Content uploaded by Zvi Vogel
Author content
All content in this area was uploaded by Zvi Vogel
Content may be subject to copyright.
Proc.
Nati.
Acad.
Sci.
USA
Vol.
89,
pp.
7717-7721,
August
1992
Neurobiology
How
the
mongoose
can
fight
the
snake:
The
binding
site
of
the
mongoose
acetylcholine
receptor
(ligand
binding/a-bungarotoxin/polymerase
chain
reaction/immunofuorcscence
microscopy)
DoRA
BARCHAN*,
SYLVIA
KACHALSKY*,
DRORIT
NEUMANN*,
ZVI
VOGELt,
MICHAEL
OVADIAt,
ELAZAR
KOCHVAt,
AND
SARA
FUCHS*
Departments
of
*Chemical
Immunology
and
tNeurobiology,
The
Weizmann
Institute
of
Science,
Rehovot
76100,
Israel;
and
tDepartment
of
Zoology,
Tel
Aviv
University,
Tel
Aviv
69978,
Israel
Communicated
by
Michael
Sela,
April
3,
1992
(received
for
review
February
9,
1992)
ABSTRACT
The
ligand
binding
site
of
the
nicotinic
ace-
tylcholine
receptor
(AcChoR)
is
within
a
short
peptide
from
the
a
subunit
that
includes
the
tandem
cysteine
residues
at
posi-
tions
192
and
193.
To
elucidate
the
molecular
basis
of
the
binding
properties
of
the
AcChoR,
we
chose
to
study
nonclas-
sical
muscle
AcChoRs
from
animals
that
are
resistant
to
a-neurotoxins.
We
have
previously
reported
that
the
resistance
of
snake
AcChoR
to
a-bungarotoxin
(a-BTX)
may
be
ac-
counted
for
by
several
major
substitutions
in
the
ligand
binding
site
of
the
receptor.
In
the
present
study,
we
have
analyzed
the
binding
site
of
AcChoR
from
the
mongoose,
which
is
also
resistant
to
a-neurotoxins.
It
was
shown
that
mongoose
Ac-
ChoR
does
not
bind
a-BTX
in
vivo
or
in
vitro.
cDNA
fragments
of
the
a
subunit
of
mongoose
AcChoR
corresponding
to
codons
122-205
and
including
the
presumed
ligand
binding
site
were
cloned,
sequenced,
and
expressed
in
Escherichia
coi.
The
expressed
protein
agments
of
the
mongoose,
as
well
as
of
snake
receptors,
do
not
bind
a-BTX.
The
mongoose
fragment
is
highly
homologous
(>90%)
to
the
respective
mouse
frag-
ment.
Out
of
the
seven
amino
acid
differences
between
the
mongoose
and
mouse
in
this
region,
five
cluster
in
the
presumed
ligand
big
site,
close
to
cysteines
192
and
193.
These
changes
are
at
positions
187
(Trp
-
Asn),
189
(Phe
Thr),
191
(Ser
-
Ala),
194
(Pro
-+
Leu),
and
197
(Pro
--
His).
The
mongoose
like
the
snake
AcChoR
has
a
potential
glycosylation
site
in
the
binding
site
domain.
Sequence
comparison
between
species
suggests
that
substitutions
at
positions
187,
189,
and
194
are
important
in
determining
the
resistance
of
mongoose
and
snake
AcChoR
to
a-BTX.
In
addition,
it
was
shown
that
amino
acid
residues
that
had
been
reported
to
be
necessary
for
acetylcholine
binding
are
conserved
in
the
toxin-resistant
an-
imals
as
well.
The
nicotinic
acetylcholine
receptor
(AcChoR)
is
an
integral
membrane
glycoprotein
composed
of
four
types
of
subunits
present
in
a
stoichiometry
of
a213^y
(for
review,
see
refs.
1
and
2).
The
cholinergic
binding
site
of
the
receptor
is
within
the
a
subunit
(1,
3,
4)
in
close
proximity
to
a
sulfhydryl
group
(1).
A
number
of
experimental
approaches
have
been
em-
ployed
to
identify
the
ligand
binding
site
in
AcChoR
and
the
amino
acids
that
participate
in
it.
Studies
based
on
proteolytic
fragmentation
of
the
a
subunit
(5-7),
affinity-labeling
exper-
iments
(8,
9),
synthetic
peptides
(6,
7,
10),
genetic
constructs
(11,
12),
and
site-directed
mutagenesis
(13)
indicated
that
the
ligand
binding
site
of
AcChoR
is
within
a
region
of
the
a
subunit
that
contains
the
two
tandem
cysteine
residues
at
positions
192
and
193.
We
demonstrated
that
a
synthetic
dodecapeptide
corresponding
to
amino
acid
residues
185-1%
of
the
Torpedo
AcChoR
a
subunit
contains
the
essential
elements
of
the
ligand
binding
site
(7,
10).
To
analyze
the
detailed
structure
of
the
cholinergic
binding
site
of
AcChoR
and
to
elucidate
the
structural
requirements
for
agonist
vs.
a-bungarotoxin
(a-BTX)
binding,
we
studied
nonconventional
muscle
AcChoRs
of
animals
that
are
resist-
ant
to
a-BTX.
AcChoR
of
elapid
snakes
is
unique
in
its
pharmacological
properties;
it
binds
cholinergic
ligands
but,
unlike
other
muscle
AcChoRs,
it
does
not
bind
a-BTX
(14).
We
have
cloned
and
sequenced
(15)
cDNA
fragments
that
contain
the
presumed
ligand
binding
site
in
the
AcChoR
a
subunit
from
two
different
snakes.
We
demonstrated
that
in
the
binding
site
region,
in
the
vicinity
of
cysteines
192
and
193,
several
major
substitutions
occur
in
the
snake
sequence
at
positions
184
(Trp
-+
Phe),
185
(Lys
--
Trp),
187
(Trp
--
Ser),
and
194
(Pro
-+
Leu).
In
addition,
Asn-189
is
a
putative
N-glycosylation
site,
present
only
in
the
snake
(15).
These
changes
or
part
of
them
may
explain
the
lack
of
a-BTX
binding
to
snake
AcChoR.
Some
of
the
sequence
differences
observed
in
the
snake
AcChoR
could
be
specific
to
the
group.
Therefore,
we
have
extended
our
study
to
a
mammal
(mongoose)
that
is
resistant
to
neurotoxins
(16)
and
includes
snakes
in
its
diet.
We
have
cloned
and
sequenced
the
region
of
the
AcChoR
a
subunit
(residues
122-205)
that
contains
the
binding
site'domain
from
the
mongoose
and
from
an
additional
primitive
snake,
the
sand
boa
(Eryx
jaculus).§
This
region
in
the
mongoose
AcChoR
is
highly
homologous
(>90%)
to
the
corresponding
region
in
other
mammalian
AcChoRs.
Nevertheless,
there
are
five
amino
acid
differences
in
the
mongoose
sequence
that
cluster
in
a
very
limited
segment
in
the
presumed
binding
site
area.
Sequence
comparison
of
the
binding
site
domains
of
the
mongoose
and
snake
AcChoR
with
those
of
other
AcChoRs
led
us
to
propose
that
substitutions
at
positions
187, 189,
and
194
of
the
receptor
a
subunit
are
important
in
conferring
toxin
resistance
in
these
animals.
MATERIALS
AND
METHODS
Animals.
Mongooses
(Herpestes
ichneumon)
and
snakes
(the
sand
boa
Eryxjaculus
and
the
cobra
Naja
naja
atra)
were
obtained
from
The
Canadian
Center
of
Ecological
Zoology
(Tel-Aviv
University).
Mice
and
rabbits
were
from
the
Center
of
Animal
Breeding
(The
Weizmann
Institute).
RNA
Preparation
and
Northern
Blot
Analysis.
RNA
prep-
aration
and
Northern
blot
analysis
were
performed
as
de-
scribed
(17).
Abbreviations:
AcCho,
acetylcholine;
AcChoE,
acetylcholinester-
ase;
AcChoR,
acetylcholine
receptor;
a-BTX,
a-bungarotoxin;
FITC,
fluorescein
isothiocyanate;
TMR,
tetramethylrhodamine.
§The
sequence
reported
in
this
paper
has
been
deposited
in
the
GenBank
data
base
(accession
no.
M93639).
7717
The
publication
costs
of
this
article
were
defrayed
in
part
by
page
charge
payment.
This
article
must
therefore
be
hereby
marked
"advertisement"
in
accordance
with
18
U.S.C.
§1734
solely
to
indicate
this
fact.
7718
Neurobiology:
Barchan
et
al.
Preparation
and
Amplification
of
cDNA.
The
preparation
of
cDNA
and
the
polymerase
chain
reaction
(PCR)
were
per-
formed
as
described
(15).
Fragments
were
purified
and
sub-
cloned
into
M13
bacteriophage
vectors
mp18
and
mp19
or
pBluescript
KS-
for
sequencing
and
then
into
the
pET8C
vector
for
expression.
The
primer
at
the
5'
end
(GGCCATG-
GCCATCTTRAAAAGC,
where
R
=
C
or
T)
corresponded
to
a
highly
conserved
region
of
the
a
subunit
(amino
acid
residues
122-126)
and
was
designed
in
a
way
that
enabled
cloning
into
a
pET8C-derived
expression
vector
by
adding
a
restriction
site
for
Nco
I
(underlined)
and
an
initiation
codon
(marked
by
asterisks).
The
primer
at
the
3'
end
(CCGGAT
CCTCAAAAGTGRTAGGTGATRTC,
where
R
=
A
or
G)
corresponded
to
the
complementary
sequence
of
another
conserved
region
(amino
acid
residues
200-205),
and
con-
tained
a
restriction
site
for
BamHI
(underlined)
and
a
stop
codon
(marked
by
asterisks).
Expression
and
Analyses
of
Cloned
cDNA
Fragments.
The
cloned
cDNA
fragments
of
the
mongoose,
snake,
and
mouse
were
subcloned
into
Nco
I
and
BamHI
sites
of
the
expression
vector
pET8C
(18).
Cloning
sites
were
confirmed
by
DNA
sequencing,
and
induction
of
protein
expression
was
per-
formed
(18).
After
expression,
the
Escherichia
coli
suspen-
sion
(400
ml)
was
centrifuged,
cells
were
lysed
by
freezing
and
thawing
the
pellet
and
resuspended
in
phosphate-
buffered
saline
(PBS,
20
ml).
The
resuspended
material
was
sonicated
for
five
15-sec
periods
and
kept
frozen
in
aliquots
until
use.
After
centrifugation,
the
expressed
protein
was
localized
in
the
precipitate,
probably
in
inclusion
bodies.
The
proteins
were
analyzed
by
electrophoresis
in
SDS/
polyacrylamide
gel
(15%),
followed
by
blotting
and
toxin
or
antibody
overlays
as
described
(7).
Preparation
of
Antibodies.
Antibodies
to
proteins
ex-
pressed
by
the
cloned
cDNA
fragments
were
elicited
in
rabbits
by
three
immunizations
with
the
homogenized
gel
band
containing
the
8-kDa
protein
fragment
and
originating
from
0.5
ml
of
concentrated
(20
times)
cell
suspension
emulsified
in
complete
Freund's
adjuvant.
Immunofluorescence
Microscopy.
Diaphragms
were
dis-
sected
from
Wistar
rats
and
mongooses.
Areas
containing
endplates
were
quickly
frozen
and
20-,um
sections
were
incubated
for
1
hr
with
purified
anti-AcChoR
antibodies
in
PBS
containing
0.25%
gelatin
and
0.5%
bovine
serum
albu-
min
(PBS-GB).
Sections
were
washed
and
incubated
for
1
hr
with
a
mixture
containing
50
nM
tetramethylrhodamine-
conjugated
a-BTX
(TMR-a-BTX;
ref.
19)
and
fluorescein
isothiocyanate
(FITC)-conjugated
goat
anti-rabbit
IgG
(Hy-
land,
Costa
Mesa,
CA)
at
10
,g/ml.
Slides
were
transferred
to
70%
ethanol
at
-20°C
and
mounted
in
glycerol/PBS
Citifluor
(Citifluor,
London).
Fluorescence
photomicro-
graphs
were
taken
at
exposures
of
15-30
sec
on
Kodak
T-Max
400
ASA
film
processed
to
ASA
800.
Acetylcholinesterase
(AcChoE)
was
stained
by
the
method
of
Karnovsky
and
Roots
(20)
or
by
the
immunofluorescent
technique
using
rabbit
anti-Torpedo
AcChoE
serum
(80b;
FIG.
1.
Staining
of
AcChoE
in
mongoose
diaphragm
sections.
(A)
Karnovsky
staining
(20).
(B
and
C)
Immunofluorescent
staining.
(B)
FITC
fluorescence
showing
staining
of
AcChoE
in
the
endplate.
(C)
TMR
fluorescence
of
the
same
field
showing
that
TMR-a-BTX
did
not
stain
the
endplate.
(Bar
=
50
gm.)
kindly
provided
by
Palmer
Taylor,
University
of
California,
San
Diego;
ref.
21),
followed
by
FITC-conjugated
anti-rabbit
antibody,
as
described
above.
RESULTS
Resistance
of
Mongoose
to
a-BTX.
The
toxic
effect
of
a-BTX
in
the
mongoose
as
compared
with
rabbit
and
mouse
was
examined.
Intramuscular
administration
of
a-BTX
into
mongoose
in
amounts
of
0.3-2
,ug/g
of
body
weight
did
not
kill
the
mongoose,
whereas
0.1
and
0.3
,lg/g
of
body
weight
were
lethal
in
mice
and
rabbits,
respectively.
The
resistance
of
mongoose
to
a-BTX
was
not
due
to
neutralizing
factors
in
their
blood
serum,
since
preincubation
of
mongoose
serum
with
a-BTX
did
not
abolish
its
toxic
effect
upon
subsequent
injection
into
mice,
as
has
been
observed
also
for
snake
serum
(15).
Mongoose
AcChoR
Does
Not
Bind
ar-BTX.
Extracts
of
either
mongoose
or
snake
(cobra
or
sand
boa)
muscle
did
not
bind
1251-labeled
a-BTX,
whereas
extracts
of
mouse
muscle
ri
^ _
r~~B
"C
FIG.
2.
Staining
of
AcChoR
in
mongoose
(A-F)
and
rat
(G-L)
diaphragm
sections.
Sections
were
incubated
with
rabbit
anti-
denatured
Torpedo
AcChoR
antibody
(A-C
and
G-1)
or
rabbit
anti-peptide
351-368
(D-F
and
J-L)
and
then
with
a
mixture
of
TMR-a-BTX
and
FITC-conjugated
goat
anti-rabbit
IgG.
(A,
D,
G,
and
J)
Phase-contrast
images.
(B,
E,
H,
and
K)
FITC
fluorescence
demonstrating
AcChoR
staining
with
both
types
of
antibodies.
(C,
F,
I,
and
L)
TMR
fluorescence
of
the
same
fields
demonstrating
that
TMR-a-BTX
did
not
stain
the
mongoose
endplate
(A-F)
and
stained
the
rat
endplate
(G-L).
(Bar
=
50
gm.)
Proc.
Natl.
Acad.
Sci.
USA
89
(1992)
Proc.
Natl.
Acad.
Sci.
USA
89
(1992)
7719
b
28S-
18s-
1
2
1
2
FIG.
3.
Northern
blot
analysis
of
mongoose
RNA.
Poly(A)+
RNA
from
mongoose
(lane
1)
or
mouse
(lane
2)
was
gel-
electrophoresed,
blotted,
and
probed
with
mouse
AcChoR
a-subunit
cDNA
(a)
or
with
a
mongoose-specific
oligonucleotide
(33-mer)
corresponding
to
amino
acid
residues
187-197
(b)
(see
Fig.
4).
exhibited
a
specific
binding
that
could
be
displaced
by
unlabeled
a-BTX
or
by
d-tubocurarine
(data
not
shown).
As
shown
in
Fig.
1,
the
endplates
in
mongoose
diaphragm
could
be
easily
visualized
by
the
Karnovsky
method
(20),
which
is
based
on
the
activity
of
the
enzyme,
or
by
fluores-
cent
microscopy
employing
anti-AcChoE
antibodies.
Rho-
damine-conjugated
a-BTX
did
not
stain
the
mongoose
end-
plates.
AcChoR
in
the
mongoose
endplate
was
visualized
by
rabbit
antibody
against
denatured
Torpedo
AcChoR
(22)
or
against
a
synthetic
peptide
corresponding
to
residues
351-368
of
the
a
subunit
of
human
AcChoR
(23).
This
latter
antibody
is
specific
for
mammalian
AcChoR
(23).
Sections
from
mon-
goose
and
for
comparison,
from
rat
diaphragm,
were
incu-
bated
with
either
of
the
two
antibodies,
followed
by
FITC-
labeled
goat
anti-rabbit
immunoglobulins.
The
same
sections
were
also
incubated
with
TMR-conjugated
a-BTX.
As
can
be
seen
in
Fig.
2
(A-F),
both
antibodies
stained
the
mongoose
endplate
whereas
no
staining
was
observed
with
a-BTX.
In
contrast,
both
the
anti-AcChoR
antibodies
and
a-BTX
stained
the
endplate
regions
in
sections
of
the
rat
diaphragm
(Fig.
2
G-L).
Cloning
the
Binding
Site
Domain
of
Mongoose
AcChoR.
We
have
cloned
and
sequenced
a
mongoose
cDNA
fragment
that
includes
the
binding
site
domain
(i.e.,
the
segment
from
the
a
subunit
containing
the
tandem
cysteines
192
and
193).
We
first
verified
by
Northern
blot
analysis
that
mongoose
poly(A)-containing
RNA
hybridizes
with
the
mouse
AcChoR
a-subunit
cDNA
probe.
As
shown
in
Fig.
3a,
a
4-kilobase
transcript
hybridized
specifically
with
the
mouse
probe.
The
mongoose
transcript
for
the
a
subunit
is
larger
than
the
homologous
mouse
transcript
(2.3
kilobases).
The
PCR
was
used
to
amplify
the
cDNA
fragment
encom-
passing
the
binding
site
region,
from
mongoose
single-
stranded
cDNA.
The
resulting
250-base-pair
fragment
hy-
bridized
to
the
mouse
AcChoR
a-subunit
cDNA.
Sequence
analysis
of
this
amplified
mongoose
fragment
(Fig.
4)
re-
vealed
high
homology
with
the
respective
mouse
fragment,
corresponding
to
amino
acid
residues
122-205
of
the
a
subunit
(homology
of
89%
in
nucleotides
and
92%
in
amino
acids).
The
mongoose
segment
contains
the
four
cysteines
at
positions
128,
142,
192,
and
193,
thus
verifying
that
it
corresponds
to
the
AcChoR
a
subunit.
Interestingly,
5
of
the
7
amino
acid
differences
between
the
mouse
and
the
mon-
goose
fragments
concentrate
in
the
vicinity
of
the
tandem
cysteines
in
a
stretch
of
11
amino
acid
residues
(residues
187-197).
Three
of
these
5
differences
are
at
positions
187,
189,
and
194
where
major
substitutions
take
place
also
in
the
snake
AcChoR
(15),
and
one
of
them
at
position
187
creates
a
potential
N-glycosylation
site
in
the
mongoose
AcChoR.
A
synthetic
oligonucleotide
corresponding
to
amino
acids
187-
197
of
the
mongoose
sequence
hybridized
to
the
mongoose
and
not
to
the
mouse
poly(A)-containing
RNA
(Fig.
3b).
Cloning
the
Binding
Site
Domain
of
the
Sand
Boa
AcChoR.
Binding
experiments
with
muscle
Triton
extracts
from
the
sand
boa
did
not
reveal
any
significant
binding
to
1251-labeled
a-BTX
(data
not
shown).
We
have
then
PCR-amplified
the
250-base-pair
fragment
from
single-stranded
cDNA
from
the
sand
boa,
by
employing
the
primers
used
to
clone
the
mongoose
fragment.
Sequence
analysis
of
the
boa
fragment
revealed
a
very
high
similarity
to
the
cobra
and
water
snake
respective
fragments
(15).
In
this
fragment
there
is
only
one
amino
acid
difference
between
the
boa
and
cobra
(residue
170
is
histidine
in
the
boa
and
tyrosine
in
cobra
and
water
snake)
and
another
difference
between
the
boa
and
water
snake
(residue
149
is
tryptophan
in
the
boa
and
arginine
in
the
water
snake).
All
three
snakes
are
completely
identical
in
the
putative
binding
site
area.
Expression
and
Binding
Properties
of
the
Mongoose
and
Snake
Fragments.
The
cloned
fragments
corresponding
to
amino
acid
residues
122-205
of
the
mongoose,
cobra,
and,
for
comparison,
the
mouse
AcChoR
were
expressed
employing
a
pET8C-derived
expression
vector.
The
expressed
protein
fragments
were
localized
in
the
insoluble
pellet,
probably
in
inclusion
bodies.
These
expressed
fragments
have
the
ex-
pected
molecular
mass
of
8
kDa
in
SDS/polyacrylamide
gel
and
constitute
the
major
protein
in
the
pellet
(Fig.
5a).
Antibodies
against
a
synthetic
peptide
corresponding
to
residues
143-158
of
the
Torpedo
AcChoR
a
subunit
stained
all
three
fragments
(Fig.
5b),
indicating
that
the
expressed
fragments
are
indeed
from
the
AcChoR
a
subunit.
Overlay
of
the
blotted
proteins
with
1251-labeled
a-BTX
showed
that
the
toxin
binds
only
to
the
mouse
fragment
and
not
to
snake
or
mongoose
fragments
(Fig.
5c).
In
some
cases
a
very
long
exposure
of
the
blots
gave
a
faint
signal
with
the
mongoose
fragment.
122
*
140
*
A
I
F
K
S
Y
C
E
I
I
v
T
H
F P
FD
E
Q
N
C
Mongoose
GCC
ATC TTC
AAA
AGC
TAC
TGT
GAG
ATC
ATC
GTC
ACC
CAC
TTT
CCC
TTT
GAT
GAA CAG
AAC
TGC
Mouse
...
...
..
T.
...
...
...
... ...
..
T
..
T
...
...
..
C
..
...
...
S
M
K
L
G
T
W
TY
D
S
Mongoose
AGC
ATG
AAG
CTG
GGT
ACC
TGG
ACC
TAT
GAC
AGC
Mouse
...
...
...
...
..
C
...
...
...
... ...
G..
G
160
S
V
V
V
I
N
P E S
D
TCT GTG
GTT
GTC
ATC
AAC
CCG
GAA
AGC GAC
...
...
..G
C.
..T
...
...
...
..T
...
A
180
Q
P
D
L
S
N
F
M
E
S
G
E
W V
I
K
E
A
R
G
W
Mongoose
CAA
CCT
GAC
CTA
AGC
AAC
TTC
ATG
GAA
AGC
GGA
GAG
TGG
GTG
ATC
AAG
GAG
GCC
CGG
GGC
TGG
Mouse
..G
..C
...
..G
..T
...
...
...
..G
...
..G
...
...
...
...
...
..A
..T
...
...
...
*
*
200
K
H
N
V
T
Y
A
C
C
L
T
T
H
Y
LD
Mongoose
AAG
CAC
AAT
GTG
ACC
TAC
GCC
TGC
TGC
CTC
ACC
ACC
CAC
TAC
CTG
GAC
Mouse
.
TOO
...
TT.
...C
.T
.0.
W
F
S
. .
P
.
P
I
T
Y
H
F
ATC
ACC
TAC
CAC
TTC
...
...
FIG.
4.
Alignment
of
nucleotide
and
deduced
amino
acid
sequences
for
the
mongoose
PCR
fragment
of
the
AcChoR
a
subunit
and
the
corresponding
mouse
fragment.
Amino
acid
residues
are
numbered
from
122
to
205,
corresponding
to
their
position
in
the
mouse
AcChoR
a
subunit.
Cysteine
residues
are
marked
with
an
asterisk.
Nucleotides
or
amino
acids
identical
to
the
mongoose
sequence
are
designated
by
dots.
Neurobiology:
Barchan
et
al.
7720
Neurobiology:
Barchan
et
at.
94-
U-'
67
-
43x-
I-a
Mongoose
Human
Calf
Mouse
Chick
Torpedo
Snake
20P
-
144-
180
200
**
EARGWKHNVTYACCLTTHYLD
..
.S
..
S,.PD.P...
.S
.
.
.W.F...PS.P.
.
.
W.F.S
.P.
.P
...
DY
....
W.Y
.
PD.P...
DY
....
W.Y.T.
.PD.P...
DY
..FW.S.N.S..
D.P...
Rat(PC12)
DAV.TYNTRK.E..AEI-.P.
_w
4PP
-
i,
2
3
5
FiG.
5.
Analysis
of
expressed
protein
fragments.
The
expressed
protein
fragments
(20
pg
of
protein)
or
purified
Torpedo
AcChoR
(10
pg)
were
resolved
by
polyacrylamide
gel
electrophoresis
(15%
gel).
The
gels
were
stained
for
proteins
by
Coomassie
brilliant
blue
(a)
or
blotted
(b
and
c)
and
overlaid
with
rabbit
anti-peptide
143-156,
followed
by
M2I-labeled
protein
A
(b)
or
with
125I-labeled
a-BTX
(c).
Lanes:
1,
pET8C
proteins
without
inserted
DNA;
2-4,
expressed
mouse,
mongoose,
and
cobra
fragment,
respectively;
5,
Torpedo
AcChoR.
Antiserum
elicited
against
the
expressed
mongoose
fag-
ment
stained
specifically
the
endplates
of
both
mongoose
and
rat
diaphragms
in
immunofluorescence
microscopy
(Fig.
6).
To
allow
for
efficient
staining
of
the
receptor
with
the
anti-fiagment
antibodies,
the
sections
had
to
be
pretreated
with
0.2%
SDS
in
PBS
for
1
hr
to
partially
denature
the
extracellular
portion
of
the
receptor.
Similar
results
were
obtained
with
the
antibodies
against
the
expressed
mouse
fragment.
DISCUSSION
Most
mammals
are
highly
sensitive
to
snake
neurotoxins.
The
mongoose,
however,
is
resistant
to
neurotoxins
and
can
overcome
various
snakes
and
feed
on
them.
As
we
have
shown
in
this
study,
the
highly
curarimetric
toxin
a-BTX
does
not
bind
to
mongoose
AcChoR
in
vivo
or
in
vitro
and
thus
is
not
toxic
in
this
animal.
To
understand
the
molecular
basis
for
the
resistance
of
mongoose
AcChoR
to
a-BTX,
we
cloned,
sequenced,
and
expressed
a
cDNA
fragment
corre-
sponding
to
residues
122-205
of
the
mongoose
AcChoR
a
subunit.
We
have
also
cloned
and
sequenced
the
same
cDNA
fiagment
from
a
primitive
snake
Eryx
jaculus
(sand
boa),
which
like
the
other
snakes
studied
(Natrix
and
cobra)
(15),
was
shown
to
be
resistant
to
a-BTX.
Comparison
of
the
AcChoR
binding
site
domains
of
ani-
mals
that
are
susceptible
or
resistant
to
a-neurotoxins
is
an
Fir.