|
|
|
|||||||
|
|||||||||
(Received for publication, December 26, 1995) From the
The binding site for batrachotoxin, a lipid-soluble neurotoxin
acting at Na
Batrachotoxin (BTX) ( Competitive binding studies with radiolabeled neurotoxin analogues
have distinguished five distinct receptor sites for neurotoxins on the
Na BTX binding to neurotoxin receptor site 2 is allosterically
modulated by other Na Neurotoxin receptor site 2 is present on
Na
Figure 1:
Structure of BTX
and its derivatives. The positions of
Figure 2:
Competitive displacement of
[
Figure 3:
Enhancement of specific
[
Figure 4:
Specific photolabeling of purified
Na
Figure 5:
Immunoprecipitation of peptides covalently
labeled by [
The site of BTX covalent labeling of the
Na
Figure 6:
Immunoprecipitation of peptides labeled by
[
Figure 7:
SDS-PAGE analysis of tryptic peptides from
[
Experiments like the one illustrated in Fig. 7were carried out with several preparations of purified and
reconstituted Na Examination of the amino acid sequence of the
Figure 8:
Amino
acid sequence of the batrachotoxin-binding region on the Na
Less extensive cleavage with trypsin alone yields peptides that
contain the recognition sites for anti-SP1 and anti-SP11 on the
intracellular, carboxyl-terminal side of transmembrane segment IS6 (Fig. 6A). These peptides must contain sequences
extending beyond Ser-504 (Fig. 8). In contrast, cleavage with V8
protease removes the recognition site for these two antibodies (Fig. 6B) by cleavage at one of the numerous Glu
residues in the SP1 sequence (Fig. 8). Cleavage of the SP1
peptide from transmembrane segment IS6 by treatment with V8 protease
and resistance of the SP28 peptide within the
We have found that in the
presence of the pyrethroid, RU51049, and the brevetoxin, PbTx-1, both
[
Further localization of the site of covalent labeling
by [ A more precise biochemical localization of the site of
covalent labeling by [ The photoreactive azide of
[
BTX binding is also affected by binding of
toxins to more distant regions of the Na The interactive nature of
BTX binding provides evidence that neurotoxin receptor site 2 is
located in a region of the Na
Volume 271,
Number 19,
Issue of May 10, 1996 pp. 11261-11267
©1996 by The American Society for Biochemistry and Molecular Biology, Inc.
Subunit (*)
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
channel receptor site 2, was localized
using a photoreactive, radiolabeled batrachotoxin derivative to
covalently label purified and reconstituted rat brain Na channels. In the presence of the brevetoxin 1 from Ptychodiscus brevis and the pyrethroid RU51049, positive
allosteric enhancers of batrachotoxin binding, a protein with an
apparent molecular mass of 240 kDa corresponding to the
Na channel 
subunit was specifically covalently
labeled. The region of the subunit specifically photolabeled by
the photoreactive batrachotoxin derivative was identified by antibody
mapping of proteolytic fragments. Even after extensive trypsinization,
an anti-peptide antibody recognizing an amino acid sequence adjacent to
Na channel transmembrane segment IS6 was able to
immunoprecipitate up to 70% of the labeled peptides. Analysis of a more
complete digestion with trypsin or V8 protease indicated that the
batrachotoxin receptor site is formed in part by a portion of domain I.
The identification of a specifically immunoprecipitated photolabeled
7.3-kDa peptide containing transmembrane segment S6 from domain I
restricted the site of labeling to residues Asn-388 to Glu-429 if V8
protease digestion was complete or Leu-380 to Glu-429 if digestion was
incomplete. These results implicate the S6 transmembrane region of
domain I of the Na channel subunit as an
important component of the batrachotoxin receptor site.
)is a steroidal alkaloid toxin
from skin secretions of South American frogs, Phyllobates
aurotaenia and Phyllobates terribilus, which are used by
the native Indians of Colombia to make poison blowdarts and
arrows(1) . It is one of the most toxic nonproteinaceous
substances known and is capable of inducing membrane depolarization at
concentrations in the low nanomolar range. The voltage-gated
Na channels of excitable membranes are the molecular
targets of BTX, and all aspects of Na channel function
are altered upon exposure to BTX: inactivation is blocked, single
channel conductance is decreased, voltage dependence of activation is
shifted to more negative potentials, and selectivity for Na is impaired (reviewed in (2) and (3) ). channel, including neurotoxin receptor site 2 which
is occupied by the full agonist BTX and the partial agonists
veratridine, aconitine, and grayanotoxin which modulate Na channel gating(2, 3) . The binding of BTX at
neurotoxin receptor site 2 results in stabilization of an open
conformation of the channel. At least four of the five receptor sites
have been shown to be located on the 260-kDa subunit, which is
composed of four homologous domains (I-IV) containing six
putative transmembrane segments (S1-S6; Refs. 4, 5).
Site-directed mutagenesis experiments have identified amino acid
residues of the subunit that are required for high affinity
binding of tetrodotoxin and saxitoxin at neurotoxin receptor site
1(6, 7, 8) . Peptide segments from neurotoxin
receptor site 3, which binds -scorpion toxins and sea anemone
toxins, and neurotoxin receptor site 5, which binds brevetoxins, have
been identified by photoaffinity labeling and peptide mapping with
sequence-specific antibodies(9, 10, 11) , and
individual amino acid residues that are required for high affinity
binding of -scorpion toxins and sea anemone toxins to receptor
site 3 have been identified by site-directed mutagenesis(12) . channel neurotoxins. The
-polypeptide toxins from scorpion and sea anemone(13) ,
the -cyano-pyrethroid
insecticides(3, 14, 15) , and brevetoxin (14, 15) all enhance BTX binding. Other compounds such
as tetrodotoxin and saxitoxin(16) , local
anesthetics(17, 18) , and the anticonvulsants
diphenylhydantoin and carbamazepine (19) decrease BTX-binding
affinity. These results indicate that the BTX-binding site is
particularly sensitive to conformational alterations induced at
distinct sites. channels expressed from subunit cDNA
alone(20, 21) . Mapping of the peptide segments of the
subunit that form receptor site 2 by photoaffinity labeling has
not been feasible because the ligands binding at that site have
relatively low affinity and are hydrophobic, resulting in low specific
binding values. These difficulties can be circumvented by taking
advantage of the allosteric enhancement of BTX binding by pyrethroids
and the brevetoxin analogue, PbTx-1. This combination of effectors can
enhance BTX binding in a purified and reconstituted Na channel preparation up to 1000-fold, reducing the K
to a value below 1 nM(15) . In this study, this combination of toxins is used
to enhance high affinity binding and covalent incorporation of a
photoreactive, radiolabeled BTX derivative,
[
H]BTX-OAB, into purified Na channels, and the major site of incorporation is identified by
peptide mapping with anti-peptide antibodies.
Materials
Ecolume was from ICN Biomedicals, Inc.
and Soluene was from DuPont NEN. Phosphatidylcholine and
phosphatidylethanolamine used for sodium channel reconstitution were
from Avanti Polar Lipids and L-1-tosylamido-2-phenylethyl
chloromethyl ketone-treated trypsin (TPCK-trypsin, from bovine
pancreas) was from Worthington. Protease type XVII-B (from Staphylococcus aureus strain V8) and trypsin inhibitor (from
soybean) were from Sigma. Prestained molecular weight markers and
precast mini-gels were from Novex. [H]BTX-B was
obtained from DuPont NEN at a specific activity of 51.5 Ci/mmol.
RU39568 and RU51049 are synthetic pyrethroid derivatives supplied by
Roussel Uclaf (Romainville, France). PbTx-1 was generously supplied by
Chiral Corp. (Miami, FL).Synthesis of
[
[H]BTX-OABH]BTX-OAB
was synthesized as described previously (22) from
[H]anthranilic acid (47 Ci/mmol) by selective
esterification of BTX-A.[
Binding reactions were initiated by
a 4-fold dilution of 50 µl of purified and reconstituted
NaH]BTX-B Binding to Reconstituted
Na Channels channel (5-10 pmol) in standard binding
medium(23) . Reconstituted channels were incubated at 25 °C
for 16 h with [H]BTX-B and other effectors as
shown in the figure legends. Binding reactions were stopped by addition
of choline wash medium, and samples were filtered through GF/F filters
as described previously(24) . Nonspecific binding was
determined in the presence of 300 µM veratridine.Sequence-directed Antibodies
Polyclonal antisera
were raised in rabbits against synthetic Na channel
peptides (SP) corresponding to sequences of the rat brain type IIA
subunit (4, 5) as described
previously(25, 26, 27) .Preparation of Antibodies Bound to Protein
A-Sepharose
Protein A-Sepharose was swollen for 20 min at 25
°C in 0.1 M sodium phosphate, pH 8.1, to give a final
concentration of 100 mg/ml. One ml of serum was added per 0.75 ml of
swollen protein A-Sepharose and mixed by rotation at 25 °C for 30
min or at 4 °C overnight. Supernatants were removed, and the
pellets were washed five times with 10 volumes of buffer S (10 mM Tris, adjusted to pH 7.4 with HCl, 150 mM NaCl). The
pellet was resuspended in 1 volume of buffer S and used for
immunoprecipitation of photolabeled Na channel
peptides.Photolabeling of Reconstituted Sodium Channels
Rat
brain Na channels were purified by ion exchange and
hydroxylapatite chromatography and affinity chromatography on wheat
germ agglutinin-Sepharose as described previously(28) .
Solubilized Na channel in detergent solution was
quantified by [H]saxitoxin binding using a rapid
filtration technique (29) . Purified samples containing
250-350 pmol/ml Na channel in
Na
SO
medium were reconstituted into
phosphatidylcholine/phosphatidylethanolamine vesicles (65:35, v/v) as
described by Feller et al.(30) . Purified and
reconstituted Na channels were diluted with 1 volume
of standard binding medium (130 mM choline chloride, 50 mM Hepes, adjusted to pH 7.4 with Tris, 5.5 mM glucose, 0.8
mM MgSO, 5.4 mM KCl; (31) ) and
incubated with 100 nM PbTx-1, 20 µM RU51049, and
[H]BTX-OAB to give a final concentration of
100-125 nM for 4 h in the dark at 25 °C. An aliquot
of this mixture was incubated in the presence of 300 µM veratridine for the determination of nonspecific binding and
nonspecific photolabeling. Dithiothreitol was added to a final
concentration of 0.25 mM (to reduce nonspecific photolabeling)
immediately before ultraviolet irradiation at 4 °C for 30 min using
a germicidal lamp (
= 254 nm; UVP, Inc.) that
was placed 5 cm from the sample. Toxin bound to sodium channels was
then separated from free toxin by rapid gel filtration on 2-ml Sephadex
G-50 spin columns. This photolabeled sample contained specifically
incorporated [H]BTX-OAB at 300-600
cpm/µl, corresponding to labeling of approximately 2% of the total
subunits.Proteolytic Cleavage and Immunoprecipitation of
[
Photolabeled NaH]BTX-OAB-Labeled
Subunits channels purified by
Sephadex chromatography were solubilized with 0.1% Triton X-100 and
treated with TPCK-trypsin at the concentrations shown in the figure
legends. Samples were digested for 1 h at 37 °C and then treated
with 50 µg/ml trypsin inhibitor, 1% Triton X-100, 150 mM NaCl, and 1 mg/ml globulin-free bovine serum albumin for 10 min at
4 °C. Digested and photolabeled Na channel
(approximately 10,000 cpm of [H]BTX-OAB per
sample) was incubated with mixing by rotation with 100 µl of
antibody bound to protein A-Sepharose overnight at 4 °C.
Supernatants were removed, and the pellets were washed three times with
5 volumes of buffer S. The proteins were solubilized from the pellet by
incubation with 8% SDS and analyzed by scintillation spectroscopy.SDS-PAGE and Gel Slicing
Photolabeled samples were
prepared for SDS-PAGE analysis by solubilization with 0.1% Triton X-100
and digestion with 100 µg/ml TPCK-trypsin overnight at 37 °C.
Digested samples were incubated with 50 µg/ml trypsin inhibitor for
10 min at 4 °C and then reduced for 10 min at 25 °C with 10
mM dithiothreitol in the presence of 0.1% (w/v) SDS in 50
mM ammonium bicarbonate buffer adjusted to pH 7.8.
Iodoacetamide (70 mM final concentration) was used to
carboxymethylate the samples by incubation at 25 °C for 1 h.
Further digestion of samples was carried out by incubation with 100
µg/ml V8 protease overnight at 37 °C. Digestion was terminated
by addition of 200 µg/ml aprotinin and 10 µg/ml leupeptin for
10 min at 4 °C. Samples were immunoprecipitated by incubation with
1% Triton X-100, 150 mM NaCl, and protein A-Sepharose-bound
antibody overnight with rotation at 4 °C. Supernatants were
removed, and the pellets were washed two times with 5 volumes of buffer
S. For analysis of [H]BTX-OAB bound to
immunoprecipitated Na channel peptides, a variety of
gel systems were used as described below. Prestained molecular mass
standards were used to determine the molecular mass of the
Na channel peptides. To determine protein-bound
radioactivity, individual gel lanes were manually cut into 1-3-mm
slices, and radioactivity was eluted in 5% (v/v) Soluene in Ecolume
according to the manufacturer's instructions.Electrophoresis Methods
The 7% porous gel system
of Doucet et al. (32) was used to analyze
radiolabeling of the intact Na channel subunit.
Three gel systems were used to analyze small peptides in this study.
System 1 was a Tricine-based gel system described by
Schägger and van Jagow ((33) ; stacking
gel, 4% T, 3% C; spacer gel, 10 T%, 3% C; separating gel, 16.5% T, 6%
C, where T is total monomer and C is cross-linker) poured as a large
gel. System 2 was a commercially prepared 16.5% Tricine mini-gel system
from Novex. System 3 was a 10-20% gradient Tricine mini-gel
system from Novex.
Characterization of [
Previous
work (15) has demonstrated specific, high affinity binding of
[H]BTX-OAB
Binding to Purified and Reconstituted Sodium ChannelsH]BTX to neurotoxin site 2 on sodium channels
purified from rat brain and reconstituted in phospholipid vesicles. In
order to demonstrate that the radiolabeled, photoreactive derivative of
BTX, [H]BTX-OAB (Fig. 1), has the same
binding characteristics as [H]BTX-B, we examined
the binding of [H]BTX-B and
[H]BTX-OAB to the same reconstituted
Na channel preparation. Specific binding of both
[H]BTX-B and [H]BTX-OAB is
inhibited by saturation of receptor site 2 by unlabeled veratridine
with half-maximal inhibition at approximately 8 µM (Fig. 2, solid symbols), in good agreement with
the value of 7 µM observed for synaptosomes(13) .
Using a concentration of 300 µM unlabeled veratridine to
measure nonspecific binding, specific binding was determined to be
92-94% of total binding using either radiolabeled BTX derivative.
No saturable component of [H]BTX-B binding was
observed in parallel experiments with sodium channels that had been
heat-inactivated by incubation at 60 °C for 2 min (Fig. 2, open circles) or with phospholipid vesicles without added
protein (Fig. 2, open squares). The loss of binding at
an elevated temperature provides evidence that conformational integrity
of the purified Na channel is a requirement for high
affinity interaction with BTX, as previously observed for
brevetoxins(11) , saxitoxin, and -scorpion
toxin(24) .
H labels are shown by asterisks. Batrachotoxinin-A (BTX-A) is the
nontoxic natural precursor of batrachotoxin from which both
radiolabeled derivatives (bottom two structures) were
synthesized.
H]BTX-B bound to reconstituted Na channels by unlabeled veratridine. Binding of
[H]BTX-B (17 nM, solid circles)
and [H]BTX-OAB (10 nM, solid
squares) to reconstituted Na channels was
measured as described under ``Experimental Procedures'' in
the presence of the indicated concentrations of veratridine. Control
binding of [H]BTX-B to heat-inactivated and
reconstituted Na channels (open circles) and
reconstituted phospholipids (open squares) was also
determined.
Synergistic Effects of Pyrethroids and Brevetoxins on BTX
Binding
[H]BTX-B binding to purified and
reconstituted rat brain Na channel is enhanced by
pyrethroids(15) . In that study, specific
[H]BTX-B binding was 99% of total binding in the
presence of 10 µM RU39568, the highest level of specific
BTX binding yet demonstrated. In the present study, half-maximal
enhancement of [H]BTX-B binding by RU39568 was
observed at approximately 2 µM (Fig. 3A, solid circles), consistent with previous studies(15) ,
and the half-maximal effect of RU51049 was achieved at about 600 nM (Fig. 3A, solid squares). The maximal
enhancement of BTX binding in the presence of 100 µM RU51049 is approximately 30% greater than that observed with
RU39568. Binding of the pyrethroids, RU51049 and RU39568, and the
brevetoxin, PbTx-1, to their receptor sites in purified Na channel preparations reconstituted into phospholipid vesicles
leads to a marked increase in specific binding of BTX at receptor site
2 (Fig. 3B). PbTx-1 increases the level of specific
binding observed in the presence of a saturating concentration of
either pyrethroid. No specific binding of
[H]BTX-B was observed with purified and
solubilized Na channel preparations that had not been
reconstituted into vesicles, which is consistent with previous
studies(29) .
H]BTX-B binding by RU39568, RU51049, and PbTx-1. A, [H]BTX-B binding to reconstituted
Na channels was measured as described under
``Experimental Procedures'' in the presence of 5 nM [H]BTX-B and increasing concentrations of
RU39568 (solid circles) or RU51049 (solid squares). B, [H]BTX-B binding was measured in the
presence of 0.5 nM [H]BTX-B and 10
µM RU pyrethroid with and without 100 nM PbTx-1
in the presence (stippled bars) or absence (solid
bars) of 300 µM veratridine for determination of
nonspecific and total binding, respectively. Binding of
[H]BTX-B to purified soluble Na channel was measured in the presence of 10 µM RU51049 and 100 nM PbTx-1. Samples were incubated at 25
°C for 17 h for determination of maximum allosteric
interaction.
Specific Photolabeling of Reconstituted Sodium Channels
by [
Purified and reconstituted NaH]BTX-OAB in the Presence of RU51049 and
PbTx-1 channels
were specifically photolabeled by [H]BTX-OAB in
the presence of 20 µM RU51049 and 100 nM PbTx-1.
Analysis of specifically labeled, reconstituted Na channels by SDS-PAGE, gel slicing, and scintillation counting
revealed a single peak of covalently incorporated
[H]BTX-OAB with an apparent molecular mass of 240
kDa, consistent with covalent labeling of the subunit (Fig. 4, solid circles). Samples labeled in the
presence of an excess of unlabeled veratridine did not contain any
[H]BTX-OAB covalently attached to proteins (Fig. 4, open circles). Some degradation of the
specifically labeled protein was observed as a band of radioactivity at
approximately 50 kDa, which is expected since a 4-h incubation at 25
°C was used in this experiment. Consistent with this result,
previous studies have indicated that the Na channel
subunit contains a relatively protease-resistant domain of
50-70 kDa (9) .
channels. Reconstituted Na channels were purified and photolabeled with 90 nM [H]BTX-OAB as described under
``Experimental Procedures'' in the absence (solid
circles) or presence (open circles) of 300 µM veratridine. Samples were analyzed by SDS-PAGE (7% porous reducing
gel system; Doucet et al.(32) ), and the incorporation
of photolabel was determined in 3-mm gel slices by extraction and
scintillation counting as described under ``Experimental
Procedures.'' The migration of molecular mass standards (kDa) are
indicated by arrows.
Isolation and Characterization of Smaller Proteolytic
Fragments Labeled with [
To
localize the position of the covalently attached photolabel, the
reconstituted NaH]BTX-OAB channel labeled with
[H]BTX-OAB was digested at lysine and arginine
residues with TPCK-trypsin. The digested, photolabeled fragments were
then probed by immunoprecipitation with a series of sequence-directed
antibodies recognizing regions of the subunit within or bordering
each of the four homologous domains as illustrated in Fig. 5A. All of the antibodies specifically
immunoprecipitated the intact subunit. The amount of
radioactivity bound by all of the Na channel
antibodies progressively decreased with increasing trypsin
concentrations, suggesting that the antibody binding sites were
separated from the [H]BTX-OAB incorporation site
upon proteolysis (Fig. 5B). After cleavage of the
[H]BTX-OAB-labeled subunit with increasing
concentrations of trypsin, anti-SP1, directed against a sequence
immediately following transmembrane segment IS6, immunoprecipitated
over 70% of the total specifically photolabeled channel at the highest
concentration of trypsin tested (Fig. 5B). The
antibodies within or bordering domains II, III, and IV recognized only
a small fraction of the radioactivity (less than 20%) after extensive
trypsinization. These results indicate that a substantial fraction of
covalently incorporated [H]BTX-OAB is located
within domain I.
H]BTX-OAB. A, recognition
sites of anti-peptide antibodies. Antibodies directed against synthetic
peptides corresponding to different sequences of the subunit of
the type IIA sodium channel were prepared as described (Gordon et
al.(25, 26) ). Antibodies were directed against
synthetic sodium channel peptides corresponding to the amino acid
sequences 355-372 (SP31), 382-400 (SP28),
427-445 (SP1), 468-504 (SP11),
531-547 (SP32), 708-722 (SP15),
1145-1164 (SP20), 1480-1498 (SP14),
1541-1561 (SP19), 1736-1753 (SP29),
1789-1798 (SP13). B, immunoprecipitation of
[H]BTX-OAB-labeled Na channel
peptide fragments from proteolytic cleavage by trypsin. Photolabeled
reconstituted Na channels were digested with
increasing concentrations of TPCK-trypsin, and the resulting peptide
fragments were probed with the indicated antibodies as described under
``Experimental Procedures.'' TPCK-trypsin concentrations for
each antibody treatment were as follows (from left to right): 0.3, 1,
10, 100 µg/ml. Values are expressed as the percentage of total cpm
immunoprecipitated without trypsin
treatment.
channel subunit was more precisely localized
by more extensive proteolytic cleavage at lysine and arginine residues
with trypsin and at glutamic acid and aspartic acid residues with V8
protease from S. aureus. Under the conditions for proteolysis
described in Fig. 6A, approximately 50% of the total
radioactivity was precipitable by anti-SP11 antibody and 70% by
anti-SP1 antibody after cleavage with 10 µg/ml trypsin for 1 h.
These antibodies, directed against peptides within the first 60 amino
acids at the intracellular side of transmembrane segment IS6,
completely lose their ability to recognize subunit-bound
[H]BTX-OAB upon digestion with 100 µg/ml V8
protease for 1 h (Fig. 6B). However, anti-SP31 and
anti-SP28, antibodies directed against peptides within the first 50
amino acids extracellular to transmembrane segment IVS6, were able to
precipitate 30-60% of radioactivity after cleavage with trypsin (Fig. 6A) and over 50% of the total specific
radioactivity after cleavage with V8 protease (Fig. 6B). Anti-SP15, anti-SP14, and anti-SP19, which
are directed against sequences farther toward the carboxyl terminus, do
not immunoprecipitate a significant amount of photolabel (Fig. 6, A and B). These results indicate that
the site of [H]BTX-OAB covalent labeling is
located to the amino-terminal side of the anti-SP1 recognition peptide.
H]BTX-OAB after proteolytic cleavage by
TPCK-trypsin and V8 protease. Photolabeled and reconstituted
Na channel was digested in the presence of 10
µg/ml TPCK-trypsin (A) or 100 µg/ml V8 protease (B) for 2 h at 37 °C, and the resulting peptide fragments
were probed with the indicated antibodies as described under
``Experimental Procedures.'' Values are expressed as the
percentage of total cpm immunoprecipitated by each antibody with no
prior protease treatment.
Photolabeling of a 7-kDa Peptide by
[
A complete tryptic digest was
obtained by covalently labeling the purified and reconstituted
NaH]BTX-OAB channel with [H]BTX-OAB and
subjecting the labeled protein to digestion with 100 µg/ml trypsin
overnight at 37 °C and then with V8 protease under the same
conditions overnight. This extensively digested preparation was used to
restrict further the sites of labeling by
[H]BTX-OAB to small Na channel
peptides recognized by anti-SP28. Based on the results of Fig. 6, we would expect the smallest labeled peptides from such
complete digestions to be recognized by anti-SP28, but not anti-SP1.
Labeled peptide fragments were incubated with either anti-SP1,
anti-SP28, or preimmune rabbit IgG and analyzed by SDS-PAGE in highly
cross-linked gels. Anti-SP1 and anti-SP28 both recognized photolabeled
fragments with apparent molecular masses of 60 and 32 kDa (Fig. 7A, solid symbols); however, IgG from
preimmune serum did not precipitate significant amounts of the
photolabeled channel peptides (Fig. 7A, open
circles). Only anti-SP28 precipitated a photolabeled peptide with
a molecular mass of approximately 6 kDa (Fig. 7A, solid triangles). Some reversibly bound photolabel, which was
not removed by washing prior to electrophoresis, was observed at the
gel dye front. Immunoprecipitation of the 6-kDa peptide by anti-SP28
was specific, because neither preimmune IgG nor anti-SP28 recognize
this small labeled peptide.
H]BTX-OAB-labeled Na channel. A, samples were labeled with [H]BTX-OAB,
digested with 100 µg/ml TPCK-trypsin at 37 °C overnight, then
with 100 µg/ml V8 protease at 37 °C overnight, and analyzed on
a Tricine SDS-PAGE gel (system 2) as described under
``Experimental Procedures.'' The labeled peptides
precipitated by anti-SP28 (solid triangles), anti-SP1 (solid circles), or preimmune rabbit IgG (open
circles) are shown. The migration positions of molecular mass
standards are indicated by the arrows. The gel dye front is
indicated by df. B, estimations of molecular size of
the [H]BTX-OAB-labeled peptide are shown. The
three different gel systems used are described under
``Experimental Procedures.'' The average molecular size
determined in these four experiments is 7.3
kDa.
channels, two different preparations
of [H]BTX-OAB, and three different SDS-PAGE
separation systems with similar results. Although the small
6-7-kDa peptide, which can be precipitated by anti-SP28, migrates
close to the dye front in all of the SDS-PAGE systems used, the
estimation of size is consistent in all three gel systems (Fig. 7B). These results indicate that
[H]BTX-OAB is covalently incorporated into a
peptide of about 7.3 kDa containing the anti-SP28 antibody recognition
site. subunit
near the SP28 peptide (Fig. 8) allows identification of the
photolabeled fragment obtained from extensive cleavage with trypsin and
V8 protease. In domain I, the peptide which includes all of the
anti-SP28 recognition sequence from the trypsin cleavage site, Leu-380,
to the V8 protease cleavage site, Glu-429, has a calculated molecular
mass of 7.3 kDa (Fig. 8). Trypsin cleavage within peptide SP28
was previously found to occur only under extreme
conditions(9) , so cleavage at Arg-395 or Lys-399 is unlikely.
More complete cleavage by V8 protease at Glu-387 within the SP28
sequence might yield the peptide from Asn-388 to Glu-429 with a
calculated molecular mass of 6.0 kDa, which contains all but 6 residues
of the sequence of peptide SP28. Because peptides are linked through
their amino termini to bovine serum albumin for immunization, the first
few amino acid residues of the peptide sequence may not be essential
for antibody recognition of the peptide. Thus, the possible
[H]BTX-OAB-binding peptides containing the SP28
sequence begin 13 or 21 amino acid residues on the extracellular side
of transmembrane segment IVS6, contain the entire transmembrane segment
IVS6, and 3 amino acids on the intracellular side of segment IVS6.
channel subunit. The primary structure and antibody
recognition sequences of the photolabeled peptides from domain I are
shown. Antibody recognition sequences (underlined),
transmembrane segments (boxed), and expected sites of trypsin (arrows) and V8 protease (asterisks) cleavage are
indicated.
subunit to cleavage
by trypsin as observed here are consistent with previous work on
peptide mapping of the -scorpion toxin receptor
site(9, 34) .
High Affinity Binding of a Photoreactive BTX Derivative
to Neurotoxin Receptor Site 2 on Purified and Reconstituted Sodium
Channels
The synthesis of a photoreactive, radiolabeled
derivative of BTX, [H]BTX-OAB, has made possible
the biochemical localization of a component of the BTX receptor site.
In the presence of -scorpion toxin, this ligand binds to the BTX
site of Na channels in rat brain synaptoneurosomes
with a K
of 23 nM(22) . In the
present study, the Na channel activators, RU51049 and
PbTx-1, are used to allosterically enhance
[H]BTX-OAB binding, resulting in a K of less than 1 nM and a specific
binding value of greater than 90%.H]BTX-B and [H]BTX-OAB
bound to purified and reconstituted sodium channels with similar high
affinity. No high affinity binding of [H]BTX-B
was present in heat-inactivated sodium channel preparations or to
phospholipid alone. These results establish that high affinity binding
of BTX derivatives requires the native conformation of the purified
sodium channel and that neurotoxin receptor site 2 is present in active
form on the solubilized, purified, and reconstituted sodium channel.
Covalent labeling of purified and reconstituted Na channel with [H]BTX-OAB provided the
initial material used for biochemical localization of a peptide
sequence that contributes to formation of neurotoxin receptor site 2.Location of the BTX Receptor Site Near Transmembrane
Segment IS6 in Domain I
The photoreactive derivative of BTX,
[H]BTX-OAB, is incorporated specifically into the
rat brain Na channel by photoactivation of the ligand
after equilibrium binding is complete. The specific reaction of this
derivative with a 240-kDa protein implicates the subunit in BTX
binding. Mapping of the peptide segment that is covalently labeled by
[H]BTX-OAB using sequence-directed antibodies
identifies one transmembrane segment as a component of the BTX receptor
site. After proteolytic cleavage, the amount of radioactivity bound by
each sequence-directed antibody progressively decreased as
antibody-binding sequences were separated from the
[H]BTX-OAB incorporation site by proteolysis.
However, even after extensive trypsinization, the anti-peptide antibody
recognizing an amino acid sequence adjacent to transmembrane segment
IS6 was able to immunoprecipitate up to 70% of the labeled peptides. In
contrast, antibodies recognizing sequences within or adjacent to
domains II, III, and IV do not immunoprecipitate labeled peptides.
These results implicate a segment in domain I in formation of the BTX
receptor site.H]BTX-OAB was achieved by analysis of
immunoprecipitated fragments with antibodies recognizing peptide
sequences within or near domain I after more extensive proteolytic
cleavage with TPCK-trypsin and V8 protease. After trypsinization, two
anti-peptide antibodies recognizing the amino acid sequence on the
intracellular side of transmembrane segment IS6 were able to
immunoprecipitate at least 50% of the
[H]BTX-OAB-labeled peptides, and two anti-peptide
antibodies recognizing the amino acid sequence on the extracellular
side of this same transmembrane region were able to immunoprecipitate
30-60% of the labeled peptides. In contrast, after treatment with
V8 protease, the ability of the two antibodies recognizing the
intracellular amino acid sequences to precipitate the labeled peptide
was completely lost, whereas the antibodies recognizing the sequences
extracellular to IS6 retained their ability to precipitate the labeled
peptide. These results indicate that the BTX receptor site is formed by
a portion of domain I and that the site of photolabeling does not
include the amino acids to the carboxyl-terminal side of transmembrane
region IS6.H]BTX-OAB was achieved by
SDS-PAGE analysis of cleavage products from proteolytic digestions of
labeled Na channels with both TPCK-trypsin and V8
protease. The identification of a specifically immunoprecipitated 7-kDa
peptide from domain I restricts the labeled peptide fragment to
residues Asn-388 to Glu-429 (calculated molecular mass, 6.0 kDa, Fig. 8) or to the region from Leu-380 to Glu-429 (7.3 kDa) if V8
protease cleavage is incomplete. When estimations of molecular mass
from three different gel systems are averaged together, the value for
the radiolabeled, immunoprecipitated peptide is 7.3 kDa, in excellent
agreement with the size of the predicted cleavage product from Leu-380
to Glu-429.H]BTX-OAB is positioned on an aromatic side
chain known to be important in conferring toxicity to BTX because the
precursor BTX-A, which lacks this aromatic moiety, is
nontoxic(3) . Therefore, the photoreactive side chain most
likely interacts with an integral part of the receptor site for BTX on
the Na channel subunit. Thus, our results
implicate transmembrane segment IS6 in direct interaction with bound
BTX in receptor site 2.Analysis of Covalent Labeling by Antibody Mapping of
Labeled Peptides
Photoaffinity labeling of ion channels and
other minor membrane proteins is an effective method for identification
of their functional components. However, the low level of incorporation
of most photolabels and the low abundance and hydrophobicity of the
labeled proteins often prevent high resolution analysis of the labeled
products by complete purification of the labeled peptides and amino
acid sequence determination. Identification of labeled peptides by
mapping with sequence-directed anti-peptide antibodies has proven to be
a powerful and reliable method of peptide identification. This approach
has been used to identify covalently labeled receptor sites for
dihydropyridines (35, 36) and phenylalkylamines (37) on L-type Ca
channels and for
-scorpion toxin(34) , brevetoxins (11) , and BTX
(this paper) on Na channels. Subsequent analysis by
site-directed mutagenesis has identified individual amino acid residues
within or near the site of photoaffinity labeling that are required for
binding of dihydropyridines (38) and phenylalkylamines (39) to L-type Ca channels and -scorpion
toxin to Na channels(12) . The photoactive
derivatives of small organic ligands have closely mapped the receptor
site residues (35, 36, 37, 38, 39) ;
however, the site of -scorpion toxin binding has only been
approximated by photolabeling due to the relatively large size of the
ligand(12) . Mutagenesis studies have confirmed the accuracy of
the photoaffinity labeling and antibody mapping approaches and provided
a higher resolution view of the components of these receptor sites.
Analysis of the region of transmembrane segment IS6 by site-directed
mutagenesis should yield additional insight into the molecular
requirements for high affinity binding of BTX.Allosteric Interactions with the BTX Receptor
Site
A model for allosteric interactions with the BTX receptor
site can be proposed by taking into consideration the lipophilic nature
of the toxin, its physiological effects, and the location of the
photolabeled peptide. BTX can gain access to its binding site from
either side of the membrane, since BTX-induced membrane depolarization
is observed in squid giant axon when the toxin is applied either
internally or externally to the perfusion medium(40) . The BTX
molecule is approximately 12 Å long, much smaller than the width
of the phospholipid bilayer. Therefore, the BTX-binding site may be
located within a transmembrane region of the subunit, in a
relatively inaccessible position in the resting Na channel conformation. Activation of the Na channel by repetitive depolarization dramatically accelerates BTX
binding(41) . In addition, several chemical activators,
including the pyrethroids, RU39568 and RU51049, and the brevetoxin,
PbTx-1, induce a conformational change which results in the greater
accessibility and higher affinity of BTX to its receptor site. The site
of brevetoxin binding has been localized to a transmembrane region at
the interface of domains I and IV that includes transmembrane segment
IS6(11) , to which BTX also binds. Thus, it is possible that
multiple sites for hydrophobic toxin binding include determinants
within this transmembrane segment and allow discrete changes in
three-dimensional structure of the Na channel caused
by binding of one toxin to affect the binding of others through
allosteric interactions. channel.
Site-directed mutagenesis studies have identified the amino acid
residues lining the mouth of the channel pore between transmembrane
segments 5 and 6 in each of the four domains as critical for binding of
tetrodotoxin and saxitoxin at neurotoxin receptor site
1(6, 7, 8) . The binding of these toxins to
site 1 has been shown to allosterically decrease the binding of BTX to
site 2 under some conditions(16) . channel that exhibits
considerable structural flexibility. Not only does the binding of other
toxins affect the affinity of BTX binding through allosteric alteration
of site 2, but binding of BTX itself results in a conformational change
of the channel. Upon BTX binding to the Na channel,
the flux of relatively impermeant cations increases, indicating that
the size of the ion selectivity filter is increased, inactivation is
blocked, and the voltage dependence of activation is shifted to more
negative potentials (see review by Brown et al.(42) ).
These alterations in channel function indicate that the Na channel has multiple neurotoxin binding sites that are located in
areas of great conformational flexibility. Upon binding of neurotoxins
to the Na channel, conformational changes occur that
alter the position of toxin binding sites with respect to one another
and change channel gating kinetics. By locating the precise positions
of toxin receptor sites and by understanding the relationship of these
sites to one another, the alterations of normal gating processes
following toxin binding and mechanisms of allosteric interaction among
bound toxins will be clarified.
)H]BTX-B, tritiated batrachotoxinin-A
20--benzoate, [benzoyl-2,5-];
[H]BTX-OAB, tritiated batrachotoxinin-A
orthoazidobenzoate, [benzoyl-2,5-]; PbTx, brevetoxin from Ptychodiscus brevis; TPCK-trypsin, L-1-tosylamido-2-phenylethyl chloromethyl ketone-treated
trypsin; SP, sodium channel peptide; PAGE, polyacrylamide gel
electrophoresis; Tricine, N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine; V8
protease, protease type XVII-B (from Staphylococcus aureus strain V8).
We thank Dr. John Daly for his generous gift of
batrachotoxinin-A used in this study and Chiral Corp. for the gift of
PbTx-1. We are grateful to Roussel Uclaf for supplying us with the
pyrethroids RU39568 and RU51049. We acknowledge Carl Baker for expert
technical assistance with purification of sodium channels.
©1996 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:
|
|
 |
|
  K. Kajiwara, T. Yanagita, Y. Nakashima, A. Wada, F. Izumi, and N. Yanagihara Differential Effects of Short and Prolonged Exposure to Carvedilol on Voltage-Dependent Na+ Channels in Cultured Bovine Adrenal Medullary Cells J. Pharmacol. Exp. Ther., July 1, 2002; 302(1): 212 - 218. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-L. Li, D. Hadid, and D. S. Ragsdale The Batrachotoxin Receptor on the Voltage-Gated Sodium Channel is Guarded by the Channel Activation Gate Mol. Pharmacol., April 1, 2002; 61(4): 905 - 912. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Shiraishi, T. Yanagita, H. Kobayashi, Y. Uezono, H. Yokoo, S.-I. Minami, M. Takasaki, and A. Wada Up-Regulation of Cell Surface Sodium Channels by Cyclosporin A, FK506, and Rapamycin in Adrenal Chromaffin Cells J. Pharmacol. Exp. Ther., April 12, 2001; 297(2): 657 - 665. [Abstract] [Full Text]
|
|
|
|
|
|
C. Nau, S.-Y. Wang, G. R. Strichartz, and G. K. Wang Point Mutations at N434 in D1-S6 of {micro}1 Na+ Channels Modulate Binding Affinity and Stereoselectivity of Local Anesthetic Enantiomers Mol. Pharmacol., August 1, 1999; 56(2): 404 - 413. [Abstract] [Full Text]
|
|
|
|
 |
|
 Y. Anis, B. Nurnberg, L. Visochek, N. Reiss, Z. Naor, and M. Cohen-Armon Activation of Go-proteins by Membrane Depolarization Traced by in Situ Photoaffinity Labeling of Galpha o-proteins with [alpha 32P]GTP-azidoanilide J. Biol. Chem., March 12, 1999; 274(11): 7431 - 7440. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Lee, Y. Park, T. M. Brown, and M. E. Adams Altered Properties of Neuronal Sodium Channels Associated with Genetic Resistance to Pyrethroids Mol. Pharmacol., March 1, 1999; 55(3): 584 - 593. [Abstract] [Full Text]
|
|
|
|
 |
|
 N. J. Linford, A. R. Cantrell, Y. Qu, T. Scheuer, and W. A. Catterall Interaction of batrachotoxin with the local anesthetic receptor site in transmembrane segment IVS6 of the voltage-gated sodium channel PNAS, November 10, 1998; 95(23): 13947 - 13952. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. K. Wang, C. Quan, and S.-Y. Wang Local Anesthetic Block of Batrachotoxin-Resistant Muscle Na+ Channels Mol. Pharmacol., August 1, 1998; 54(2): 389 - 396. [Abstract] [Full Text]
|
|
|
|
|
|
S.-Y. Wang and G. K. Wang Point mutations in segment I-S6 render voltage-gated Na+ channels resistant to batrachotoxin PNAS, March 3, 1998; 95(5): 2653 - 2658. [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 |