single epitope formed by the two interacting chains.
The availability of hybrid gangliosides, containing two or
more oligosaccharide chains, mimicking the cluster of oligo-
saccharide chains occurring on cell membranes, provides a
new tool to analyze anti-oligosaccharide and correlated them
with the clinical features of the pathology. In addition to this,
their use to generate monoclonal antibodies could be useful to
develop animal models. It should open new perspectives for
understanding the pathogenesis and developing a novel
therapy of GBS.
357

L Mauri et al.
Fig. 5. Portion of the 1H-13C HSQC-NOESY spectrum of ganglioside hybrid dimer GM1–GD1a mixed with DPC-d38 in 10 mM phosphate buffer (pH 7.0) D2O
solution at 2.0 mM final concentration. Mixing time was 200 ms.
chemicals were of the highest purity available. The solvents
were distilled before use and deionized water was freshly dis-
tilled in a glass apparatus. Dialysis tubes 12.000–14.000 Da
were from Medicell International Ltd (London).
Gangliosides GM1 and GD1a were extracted from calf
brain, purified to homogeneity and characterized by NMR and
MS. Lyso-GM1 and lyso-GM1 were prepared from the corre-
sponding natural gangliosides (Neuenhofer et al. 1985;
Sonnino et al. 1992).
Table II. OD of the ELISA
Patient GM1 GD1a GM1/
GM1– GD1a– GM1– IgG
GM1 GD1a GD1a antibodies to
mixture dimer dimer hybrid other
dimer gangliosides
number
GD1a
1
2
3
<0.1 <0.1
0.516
<0.1
0.134
<0.1
<0.1
0.104
<0.1
0.799 No
0.489 No
<0.1
<0.1
0.154 0.52
0.145 0.65
<0.1
GD1b,
GQ1b,
GT1a
0.583 GD1b
4
5
<0.1 <0.1
<0.1 <0.1
0.664
0.49
<0.1
<0.1
Methods
0.111 <0.1
<0.1 <0.1
0.131
0.151
<0.1
0.127
<0.1
0.742 GD1b
0.548 GD1b
0.482 No
Preparation of GM1 and of GD1a dimers. To a solution of
0.15 M adipic acid in CH2Cl2/toluene 2:1 by vol., five
equivalents of thionyl chloride were added drop wise at −10°
C (Figure 1). After standing for 10 min at the same
temperature, the reaction mixture was heated at 50°C for 3 h
under reflux. The exceeding of thionyl chloride was removed
by evaporation at reduced pressure and the product used
without further purification. Adipoyl chloride and
tributylamine were added to a solution of lyso-GM1 or
lyso-GD1a in anhydrous CH2Cl2/CH3OH 1:2 by vol.
(50 μmol/mL) to a final molar ratio of 0.5:2:1. After vigorous
stirring for 2 h at room temperature, the reaction mixture was
dried and the residue purified by flash chromatography with
silica gel 60 column equilibrated and eluted with CHCl3/
CH3OH/H2O 60:35:5, by volume. The yield of reaction, after
purification, was 70%.
6
7
8
<0.1
<0.1
<0.1
0.135 0.519
0.199 0.497
0.111 0.671
<0.1
<0.1
<0.1
<0.1
<0.1
0.678 No
9
10
11
<0.1 <0.1
<0.1 <0.1
<0.1 <0.1
0.605
0.594
0.382
0.34
0.68
No
No
<0.1
GD1b,
GQ1b,
GT1a
0.751 No
12
<0.1
0.218 0.774
<0.1
0.101
Antigens: GM1, GD1a, GM1 and GD1a mixture, GM1–GM1 dimer, GD1a–
GD1a dimer and GM1–GD1a hybrid dimer.
Materials and methods
Chemicals
LiChroprep RP18 for column chromatography (particle size,
40–63 μm), silica gel 100 (particle size, 63–200 μm) and 60
(particle size, 15–40 μm), Amberlite® IR-120 H+ form (par-
ticle size 300–900 μm) and high-performance silica gel-
precoated thin-layer plates (HPTLC, Kieselgel 60) were
obtained from Merck (Darmstadt, Germany). All the
Preparation of hybrid GM1–GD1a dimer. To a solution of
0.15 M mono-methyl adipate in CH2Cl2/toluene 2:1 by vol.,
five equivalents of thionyl chloride were added drop wise at
−10°C (Figure 2). After standing for 10 min at the same
358

Hybrid dimer ganglioside
temperature, the reaction mixture was heated at 50°C for 3 h
under reflux. The exceeding of thionyl chloride was removed
by evaporation at reduced pressure and the product used
without further purification.
purified gangliosides GM1 and GD1a, a mixture of GM1 and
GD1a, the dimers GM1–GM1, GD1a–GD1a and GM1–GD1a.
Each microtiter well was coated with 0.2 μg of each antigen.
Serum diluted 1:40 with 1% bovine serum albumin (BSA) in
PBS was added to wells, followed by the procedure as
described previously (Kaida et al. 2004). Peroxidase-
conjugated goat anti-human IgG Fc antibody (diluted 1:200,
ICN Biomedicals Inc.) was the secondary antibody. A color
reaction was obtained by incubation with 200 μL of
orthophenylenediamine dihydrochloride (40 mg/dL of
phosphate–citrate buffer, pH 5.0) at room temperature for 2
min. The reaction was stopped by the addition of 8 N H2SO4,
after which the optical density (OD) at 492 nm was read with
an ELISA reader. OD values were corrected by subtracting the
OD of an uncoated well that had been processed similarly
(Table II). When the corrected OD was >0.1, the serum was
considered positive. ELISAs were repeated twice in the same
way, and the mean OD of the two experiments was calculated.
Antibody activities against other gangliosides, such as GM2,
GM3, GD1b, GD3, GalNAc-GD1a, GT1a, GT1b and GQ1b,
also were examined as described elsewhere (Kaida et al. 2004).
Methyladipoyl chloride and tributylamine were added to
a solution of lyso-GM1 in anhydrous CH2Cl2/CH3OH 1/2
(50 μmol/mL) to a final molar ratio of 1:2:1. After vigorous
stirring at room temperature for 2 h, the reaction mixture
was dried and the residue was dissolved in 0.5 M KOH in
CH3OH. After standing at room temperature for 18 h, the
solution was neutralized and the solvent was removed. The
residue was dissolved in water and the salts removed by
dialysis. After lyophilization, the GM1 adipate was obtained
as a white powder and used without further purification.
The reaction mixture was analyzed by HPTLC, using the
solvent systems CHCl3/CH3OH/0.2% CaCl2 50:42:11 by
volume.
Dicyclohexylcarbodiimide and N-hydroxysuccinimide, both
at a final concentration of 0.02 M, were added to a 0.017 M
solution of GM1 adipate, in dry dimethylformamide. After
stirring for 30 min at room temperature, the lyso-GD1a was
added to give a final concentration of 0.02 M. The reaction
was allowed to proceed at 60°C under vigorous stirring for
1 h. The reaction mixture was dried and the residue purified
by flash chromatography with silica gel 60 column equili-
brated and eluted with CHCl3/CH3OH/H2O, 60:35:5 by vol.
The yield of reaction, after purification, was 70%.
Mass spectrometry. ESI-MS was carried out in a negative
mode on
a
ThermoQuest Finnigan LCQdeca mass
spectrometer equipped with an electrospray ion source and an
Xcalibur™ data system. Samples were dissolved in methanol
at a concentration of 20–200 ng/μL prior to direct injection
into the electrospray ionization ion-trap mass spectrometer
(MS). Ions were monitored as MS1 or MS2 product ions.
Ionization was performed under the following conditions:
spray voltage, 4 kV; sheath gas flow rate, 50 arbitrary units;
capillary temperature, 260°C; capillary voltage, −42 V. The
scanning range was m/z 200–1600, and fragmentor voltage
for collision induced dissociation was 25–90%.
High-performance TLC. Ganglioside, ganglioside derivatives,
ganglioside dimers and the reaction mixtures were analyzed
by HPTLC using the solvent systems CHCl3/CH3OH/0.2%
CaCl2, 50:42:11 by vol., CHCl3/CH3OH/0.2% CaCl2/50 mM
KCl, 50:50:4:8 by vol. or CHCl3/CH3OH/(CH3)2CHOH/50
mM KCl, 30:35:15:13 by vol., according to the polarity of
compounds.
Gangliosides, ganglioside derivatives and dimers were visu-
alized on the HPTLC plates by treatment with anisaldehyde
and p-dimethylaminobenzaldehyde spray reagents followed
by heating at 130°C; amine-containing compounds were visu-
alized by treatment with 20% methanolic ninhydrin followed
by heating at 80°C.
For HPTLC immunostaining, 2 μg of gangliosides GM1
and GD1a of dimers GM1–GM1, GD1a–GD1a and GM1–
GD1a were applied to the TLC plate and developed with the
solvent of chloroform/methanol/aqueous 0.2% CaCl2
50:45:10, by vol. The line of application for GD1a was 1 cm
above that for GM1 and the dimers. Sera positive for the IgG
anti-GM1/GD1a complex antibody were overlaid for the TLC
Nuclear magnetic resonance. Two-dimensional NMR
experiments (HSQC, heteronuclear
2 bond correlation,
heteronuclear multiple bond correlation, chemical-shift
correlated spectroscopy, HSQC-total correlation spectroscopy,
HSQC-NOESY and NOESY) were performed with JEOL
ECA-920 or JEOL ECA-600 spectrometers at 30°C. The
pulse delay time of 2 s was used. All NMR spectra were
processed and analyzed using Delta (JEOL) and Sparky
(T. D. Goddard and D. G. Kneller: SPARKY 3, University of
California, San Francisco, CA). Experiments were carried out
on the hybrid dimeric GM1–GD1a inserted in a micelle of
dodecylphosphocholine (DPC)-d38 in 10 mM phosphate
buffer (pH 7.0) D2O solution. The hybrid dimer GM1–GD1a/
DPC molar ratio was 1:52 and the final hybrid dimer GM1–
GD1a concentration was 2.0 mM.
plate at
a dilution of 1:100 (Kaida et al. 2004).
Peroxidase-conjugated goat anti-human IgG Fc antibody
(diluted 1:200, ICN Biomedicals Inc., Aurora, OH) was the
secondary antibody. Immunoreactants were made visible with
phosphate-buffered saline (PBS) containing 0.01% H2O2 and
50 mg/dL 3,3′-diaminobenzidine tetrahydrochloride.
Quantitative determinations. Gangliosides, ganglioside
derivatives and ganglioside dimers were quantified by their
sialic acid content using the HCl–resorcinol method and
Enzyme-linked immunosorbent assay. Sera from 12 GBS
patients that displayed the IgG antibody against a mixture of
GM1 and GD1a were examined for their reactivity against the
hybrid dimer GM1–GD1a. ELISA was performed using
pure N-acetylneuraminic acid as
a
reference standard
(Svennerholm 1957).
359

L Mauri et al.
Kaida K, Sonoo M, Ogawa G, Kamakura K, Ueda-Sada M, Arita M,
Motoyishi K, Kusunoki S. 2008. GM1/GalNAc-GD1a complex: A target
for pure motor Guillain-Barré syndrome. Neurology. 71:1683–1690.
Kanzaki M, Kaida K, Ueda M, Morita D, Hirakawa M, Motoyoshi K,
Kamakura K, Kusunoki S. 2008. Ganglioside complexes containing GQ1b
as targets in Miller Fisher and Guillain-Barré syndromes. J Neurol
Neurosurg Psychiatry. 79:1148–1152.
Kusunoki S, Kaida K, Ueda M. 2008. Antibodies against gangliosides and
ganglioside complexes in Guillain-Barré syndrome: New aspects of
research. Bioch Biophys Acta. 1780:441–444.
Funding
This work was supported by the Ministry of Education,
Culture, Sports, Science and Technology of Japan
(Grants-in-Aid for Scientific Research, 21390273 to S.K. and
Grants-in-Aid for Scientific Research on Innovative Areas
20107004 to K.K.), the Ministry of Health, Labor, and
Welfare of Japan (Health and Labour Sciences Research
Grants for research on intractable diseases, H23-017) to S.K.
and PRIN (Progetti di interesse nazionale) 2009 to S.S.
Lingwood D, Simons K. 2010. Lipid rafts as a membrane-organizing prin-
ciple. Science. 327:46–50.
Mauri L, Prioni S, Loberto N, Chigorno V, Prinetti A, Sonnino S. 2004.
Synthesis of radioactive and photoactivable ganglioside derivatives
for the study of ganglioside-protein interactions. Glycoconj J.
20:11–23.
Neuenhofer S, Schwarzmann G, Egge H, Sandhoff K. 1985. Synthesis of
lysogangliosides. Biochemistry. 24:525–532.
Conflict of interest
None declared.
Nobile-Orazio E, Giannotta C, Briani C. 2010. Anti-ganglioside complex
IgM antibodies in multifocal motor neuropathy and chronic immune-
mediated neuropathies. J Neuroimmunol. 219:119–122.
Abbreviations
Paparounas K, O’Hanlon GM, O’Leary CP, Rowan EG, Willison HJ. 1999.
Anti-ganglioside antibodies can bind peripheral nerve nodes of Ranvier
and activate the complement cascade without inducing acute conduction
block in vitro. Brain. 122:807–816.
Poppe L, van Halbeek H, Acquotti D, Sonnino S. 1994. Carbohydrate
dynamics at a micellar surface: GD1a headgroup transformations revealed
by NMR spectroscopy. Biophys J. 66:1642–1652.
Prince A. 1992. Adhesins and receptors of Pseudomonas aeruginosa
associated with infection of the respiratory tract. Microb Pathog.
13:251–260.
Prinetti A, Loberto N, Chigorno V, Sonnino S. 2009. Glycosphingolipid
behaviour in complex membranes. Biochim Biophys Acta. 1788:
184–193.
Sonnino S, Acquotti D, Kirschner G, Uguaglianza A, Zecca L, Rubino F,
Tettamanti G. 1992. Preparation of lyso-GM1 (II3Neu5AcGgOse4-long
chain bases) by a one-pot reaction. J Lipid Res. 33:1221–1226.
Sonnino S, Cantù L, Corti M, Acquotti D, Venerando B. 1994. Aggregative
properties of gangliosides in solution. Chem Phys Lipids. 71:21–45.
Sonnino S, Prinetti A. 2010. Gangliosides as regulators of cell membrane
organization and functions. Adv Exp Med Biol. 688:165–184.
Sonnino S, Prinetti A, Mauri L, Chigorno V, Tettamanti G. 2006. Dynamic
and structural properties of sphingolipids as driving forces for the for-
mation of membrane domains. Chem Rev. 106:2111–2125.
BSA, bovine serum albumin; DPC, dodecylphosphocholine;
ELISA, enzyme-linked immunosorbent assay; ESI, electro-
spray ionization; GBS, Guillain–Barré syndrome; GD1a,
IV3NeuAc,II3NeuAc-Gg4Cer; GD1b, (II3NeuAc)2-Gg4Cer;
GM1, II3NeuAc-Gg4Cer; GQ1b, (IV3NeuAc)2,(II3NeuAc)2-
Gg4Cer; GT1a, (IV3NeuAc)2,II3NeuAc-Gg4Cer; GSC,
ganglioside complex; HPTLC, high-performance thin-layer
chromatography; HSQC, heteronuclear single-quantum coher-
ence; MS, mass spectrometry; NMR, nuclear magnetic reson-
ance; NOESY, nuclear overhauser effect spectroscopy; OD,
optical density; PBS, phosphate-buffered saline; TLC, thin-
layer chromatography.
References
Brocca P, Berthault P, Sonnino S. 1998. Conformation of the oligosaccharide
chain of GM1 ganglioside in a carbohydrate enriched surface. Biophys
J. 74:309–318.
Greenshields KN, Halstead SK, Zitman FM, Rinaldi S, Brennan KM,
O’Leary C, Chamberlain LH, Easton A, Roxburgh J, Pediani J, et al. 2009.
The neuropathic potential of anti-GM1 autoantibodies is regulated by the
local glycolipid environment in mice. J Clin Invest. 119:595–610.
Kaida K, Kamakura K, Ogawa G, Ueda M, Motoyoshi K, Arita M, Kusunoki
S. 2008. GD1b-specific antibody induces ataxia in Guillain-Barré syn-
drome. Neurology. 71:196–201.
Kaida K, Kusunoki S. 2010. Antibodies to gangliosides and ganglioside com-
plexes in Guillain–Barré syndrome and Fisher syndrome: Mini-review.
J Neuroimmunol. 223:5–12.
Kaida K, Morita D, Kanzaki M, Kamakura K, Motoyoshi K, Hirakawa M,
Kusunoki S. 2004. Ganglioside complexes: As new target antigens in
Guillain-Barré syndrome. Ann Neurology. 56:567–571.
Svennerholm L. 1957. Quantitative estimation of sialic acids. II. A colori-
metric resorcinol-hydrochloric acid method. Biochim Biophys Acta.
24:604–611.
Valiente O, Mauri L, Casellato R, Fernandez LE, Sonnino S. 2001.
Preparation of deacetyl-, lyso-, and deacetyl-lyso-GM(3) by selective alka-
line hydrolysis of GM3 ganglioside. J Lipid Res. 42:1318–1324.
Van Doorn P, Ruts L, Jacobs B. 2008. Clinical features, pathogenesis, and
treatment of Guillain-Barré syndrome. Lancet Neurol. 7:939–950.
van Zanten TS, Gomez J, Manzo C, Cambi A, Buceta J, Reigada R,
Garcia-Parajo MF. 2010. Direct mapping of nanoscale compositional con-
nectivity on intact cell membranes. Proc Natl Acad Sci USA.
107:15437–15442.
Kaida K, Morita D, Kanzaki M, Kamakura K, Motoyoshi K, Hirakawa M,
Kusunoki S. 2007. Anti-ganglioside complex antibodies associated with
severe disability in GBS. J Neuroimmunol. 182:212–218.
Willison HJ, Yuki N. 2002. Peripheral neuropathies and anti-glycolipid anti-
bodies. Brain. 125:2591–2625.
360