|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 276, Issue 34, 32274-32281, August 24, 2001
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the
Received for publication, April 12, 2001, and in revised form, June 11, 2001
Tetanus toxin, a member of the family of
Clostridial neurotoxins, is one of the most potent toxins
known. The crystal structure of the complex of the COOH-terminal
fragment of the heavy chain with an analogue of its ganglioside
receptor, GT1b, provides the first direct identification and
characterization of the ganglioside-binding sites. The ganglioside
induces cross-linking by binding to two distinct sites on the Hc
molecule. The structure sheds new light on the binding of
Clostridial neurotoxins to receptors on neuronal cells and
provides important information relevant to the design of anti-tetanus
and anti-botulism therapeutic agents.
Tetanus toxin (TeNT)1
and the botulinum toxins (BoNTs) are extremely potent neurotoxins
produced by the anaerobic bacteria, Clostridium
tetani and Clostridium botulinum,
respectively. The toxins are structurally and functionally related,
each being synthesized as a 150-kDa single polypeptide that is
subsequently proteolytically cleaved to give a 50-kDa amino-terminal
L-chain, disulfide bonded to a 100-kDa carboxyl-terminal H-chain (1,
2). The L-chain has a metalloprotease activity and is responsible for
the toxicity (3). The heavy chain can be cleaved into two fragments,
HN and HC. Each fragment is thought to have a
distinct function, the HC fragment for binding to sensitive
cells and subsequent internalization into vesicles (4-6), and the
HN fragment for the translocation of the L-chain across the
vesicular membrane (4, 7). Although tetanus and botulinum toxins bind
at the nerve terminals of the neuromuscular motor junction they
exhibit different clinical symptoms. BoNTs act locally at the
peripheral nervous system by disrupting neurotransmitter release and
causing flaccid paralysis. In contrast, TeNT activity occurs at the
central nervous system, following retrograde transport from the axon
terminal to inhibitory neurons within the spinal cord. Proteolytic
cleavage of vesicle-associated membrane protein by TeNT L-chain blocks neurotransmitter release (3), by preventing formation of the synaptic
SNARE complex (8).
A two-receptor model incorporating ganglioside and protein receptors
has been proposed to explain the distinct sites of action of TeNT and
BoNTs (5, 9, 10). These models attempt to explain the experimental
observations that are largely inconsistent with a sole ganglioside
receptor, such as the difference in the binding affinity measured
in vivo (10 Ganglioside molecules are a class of glycosphingolipids found in
high percentages in the membranes of neuronal cells (18). Most of them
contain a common "core" (GM1 ganglioside) consisting of
Gal( The x-ray crystal structure of the HC fragment (27, 28)
shows that the protein has two domains. The amino-terminal domain has a
jelly-roll lectin-like fold and the carboxyl-terminal is a Ganglioside Synthesis--
The fully protected derivative of
GT1b oligosaccharide (290 mg), compound 27 in Ishida et al.
(32), was converted into the desired product (87 mg, 53% in three
steps) by hydrogenolytic removal of benzyl groups over Pd-C catalyst,
de-O-acylation with sodium methoxide in methanol and
subsequent saponification of methyl esters and lactone group by
treatment with 0.2 M KOH. The product was purified using a
Sephadex LH-20 (MeOH) column.
Protein Expression and Purification--
TeNT HC
protein was prepared from Escherichia coli BL21 (pKS1) by
induction with isopropyl-1-thio- Crystallization--
Protein (2.8 mg/ml) and the GT1b analogue
(GT1b-
After a few days, plate-like crystals (thickness less than 10 µm)
grew in the first conditions and rod-like crystals (thickness about 50 µm) in the second conditions. X-ray diffraction data from the
plate-like crystals were collected at the ESRF microfocus beamline
(ID13, Crystal Structure Determination--
For both crystal forms,
molecular replacement was performed using AMoRe (34) with the
HC coordinates of the lactose-HC complex (PDB
access code 1dll) (28) as the search model. After rigid body refinement
of the protein (both crystal forms), several cycles of refinement using
Refmac (35) and rebuilding with QUANTA, the carbohydrate ligand was
modeled into weighted difference electron density
(2mFo
Further rebuilding and refinement proceeded using QUANTA, Refmac (35),
and ARP (37). The protein structure geometry was analyzed using
PROCHECK (38). Buried surfaces were calculated using GRASP (39) with a
1.4-Å probe radius.
The initial three-dimensional model of GT1b-OS was constructed with the
SWEET software (40) and superimposed on the GT1b- The complex of HC + GT1b-
The Crystal Structure of Tetanus Toxin Hc Fragment
Complexed with a Synthetic GT1b Analogue Suggests Cross-linking between
Ganglioside Receptors and the Toxin*
,,,
Department of Chemistry, University of
Glasgow, Glasgow, G12 8QQ, Scotland, the § Department of
Applied Bio-organic Chemistry, Gifu University, Gifu 501-1193, Japan,
and the ¶ Department of Biochemistry, Imperial College of Science
and Technology Medicine, London SW7 2AZ, United Kingdom
ABSTRACT
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
12 M) and in
vitro (106 to 108 M) (5,
11, 12) and the low and high affinity sites that have been measured in
rat brain membranes (13, 14). Although synaptotagmin in combination
with gangliosides has been shown to act as a receptor for BoNT/A, -B,
and -E (15-17) characterization of the protein receptors for other
toxins is extremely limited. A putative receptor for TeNT has been
described, but the protein(s) remain to be characterized at the
molecular level (9, 10).
1-3)GalNAc(1-4)(NeuAc(
2-3))Gal(1-4)Glc(1-1)Cer
to which one or more N-acetylneuraminic acids (sialic acids)
are bound (18) (Fig. 1a). They
were the first membrane components found to have BoNT and TeNT binding
activity (19-21). Later studies have identified the gangliosides of
series b, especially GT1b (Fig. 1a) and GD1b, to have the
highest affinity (11, 22-26). Binding studies of TeNT to brain
membranes and purified gangliosides have shown that the
Gal(1-3)GalNAc(1-4)(NeuAc(2-8)NeuAc(2-3))Gal(1-4) moiety (i.e. Gal4-GalNAc3-Sia7-Sia6-Gal2 in Fig.
1a) is essential for the binding of TeNT or HC
fragment (22, 23).
View larger version (37K):
[in a new window]
Fig. 1.
Schematic representations of a, GT1b
ganglioside. The monosaccharide subunit labels used in this paper are
shown. GD1b lacks Sia5. The shaded area corresponds to the
GM1 ganglioside and the boxed outline encloses the portion
of GT1b essential for binding TeNT. b, the oligosaccharide
of the synthetic GT1b ganglioside analogue (GT1b- A) used in this
study. Sia6 is the -anomer. The ceramide of the cellular GT1b
has been replaced by CH2CH2Si(Me)3
to increase solubility.
-trefoil.
The COOH-terminal domain of the heavy chain of BoNT/A (29) has been
shown to have a similar structure. Biochemical and mutational studies
(4, 6, 30) with TeNT and native and recombinant HC
fragments have shown that the -trefoil domain of the HC
fragment contains the ganglioside and cell-binding site(s). Photoaffinity labeling studies have implicated the area near residue His1293 as a site where ganglioside binds (31). Our own
studies on the crystal structure of TeNT HC complexed with
lactose (Gal(1-4)Glc) supports this association (28). Further
structural studies on HC crystals soaked in solutions
containing the carbohydrate units of gangliosides (galactose,
N-acetylgalactosamine and sialic acid) (28), identified
three more areas where the monomer sugar units bound. These results
suggested that additional sites on the HC molecule may be
involved in recognition of ganglioside receptors and that a single
ganglioside could bind simultaneously to more than one TeNT molecule.
Our attempts to crystallize a complex between TeNT HC and
native GT1b ganglioside have not been successful. However, crystals
could be grown of a complex formed between TeNT HC and a
synthetic GT1b analogue. This analogue has the ceramide replaced by a
2-(trimethylsilyl)ethyl group and a 2-3 linkage from the disialic
acid arm to the central galactose unit, i.e. Sia6 is the
-anomer (Fig. 1b). The crystal structure of this complex
shows cross-linking between the GT1b analogue and TeNT HC.
MATERIALS AND METHODS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
-D-galactopyranoside to 1 mM for 4 h as described by Sinha et al.
(33). Cells were then harvested, lysed by sonication, and the
HC protein purified by nickel affinity chromatography using
Ni-NTA resin (Qiagen, Crawley, Sussex, United Kingdom) followed by
gel-filtration on a Superdex-200 column. The protein was concentrated
to 2.8 mg/ml using a 30,000 cutoff MicrocepTM
microconcentrator (Pal Gelman Sciences) pretreated overnight with 10%
glycerol to prevent nonspecific interactions of protein with the membrane.
A) were mixed in a range of final molar ratios (1:2 to1:10).
Crystals were produced from all mixtures by vapor diffusion against a
1-ml reservoir solution containing either 20% PEG 4K and 0.2 M imidazole-malate (pH 7.0), or 20% PEG 8K, 0.2 M Na2KPO4, and 0.1 M
TRIS (pH 8.5) using a sitting drop containing 2 µl of the
HC + heptasaccharide solution and 2 µl of reservoir solution.
= 0.782 Å) with a Mar CCD detector set at a distance
of 170 mm from the crystal. With an initial exposure time of 5 s
per frame, 40 frames were collected from a single crystal with
intervals of 1° of rotation about an axis perpendicular to the x-ray
beam. The crystal was then translated 4 times to reduce the effects of
damage caused by the intense beam. Following the first translation 50 frames were collected with 5-s exposures; then 23 frames with 20-s
exposures; then 20 frames with 10-s exposures and finally 40 frames,
with 10-s exposures. Each frame covered 1° rotation of the crystal.
Data from the rod-like crystal were collected at the Daresbury SRS,
station 7.2 ( = 1.488 Å) with a Mar240 image plate detector
set at a distance of 173 mm from the crystal. An exposure at a 10,000 dose/frame and angular interval of 1° was used. 173 frames were
collected from a single crystal. All data sets were collected under
cryo-cooling conditions (100 K). The crystals were placed briefly in a
well of cryo-protectant (15% PEG 400) then collected in a loop and
flash-cooled in a stream of dry nitrogen at 100 K. Both crystal forms
were monoclinic, space group P21, but the a cell
dimension was doubled in the plate-like crystals, which contain two
molecules in the asymmetric unit as a consequence. Table I shows the
data statistics.
DFc and
mFoDFc) maps (Fig.
2). Omit maps were used for building
ambiguous areas. The initial model and the restraints dictionary for
the carbohydrate were generated using coordinates for the
monosaccharide units of the ligand taken from the Cambridge Structural
Data base (CSD) and the Protein Data Bank. -Neuraminic acid (CSD
code ANEU) was used for the -anomeric sialic acid,
N-acetyl--galactosamine (CSD code AOGAPY) for
N-acetylgalactosamine, -lactose (CSD code BLACTO) for
galactose and glucose, and coordinates for -anomeric sialic acid
were taken from the structure of the complex of cholera toxin with GM1
ganglioside (Protein Data Bank ID 2chb (36)).
View larger version (57K):
[in a new window]
Fig. 2.
Stereo view of a weighted
(mFo DFc, calculated
by Refmac (35)) difference electron density map of the
GT1b-A contoured at
2
, calculated before building the ganglioside
model. The clear density allows the unambiguous location of
all the subunits GT1b. The refined GT1b-A coordinates are shown with
the disordered terminal 2-(trimethylsilyl)ethyl group.
A structure using
LSQKAB (41). The fit was improved with cycles of manual adjustment to
the glycosidic torsion angles of the GT1b-OS (QUANTA) followed by
superposition with LSQKAB. Excluding Sia6, the r.m.s. difference
between the overlapped oligosaccharides is 1.3 Å. No energy
minimization was performed. The figures were prepared using
SETOR (42) and GRASP (39) and were edited and composed using the GIMP
(www.gimp.org/).
RESULTS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
A crystallizes in two
forms: one with two molecules in the asymmetric unit and the other with
one. The final models of the two crystal forms of HC
consist of residues 865 to 1315 of TeNT and all the carbohydrate
residues. The terminal 2-(trimethylsilyl)ethyl group is disordered in
all copies of the ganglioside and is modeled with two conformations.
The model in the larger cell has 240 waters and in the smaller cell 120 (Table I). For both forms 80% of the
residues are in the most favored regions of the Ramachadran plot as
defined by PROCHECK (38) and no residues are in the disallowed
regions.
Data statistics and refinement
The r.m.s. differences in the C atom positions of the three copies
of the protein (excluding flexible loop residues 865-875, 937-945,
977-989, 1060-1070, 1177-1194, and 1201-1209), are 0.57, 0.62, and
0.60 Å for the pairs A and B, A and C, and B and C, respectively (Fig.
3). The three independent copies of the
protein in the two crystal forms provide the same general information on the HC-carbohydrate interactions. All have two binding
sites on the HC molecule. One interacts with the
Gal4-GalNAc3 and the other with Sia7-Sia6 moieties of different
ganglioside molecules (Table II), so that
the ganglioside acts as a cross-linker for the protein (Figs.
4 and
5).
|
|
|
|
Gal4-GalNAc3-binding Site--
The Gal-GalNAc-binding site is a
narrow groove formed by the side chains of residues Trp1289
and His1271 with Tyr1290 forming the base (Fig.
6a). On one side of the
groove, Trp1289 packs against His1293, while on
the other His1271 packs against Phe1218. A
hydrogen bond (3.5 Å in molecule A; 3.4 Å in molecule B; 3.5 Å in
molecule C) from His1293 ND1 to the main chain carbonyl of
Trp1289 contributes in holding the Trp in position.
Tyr1290 forms only hydrophobic contacts. The location of
His1293 is in agreement with photoaffinity labeling studies
that place it close to a GD1b-binding site (31). This is also the site where lactose binds to HC (28). The Gal4-GalNAc3 units of
the oligosaccharide interact extensively with this site. The
hydrophobic faces of both sugar rings are stacked against the indole
ring of Trp1289 and the polar faces are hydrogen bonded to
the protein (Table II). There are some differences in the number of
hydrogen bonds found in the three copies of the complexes (Table II)
but because of the resolution of the structures the significance of
these differences is uncertain. Common hydrogen bonds are formed
between the side chain of His1271 and OH-6, OH-4 and O-5 of
Gal4 and between the main chain carbonyl oxygen of Thr1270
and OH-4 of Gal4. GalNAc3 interacts via a hydrogen bond between OH-4
and Asp1222 OD and between OH-4 and His1271.
The stacking of the hydrophobic face of galactose against aromatic rings and the involvement of OH-4 in specificity determination are
common characteristics in carbohydrate-protein interactions (43, 44).
There are also water-mediated hydrogen bonds that are not common in all
copies. The buried surface area between protein and carbohydrate for
the Gal4-GalNAc3-binding site is 694, 740, and 700 Å2 for
copies A, B, and C, respectively.
|
Sia7-Sia6-binding Site-- This binding site is a shallow pocket, involving residues Arg1226, Asn1216, Asp1214, Asp1147, and Tyr1229 in interactions with the disialic acid moiety (Fig. 6b). The shape of the binding site and the mode of sialic acid-protein interactions corresponds to the second class described by Stehle and Harrison (45) (one side of the sialic acid ring is in contact with the protein, O-4 is exposed and the glycerol group interacts with the protein). Again, there are differences in the number of H-bonds found in the copies of the complexes (Table II). Common interactions are the hydrogen bonds between OD-1 and OD-2 of Asp1147 and O-4 and the acetamido-N-5 of Sia6 and between ND-2 of Asn1216 and O-10 of Sia6. The terminal Sia7 interacts more than Sia6 with the protein. The interactions of Sia7 observed in all molecules are a salt bridge between Arg1226 and the sialic acid carboxylate group and hydrogen bonds between O-1A and the amide NH of Asn1216; between O-4 and the carbonyl oxygen of Asp1214; and between OH-8 and Tyr1229 hydroxyl group. These interactions of the sialic acid carboxylate and glycerol groups with the protein have been found to be specificity determinants in lectins (43, 44). The buried surface area between protein and carbohydrate for the Sia6-Sia7-binding site is 920, 956, and 903 Å2 for copies A, B, and C, respectively.
Carbohydrate Conformation--
The 
and 
angles of the
glycosidic bonds of the oligosaccharide molecule are listed in Table
III. The Gal4(1-3)GalNAc3 and GalNAc3(1-4)Gal2 glycosidic bonds, given the resolution of the structure, appear to be relatively flexible, as observed in the NMR
structure of the similar GD1b ganglioside (46). The values of the
Sia5(2-3)Gal4, and Sia7(2-8)Sia6 angles lie in minima predicted
by theoretical calculations (47). There are intra-carbohydrate hydrogen
bonds (Table II) and water mediated ones, but as they are not present
in all copies, we believe they are not critical in determining the
conformation of GT1b-A.
|
Comparison with the HC-Lac, NGA, and Sia Soaks--
We
have compared the HC-GT1b-A structure with our
previously reported structures of HC complexed with
carbohydrates (28). The overlays used C atoms within spheres of
radius 10 Å around the binding site of the lactose complex and 12 Å for the N-acetylgalactosamine and sialic acid complexed
structures. The r.m.s. deviations ranged from 0.29 to 0.35 Å for the
lactose complex (0.29, 0.30, and 0.35 for molecules A, B, and C,
respectively), 0.35-0.5 Å for the N-acetylgalactosamine complex (0.37, 0.35, and 0.50 for molecules A, B, and C, respectively) and 0.37-0.51 Å for sialic acid complex (0.37, 0.37, and 0.51 for
molecules A, B, and C, respectively). The main difference between the
HC-GT1b-A and N-acetylgalactosamine and
sialic acid complex structures is the conformation of
Asn1220 which has moved about 1.0 Å in order to leave
space for Glc1 of the ganglioside. The galactose sugar of the lactose
complex packs against Trp1289 and forms similar hydrogen
bonds as Gal4 of GT1b-A although His1271 is moved by
about 0.4 Å in the lactose complex structure leading to a loss of the
hydrogen bond observed between Gal4 and ND1His1271 in the
HC -GT1b-A complex.
The binding of N-acetylgalactosamine mimics that of Sia6.
Hydrogen bonds are formed to Arg1226 via a water molecule
while a SO
A. These
interactions are the direct hydrogen bonds of O-6 with NH2
Asp1226, O-1B with NH1 Asp1226, O-1B with NH
Asn1216, and the water-mediated H-bonds of O-4 with OD-1
and OD-2 Asp1147, O-2 to CO Asp1214, and O-1A
to CO Asp1214.
Comparison of HC with BoNT/A--
An overlay of the
C atoms of this structure with those from the homologous
HC fragment of BoNT/A shows a spatial coincidence in the
Gal4-GalNAc3-binding site (when residues contained in a sphere of
radius 12 Å around Gal4 are overlaid, the r.m.s. differences between
C atoms are: 0.77, 0.74, and 0.76 Å for BoNT/A to TeNT HC molecules A, B, and C, respectively). The key residues
Trp1289, His1271, His1293, and
Phe1218 of TeNT have counterparts in BoNT/A residues
Trp1265, His1252, Gln1269, and
Phe1254, respectively. The coincidence of TeNT
Phe1218 and BoNT/A Phe1254 is noteworthy as the
side chains point into the space from opposite directions (Fig.
7). In addition, the main chain hydrogen
bonds contributing to the shape of the Gal4-GalNAc3-binding site in TeNT are the same in both structures, strongly indicating that this
region of BoNT/A will also bind these ganglioside units. Gln1269 in BoNT/A is able to play the same role as
His1293 in TeNT with a similar stacking of the side chain
against the Trp ring. Fluorescence quenching experiments (48) have been interpreted (49) as evidence for the presence of Trp1265 in
the ganglioside-binding site of BoNT/A. Data from neutralizing antibodies (50) also implicate this region (residues 1266-1272) of
BoNT/A in ganglioside binding. A similar cleft is formed by Trp1261 and His1240 in the structure of BoNT/B
(51). A crystallographic study of BoNT/B complexed with sialyllactose
shows the carbohydrate bound to this cleft, but with the sialic acid
intercalating between the Trp and His residues (51). This mode of
binding contrasts with that of TeNT where the structures of complexes
with a number of carbohydrates containing sialic acid
(28)2 always show the sialic
acid binding in the Sia7-Sia6 site with a conservation of the salt
bridge between the carboxylate and Arg1226 and the hydrogen
bonds between Asp1147 and the sialic acid. These data, and
the lack of structural similarity between the BoNT and TeNT structures
at the TeNT disialic acid-binding site could indicate fundamental
differences in the way the toxins interact with gangliosides.
Alternatively, the binding of sialyllactose to BoNT/B may not reflect
the binding of the complete ganglioside. Our earlier studies (28)
showed that the positions of binding of some monosaccharide units to
TeNT HC differs to that found for the ganglioside
analogue.
|
Comparison with Other Glycosphingolipid-binding
Toxins--
Cholera toxin and Shiga-like toxin are
glycosphingolipid-binding toxins where structural data are available
for the proteins complexed with their carbohydrate receptors. The
cholera toxin receptor, ganglioside GM1, is similar to GT1b (Fig. 1)
but lacks the two terminal sialic acids (Sia5 and Sia7 in Fig.
1a). In contrast with the two ganglioside sites in TeNT, in
cholera toxin (36) there is one binding site in each of the five
-subunits, with the terminal galactose and sialic acid forming the
main interactions. This is very similar to the TeNT
HC-GT1b-A interactions where the galactose Gal4 and
disialo-group Sia7-Sia6 (Fig. 1) make almost all the interactions with
the protein. There is some small contribution from the GalNAc3 unit but
no contributions from the Gal2-Glc1 disaccharide portion. In both
toxins the galactose is packed against a Trp ring and is extensively
hydrogen bonded to the protein. The sialic acid-binding site in cholera
toxin is a shallow groove with only one side of the carbohydrate ring
making contacts with the protein. In contrast, tetanus toxin interacts
extensively with the terminal sialic acid (Sia7) of the ganglioside via
its carboxylate, OH-4, acetamido, and glycerol groups. The low affinity of tetanus toxin for GM1 (22, 23) is due to the loss of these strong
interactions with sialic acid.
In Shiga-like toxins the glycosphingolipid receptor,
Gal(1-4)Gal(1-4)Glc(1-1)-ceramide, does not contain sialic
acid. There are three binding sites per B-subunit. In two binding sites
galactose shows aromatic packing against Trp or Phe residues with
extensive hydrogen bonds to the protein, as observed in tetanus toxin.
In the third there is no aromatic stacking.
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The crystal structure of a complex between the HC fragment of TeNT and a synthetic analogue of GT1b reported here provides the first direct structural evidence of the interaction between the protein toxin and ganglioside. The observation of two separate binding sites on both the protein and carbohydrate components is unexpected, but nonetheless demonstrates the specificity of the interaction and is consistent with published biochemical data. Similar interactions were described in our earlier report of the crystal structures of TeNT HC soaked with lactose, N-acetylgalactosamine, and sialic acid (28).
These structural data demonstrate that two sites on the GT1b molecules, the Gal4-GalNAc3 and Sia7-Sia6 groups, provide the key interactions for the TeNT-ganglioside recognition. The results are entirely consistent with the similar binding affinites observed for GT1b and GD1b gangliosides (22, 23) which differ only in GT1b possessing a terminal sialic acid on the GalNAc3 unit. The structure shows no significant interaction between this terminal sialic acid (Sia5) and the HC fragment. In contrast, the extensive interactions of the terminal sialic acid (Sia7) with the HC fragment explain why in GM1, which lacks this sugar, the affinity of the binding is considerably lower (1, 2, 23, 24).
Recent biochemical characterizations of mutant HC molecules
are consistent with this structure (33, 52). Mutant HC
containing a deletion of residues 1214-1219 is severely impaired in
its ability to bind GT1b, having 0.5% of the activity of wild-type
HC protein. This mutant lacks Asp1214 and
Asn1216, residues which interact with the disialic acid
present in GT1b. Another mutant containing deletions in the
Gal4-GalNAc3-binding site, e.g.
(His1271-Asp1282) also shows reduced GT1b
binding activity (1% of wild-type) and both mutants have reduced
binding to primary motorneurons, suggesting that ganglioside
binding is essential for binding to target cells. Replacement of
His1293 with either Ala or Ser also reduces GT1b binding
(12-43% of wild-type). This residue does not interact directly with
the ganglioside, but forms part of the structural framework of the
GalNAc4-Gal3-binding site. The loss of binding correlates with the
reduction in volume of the side chain. A similar effect has been
reported for mutations of Tyr1290 to Phe, Ala, or Ser (52).
This residue forms the base of the binding site, making hydrophobic
interactions with the carbohydrate. Changes in the volume and
hydrophobicity of the side chain would disturb the structural integrity
of the binding site. Similar effects explain the mutagenesis data of
Halpern and Loftus (6) who showed that deletion of the COOH-terminal 5 or 10 residues reduced binding to both ganglioside GT1b and to primary
spinal cord neurones. These residues do not interact with the GT1b
analogue, and the reduction in ganglioside binding is probably due to
alteration in the structure or stability of the HC molecule.
Simultaneous binding through both groups on the ganglioside involves
more than one protein molecule and suggests that cross-linking might
occur in vivo. This would depend on the membrane-attached, wild-type gangliosides making the same interactions as in the crystal
complex. The carbohydrate of the synthetic ganglioside used in this
work differs from that of wild-type GT1b only in the Sia6-Gal2 linkage
which is , in the wild-type, and , in this work (Fig. 1). This
difference will not affect the binding of the Gal-GalNAc units of the
ganglioside to the protein. With regard to the binding of the disialyl
group, since identical interactions between the sialic acid units and
proteins are observed in the crystal structure of a complex between
TeNT HC and disialyllactose, where the Sia-Gal linkage is
,2 this mode of binding is not influenced by the
anomeric stereochemistry of the Sia-Gal linkage.
Although wild-type GT1b can bind at each site on the protein,
cross-linking will occur only if the relative positions of the two arms
of the ganglioside can interact simultaneously with more than one
protein. A simple modeling experiment suggests this is possible (Fig.
8). A model of GT1b-OS can be fitted to
the structure of GT1b-A to retain the cross-linking found in the
crystal, and with the formation of the same H-bonds and salt bridge
between Sia7 and the protein. There are no steric clashes in this
model.
|
The biological consequences of multivalent cross-linking would be to
increase the binding affinity of the protein to the ganglioside and to
induce a clustering of the toxin on the cell surface. Indeed binding of
TeNT to the cell surface is characterized by punctate staining in a
variety of neuronal cells types, including rat dorsal root ganglia (30)
and mouse spinal cord neurons (4, 25). While this punctate staining has
been inferred (4) to reflect the distribution of distinct zones where
neurotransmitter release and endocytosis are proposed to take place, it
could also could be explained in part by clustering of toxin with
gangliosides or ganglioside-protein complexes. The clustering effect
may be important in enhancing the process of internalization.
| |
ACKNOWLEDGEMENTS |
|---|
We thank ESRF (ID13) and SRS-Daresbury (7.2) station personnel for assistance with data collection and A. Roszak and S. Prince for helpful discussions.
| |
FOOTNOTES |
|---|
* This work was supported by Grants 053570 and 051615 from The Wellcome Trust.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The atomic coordinates and the structure factors (code 1fv2 and 1fv3) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).
To whom correspondence should be addressed. Tel.:
44-141-3305954; Fax: 44-141-3304888; E-mail:
neil@chem.gla.ac.uk.
Published, JBC Papers in Press, June 19, 2001, DOI 10.1074/jbc.M103285200
2 C. Fotinou, P. Emsley, I. Black, N. F. Fairweather, and N. W. Isaacs, unpublished data.
| |
ABBREVIATIONS |
|---|
The abbreviations used are:
TeNT, tetanus
neurotoxin;
HC, COOH-terminal domain of heavy chain of
TeNT;
BoNT, botulinum neurotoxin;
GalNAc, N-acetylgalactosamine;
Sia, sialic acid;
Cer, ceramide;
PEG, polyethylene glycol;
GD1b, Gal(1-3)GalNAc(1-4)(NeuAc(2-8)NeuAc(2-3))Gal(1-4)Glc(1-1)Cer;
GT1b, NeuAc(2-3)Gal(1-3)GalNAc(1-4)(NeuAc(2-8)NeuAc(2-3))Gal(1-4)Glc(1-1)Cer;
GT1b-OS, oligosaccharide part of the GT1b ganglioside;
GT1b-A, synthetic GT1b oligosaccharide analogue;
r.m.s., root mean
square.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
| 1. | Montecucco, C., and Schiavo, G. (1994) Mol. Microbiol. 13, 1-8 |
| 2. | Herreros, J., Lalli, G., Montecucco, C., and Schiavo, G. (1999) in The Comprehensive Sourcebook of Bacterial Protein Toxins (Alouf, J. E. , and Freer, J. H., eds) , pp. 202-228, Academic Press |
| 3. | Schiavo, G., Benfenati, F., Poulain, B., Rossetto, O., de Laureto, P. P., DasGupta, B. R., and Montecucco, C. (1992) Nature 359, 832-839 |
| 4. | Lalli, G., Herreros, J., Osborne, S. L., Montecucco, C., Rossetto, O., and Schiavo, G. (1999) J. Cell Sci. 112, 2715-2724 |
| 5. | Montecucco, C. (1986) Trends Biochem. Sci. 11, 314-317 |
| 6. | Halpern, J. L., and Loftus, A. (1993) J. Biol. Chem. 268, 11188-11192 |
| 7. | Halpern, J. L., and Neale, E. A. (1995) Curr. Top. Microbiol. Immunol. 195, 221-241 |
| 8. | Sollner, T., Whitehart, S. W., Brunner, M., Erdjument-Bromage, H., Geromanos, S., Tempst, P., and Rothman, J. E. (1993) Nature 362, 318-324 |
| 9. | Herreros, J., Lalli, G., Montecucco, C., and Schiavo, G. (2000) J. Neurochem. 74, 1941-1950 |
| 10. | Herreros, J., Lalli, G., and Schiavo, G. (2000) Biochem. J. 347, 199-204 |
| 11. | Janshoff, A., Steinem, C., Sieber, M., elBaya, A., Schmidt, M. A., and Galla, H. J. (1997) Eur. Biophys. J. 26, 261-270 |
| 12. | Johnson, E. A. (1999) Annu. Rev. Microbiol. 53, 551-575 |
| 13. | Pierce, E. J., Davison, M. D., Parton, R. G., Habig, W. H., and Critchley, D. R. (1986) Biochem. J. 236, 845-852 |
| 14. | Bakry, N., Kamata, Y., Sorensen, R., and Simpson, L. L. (1991) J. Pharmacol. Exp. Ther. 258, 613-619 |
| 15. | Nishiki, T., Kamata, Y., Nemoto, Y., Omori, A., Ito, T., Takahashi, M., and Kozaki, S. (1994) J. Biol. Chem. 269, 10498-10503 |
| 16. | Li, L., and Singh, B. R. (1998) J. Nat. Toxins 7, 215-226 |
| 17. | Kozaki, S., Kamata, Y., Watarai, S., Nishiki, T., and Mochida, S. (1998) Microb. Pathog. 25, 91-99 |
| 18. | Karlsson, K. A. (1998) Acta Biochim. Pol. 45, 429-438 |
| 19. | Van Heyningen, W. E., and Miller, P. (1961) J. Gen. Microbiol. 24, 107-119 |
| 20. | Simpson, L. L., and Rapport, M. M. (1971) J. Neurochem. 18, 1341-1343 |
| 21. | Van Heyningen, W. E. (1974) Nature 249, 415-417 |
| 22. | Holmgren, J., Elwing, H., Fredman, P., and Svennerholm, L. (1980) Eur. J. Biochem. 106, 371-379 |
| 23. | Angstrom, J., Teneberg, S., and Karlsson, K. A. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 11859-11863 |
| 24. | Fishman, P. H. (1982) J. Membr. Biol. 69, 85-97 |
| 25. | Williamson, L. C., Bateman, K. E., Clifford, J. C. M., and Neale, E. A. (1999) J. Biol. Chem. 274, 25173-25180 |
| 26. | Kitamura, M., Takamiya, K., Aizawa, S., Furukawa, K., and Furukawa, K. (1999) Biochim. Biophys. Acta 1441, 1-3 |
| 27. | Umland, T. C., Wingert, L. M., Swaminathan, S., Furey, W. F., Schmidt, J. J., and Sax, M. (1997) Nat. Struct. Biol. 4, 788-792 |
| 28. | Emsley, P., Fotinou, C., Black, I., Fairweather, N. F., Charles, I. G., Watts, C., Hewitt, E., and Isaacs, N. W. (2000) J. Biol. Chem. 275, 8889-8894 |
| 29. | Lacy, D. B., Tepp, W., Cohen, A. C., DasGupta, B. R., and Stevens, R. C. (1998) Nat. Struct. Biol. 5, 898-902 |
| 30. | Figueiredo, D., Turcotte, C., Frankel, G., Li, Y., Dolly, O., Wilkin, G., Marriott, D., Fairweather, N., and Dougan, G. (1995) Infect. Immun. 63, 3218-3221 |
| 31. | Shapiro, R. E., Specht, C. D., Collins, B. E., Woods, A. S., Cotter, R. J., and Schnaar, R. L. (1997) J. Biol. Chem. 272, 30380-30386 |
| 32. | Ishida, H.-K., Ishida, H., Kiso, M., and Hasegawa, A. (1994) Tetrahedron Asymmetry 5, 2493-2512 |
| 33. | Sinha, K., Box, M., Lalli, G., Schiavo, G., Schneider, H., Groves, M., Siligardi, G., and Fairweather, N. (2000) Mol. Microbiol. 37, 1041-1051 |
| 34. | Navaza, J. (1994) Acta Cryst. Sec. A 50, 157-163 |
| 35. | Murshudov, G. N., Vagin, A. A., and Dodson, E. J. (1997) Acta Cryst. Sec. D 53, 240-255 |
| 36. | Merritt, E. A., Sarfaty, S., van den Akker, F., L'Hoir, C., Martial, J. A., and Hol, W. G. J. (1994) Protein Sci. 3, 166-175 |
| 37. | Lamzin, V. S., and Wilson, K. S. (1993) Acta Cryst. Sec. D 49, 129-147 |
| 38. | Laskowski, R. A., Macarthur, M. W., Moss, D. S., and Thornton, J. M. (1993) J. Appl. Cryst. 26, 283-291 |
| 39. | Nicholls, A., Sharp, K. A., and Honig, B. (1991) Proteins 11, 281-296 |
| 40. | Bohne, A., Lang, E., and von der Lieth, C. W. (1998) J. Mol. Model. 4, 33-43 |
| 41. | Collaborative Computational Project Number 4. (1994) Acta Cryst. Sec. D 50, 760-763 |
| 42. | Evans, S. V. (1993) J. Mol. Graph. 11, 134-138 |
| 43. | Rini, J. M. (1995) Annu. Rev. Biophys. Biomol. Struct. 24, 551-577 |
| 44. | Weis, W. I., and Drickamer, K. (1996) Annu. Rev. Biochem. 65, 441-473 |
| 45. | Stehle, T., and Harrison, S. C. (1997) The EMBO J. 16, 5139-5148 |
| 46. | Acquotti, D., Fronza, G., Ragg, E., and Sonnino, S. (1991) Chem. Phys. Lipids 59, 107-125 |
| 47. | Veluraja, K., and Rao, V. S. R. (1984) Carbohydr. Polymers 4, 357-375 |
| 48. | Kamata, Y., Yoshimoto, M., and Kozaki, S. (1997) Toxicon 35, 1337-1340 |
| 49. | Lacy, D. B., and Stevens, R. C. (1999) J. Mol. Biol. 291, 1091-1104 |
| 50. | Kubota, T., Watanabe, T., Yokosawa, N., Tsuzuki, K., Indoh, T., Moriishi, K., Sanda, K., Maki, Y., Inoue, K., and Fujii, N. (1997) Appl. Envirom. Microbiol. 63, 1214-1218 |
| 51. | Swaminathan, S., and Eswaramoorthy, S. (2000) Nat. Struct. Biol. 7, 693-699 |
| 52. | Sutton, J. M., Chow-Worn, O., Spaven, L., Silman, N. J., Hallis, B., and Shone, C. C. (2001) FEBS Lett. 493, 45-49 |
| 53. | Leslie, A. G. W. (1992) Joint CCP4 and ESF-EACMD Newsletter on Protein Crystallography , Daresbury Laboratory, Warrington, UK |
| 54. | Evans, P. R. (1993) in Proceedings of the CCP4 Study Weekend (Sawyer, L. , Isaacs, N. W. , and Bailey, S., eds) , pp. 114-122, SERC Daresbury Laboratory, Warrington, UK |
Copyright © 2001 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Fischer, C. Garcia-Rodriguez, I. Geren, J. Lou, J. D. Marks, T. Nakagawa, and M. Montal Molecular Architecture of Botulinum Neurotoxin E Revealed by Single Particle Electron Microscopy J. Biol. Chem., February 15, 2008; 283(7): 3997 - 4003. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Rummel, T. Eichner, T. Weil, T. Karnath, A. Gutcaits, S. Mahrhold, K. Sandhoff, R. L. Proia, K. R. Acharya, H. Bigalke, et al. Identification of the protein receptor binding site of botulinum neurotoxins B and G proves the double-receptor concept PNAS, January 2, 2007; 104(1): 359 - 364. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Attrill, A. Imamura, R. S. Sharma, M. Kiso, P. R. Crocker, and D. M. F. van Aalten Siglec-7 Undergoes a Major Conformational Change When Complexed with the {alpha}(2,8)-Disialylganglioside GT1b J. Biol. Chem., October 27, 2006; 281(43): 32774 - 32783. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Qazi, D. Sesardic, R. Tierney, Z. Soderback, D. Crane, B. Bolgiano, and N. Fairweather Reduction of the Ganglioside Binding Activity of the Tetanus Toxin HC Fragment Destroys Immunogenicity: Implications for Development of Novel Tetanus Vaccines. Infect. Immun., August 1, 2006; 74(8): 4884 - 4891. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Moore, S. E. S. Bailey, Z. Benmechernene, C. Tzitzilonis, N. J. E. Griffiths, M. Virji, and J. P. Derrick Recognition of Saccharides by the OpcA, OpaD, and OpaB Outer Membrane Proteins from Neisseria meningitidis J. Biol. Chem., September 9, 2005; 280(36): 31489 - 31497. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. X. Moss, S. P. Matthews, D. J. Lamont, and C. Watts Asparagine Deamidation Perturbs Antigen Presentation on Class II Major Histocompatibility Complex Molecules J. Biol. Chem., May 6, 2005; 280(18): 18498 - 18503. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Rummel, T. Karnath, T. Henke, H. Bigalke, and T. Binz Synaptotagmins I and II Act as Nerve Cell Receptors for Botulinum Neurotoxin G J. Biol. Chem., July 16, 2004; 279(29): 30865 - 30870. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. S. Alphey, H. Attrill, P. R. Crocker, and D. M. F. van Aalten High Resolution Crystal Structures of Siglec-7. INSIGHTS INTO LIGAND SPECIFICITY IN THE SIGLEC FAMILY J. Biol. Chem., January 24, 2003; 278(5): 3372 - 3377. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. C. Yowler, R. D. Kensinger, and C.-L. Schengrund Botulinum Neurotoxin A Activity Is Dependent upon the Presence of Specific Gangliosides in Neuroblastoma Cells Expressing Synaptotagmin I J. Biol. Chem., August 30, 2002; 277(36): 32815 - 32819. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Nowakowski, C. Wang, D. B. Powers, P. Amersdorfer, T. J. Smith, V. A. Montgomery, R. Sheridan, R. Blake, L. A. Smith, and J. D. Marks Potent neutralization of botulinum neurotoxin by recombinant oligoclonal antibody PNAS, August 20, 2002; 99(17): 11346 - 11350. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |