Biochimica et Biophysica Acta 1717 (2005) 27 – 33

http://www.elsevier.com/locate/bba

Inhibition of crotoxin binding to synaptosomes by a receptor-like protein

from Crotalus durissus terrificus (the South American rattlesnake)B
Roberta Márcia Marques dos Santos a,b, Leida Calegário Oliveira c, Maria Inácia Estevão-Costa a,b,
Maria Elena de Lima b,c, Marcelo Matos Santoro b, Consuelo Latorre Fortes-Dias a,*
a
Centro de Pesquisa e Desenvolvimento, Fundação Ezequiel Dias (FUNED), R. Conde Pereira Carneiro 80, CEP 30510-010, Belo Horizonte, Minas Gerais, Brazil
b
Departamento de Bioquı́mica e Imunologia, Universidade Federal de Minas Gerais,
Av. Antônio Carlos 6627, CEP 31161-970, Belo Horizonte, Minas Gerais, Brazil
c
Departamento de Fisiologia e Biofı́sica, Universidade Federal de Minas Gerais, Av. Antônio Carlos 6627, CEP 31161-970, Belo Horizonte, Minas Gerais, Brazil

Received 5 April 2005; received in revised form 29 May 2005; accepted 17 June 2005
Available online 3 October 2005

Abstract

Crotoxin (Ctx) is a potent neurotoxin of the venom of Crotalus durissus terrificus (the South American rattlesnake). Ctx is a heterodimer
composed of CB, a toxic PLA2 subunit, and CA, a non-toxic and non-enzymatic subunit, that potentiates the neurotoxicity of CB in vivo. The
deleterious action of Ctx upon C. d. terrificus snakes themselves is known to be prevented by a PLA2 inhibitor (CNF) present in their blood
serum. CNF acts by replacing CA in Ctx, thus forming a new stable complex CNF – CB. This complex no longer interacts with the target receptor
(TR) to deliver CB to cause its lethal effect. Furthermore, CNF – CB seems to be reminiscent of the interaction Ctx – TR at the pre-synaptic site. In
the present work, the binding competition between rat brain synaptosomes (TR) and CNF for Ctx was investigated. Radiolabeled Ctx, made of CA
and one isoform of CB (CA – 125ICB2), was used as ligand. The competition by unlabeled Ctx was taken as a reference. The potency of CNF as a
competitor was evaluated under different incubation conditions with varying time scale addition of reagents (CA – 125ICB2, synaptosomes and
CA – CB2 or CNF). CNF was able to inhibit the binding of the toxin to synaptosomes as well as to partially displace the toxin already bound to its
membrane target. The mechanisms of competition involved were discussed and a previous schematic model of interactions between Ctx, TR and
CNF was updated.
D 2005 Elsevier B.V. All rights reserved.

Keywords: Crotoxin; Phospholipase A2; Phospholipase A2 inhibitor; Synaptosome; CNF; Crotalus

1. Introduction tion of CB to membrane sites other than its target receptor on

pre-synaptic membranes [3]. Several isoforms of CA and CB
Crotalus durissus terrificus (the South American rattle- have been isolated, each one displaying slightly different
snake) venom is mainly composed of a potent neurotoxic enzymatic and pharmacological activities [4,5]. The ultimate
protein, crotoxin (Ctx), which is known to contain the bulk of properties of Ctx are determined by the isoform of CB present
the lethal toxicity of the crude venom. Crotoxin is a h- in the heterodimeric complex [2].
neurotoxin consisting of a heterodimer of a non-toxic, non- It has long been noted that C. d. terrificus snakes, as well as
enzymatic acidic protein CA and a basic protein CB with other snake species, are resistant to envenomation by their own
phospholipase A2 activity [1]. CA and CB are tightly linked in venom, due to the presence of toxin inhibitors in their blood. A
the Ctx complex by non-covalent, electrostatic forces [2]. The primary function ascribed to these inhibitors has been the
pharmacological action of Ctx is enhanced in vivo by CA that prevention of the deleterious action of toxins upon the snakes
acts as a chaperone, thus preventing the non-specific adsorp- themselves, in case of an eventual leaking of the contents of the
venom gland or a bite by another snake. In the case of C. d.
i
This work is part of the Doctoral thesis of R.M.M. dos Santos.
terrificus, it was shown that the lethal activity of the crude
* Corresponding author. Fax: +55 31 3371 1753. venom and of Ctx in mice can be neutralized by a protein
E-mail address: [email protected] (C.L. Fortes-Dias). inhibitor present in the homologous blood. The whole C. d.
0005-2736/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbamem.2005.06.014
28 R.M.M. dos Santos et al. / Biochimica et Biophysica Acta 1717 (2005) 27 – 33

terrificus plasma inhibits the phospholipase A2 enzymatic CB-agarose column, previously equilibrated with 0.1 M Tris – HCl/0.5 M NaCl
activity of Ctx in vitro as efficiently as a commercial anti- pH 8.0 buffer. This column was incubated overnight at 4 -C, with periodical
and automatic inversion, to assure the best contact between the matrix and the
serum, used in the treatment of victims of C. durissus sample solution. Then, the column was successively washed with 0.15 M
snakebites [6]. The antiserum is produced by hyper immuni- phosphate-citrate buffer of decreasing pH (7.0, 6.0, 5.0 and 4.0). Fractions (1.5
zation of horses with the whole venom and needs further ml) containing CNF were eluted with the same buffer at pH 3.0 and collected in
processing aiming at the concentration of the immunoglobulin tubes containing 150 Al of 1.0 M Tris – HCl pH 8.0 buffer. The final CNF
preparation was analyzed by SDS-PAGE in a 12.5% minigel (Bio-Rad
fraction and, consequently, the enhancement of its neutralizing
Laboratories, Inc.) according to Laemmli [16], before and after deglycosyla-
potency. An acidic glycoprotein, present in the a1-globulin tion. Deglycosylation was performed by incubating 2.5 Ag of CNF with 1 mU
fraction of the snake blood plasma, was purified and of PNGase F (2.5U/ml, Bio-Rad Laboratories, Inc.) in 50 mM Na2HPO4 buffer,
characterized, later on, as a Ctx inhibitor [7]. The native pH 7.5, at 37 -C for 24 h.
protein, named CNF [8] or CICS [9], exists as an oligomer of
molecular mass around 140 kDa. The aggregate is formed by 6 2.2. Crotoxin
to 8 single-polypeptide-chain subunits of one type [8]. The
C. d. terrificus snake venom was used for crotoxin purification [17,18].
exact number of subunits in the oligomer has not been clearly
Further fractionation into CA and CB isoforms was performed by reverse phase
determined yet. The primary structure of the monomer was chromatography [8].
deduced from the cDNA nucleotide sequence. It is composed Briefly, 0.5 mg of Ctx were loaded per run on a Sephasil Peptide C18 5A
of 181 amino acid residues with a calculated mass of 20.06 4.6/250 column (Pharmacia Biotech). The mobile phases were 0.1% TFA in
kDa and contains a putative N-glycosylation site. When CNF is H2O (A) and 0.1% TFA in acetonitrile (B). Fractions of 0.5 ml were eluted with
a gradient from 25% to 35% of B in 40 ml. CA was used as a mixture of the
incubated with Ctx, it displaces CA in the toxin complex and
isoforms present in the preparation, eluted between 29.3 and 30% of B. Only
binds to CB. The exchange reaction between CA and CNF the isoform CB2 (34.1 – 34.3% of B) was employed in the experiments.
leads to the formation of a new stable complex, CNF – CB that
no longer delivers CB to its target receptor. Quantitative 2.3. Iodination procedure
analysis of the CNF –CB complex demonstrated that it is most
likely formed by one CB per subunit of CNF and it is To avoid using of a mixture of CB isoforms or concomitant labeling of CA,
completely devoid of CA [8]. isolated CB2 was firstly radio iodinated using the chloramine-T method [19 –
21]. Then, the Ctx complex was reformed by incubating 125ICB2 with the
During the last 15 years, a series of natural PLA2 inhibitors
corresponding amount of CA in the native complex, for 15 min at room
(PLIs) have been isolated and characterized from the blood temperature. CA – 125ICB2 was loaded onto a 0.8
18 cm Sephadex G-25
plasma of snakes from different families ([10 –12] for reviews). column, previously calibrated with the unlabeled CA – CB2 complex. Fractions
The increasing number of isolated PLIs led to the proposal of of 0.75 ml were eluted with 0.1% BSA in PBS and the radioactivity present in
three different structural classes of blood inhibitors (a, h and g) aliquots of 5 Al each was counted in a Gamma Counter. Specific activities
ranged from 3700 to 4200 cpm/fmol of 125ICB2.
[13]. More recently, the sub classification of the g-type PLIs in
two subclasses, I and II, according to their heteromeric or
2.4. Synaptosomes
homomeric character, respectively, was proposed [12]. CNF
was placed in the subclass II of the g-type inhibitors, which Synaptosomes were prepared as described [22]. Briefly, the cerebral cortex
comprises homomeric PLIs with two structural units of highly of Wistar female rats was dissected and immersed in 0.32 mM sucrose, 1 mM
conserved three-finger motifs. CNF inhibits PLA2s from other EDTA, 0.25 mM DTT, pH 7.4 at 10% w/v. The tissue was homogenized and
snake venoms but no such effect was observed on the centrifuged at 1,000g for 10 min. Two ml of the supernatant were loaded onto
the top of a discontinuous Percoll gradient (23%, 15%, 10% and 3%) prepared
mammalian PLA2 tested so far [14,15]. This property would in the sucrose solution. After centrifugation at 32,500
g for 5 min, the
seem to be advantageous if CNF is considered as a model for synaptosomal fraction was collected at the interface between 15 and 23% of
the development of alternative drugs for the treatment of snake Percoll and diluted 4-fold in buffer A (25 mM HEPES, 10 mM glucose, 140
bites. mM choline, 5.4 mM KCl, 0.8 mM MgSO4, 1.8 mM CaCl2, 0.2% BSA, pH
It has been suggested that the interaction between CNF and 7.4). Synaptosomes were pelleted by centrifugation at 15,000
g for 15 min
and resuspended in about 8 ml of buffer A. This step was repeated until
CB may be reminiscent of the interaction of Ctx with its target complete removal of the Percoll in the sample was achieved. The final
receptor (TR) at the neuromuscular transmission site in the protein concentration was determined using bovine serum albumin as a
presynaptic cells [8,9]. Based on that, we decided to investigate standard [23].
more closely the effect of CNF on the interaction between Ctx
and TR, here represented by rat brain synaptosomes. Particular 2.5. General conditions for the binding assays
attention was devoted to this action on the Ctx already bound to
The binding experiments were performed based on Degn et al. [21].
TR, a condition that simulates the human envenomation by C.
Triplicates were used throughout and incubations were performed in a rocking
d. terrificus snake bite. water bath at 37 -C for 1 h, unless otherwise stated. After the incubation period,
the samples were filtered through a GFB disc, previously soaked in buffer A
2. Materials and methods containing 20 mg/ml of BSA. The radioactivity on the filters was counted for 1
min in a Gamma Counter, following two washings with 5 ml of buffer B (5 mM
2.1. Crotalus Neutralizing Factor (CNF) HEPES, 10 mM glucose, 140 mM choline, 5.4 mM KCl, 0.8 mM MgSO4, 1.8
mM CaCl2, 7.0% BSA, pH 7.4). Control samples were prepared with
CNF was purified from the plasma of C. d. terrificus snakes in a two step CA – 125ICB2 alone (total binding) and in the presence of a 1000-fold excess
procedure. First, a preparative isoelectric focusing using ampholytes within pH of CA – CB2 (nonspecific binding). Values for nonspecific binding to the discs,
range 3 – 10 was performed. CNF-containing fractions were then loaded on a obtained by filtering equivalent concentrations of CA – 125ICB2 alone, were
 R.M.M. dos Santos et al. / Biochimica et Biophysica Acta 1717 (2005) 27 – 33 29

Table 1
Incubation conditions, potency (IC50) and equilibrium dissociation constants (K d) in binding assays of Ctx (CA – 125ICB2) to rat brain synaptosomes (S), in the
presence of CNF or CA – CB2 as competitors

n.d. = not determined.

a
This value was obtained by extrapolation of the corresponding curve (C) in Fig. 3.

subtracted from the corresponding test samples. Data analyses were performed [21,24]. CA –CB2 and CA were able to completely inhibit
using the GraphPad Prism version 4 software. the specific binding of CA – 125ICB2 to rat brain synaptosomes
in a dose-dependent manner, with IC50 of 16 nM and 97 nM,
2.6. Competition experiments
respectively (Fig. 1). The competition profile of CB2 was
Varying concentrations of CA, CB2 or CA – CB2 in the range 1011 to 106 concentration-dependent, as it acted as an inhibitor at
M were used in the presence of a fixed concentration of 1010 M of intermediate concentrations only (108 to 106). After that, a
CA – 125ICB2. For CNF, concentrations between 1012 and 106 M were tendency to an increase in the CA – 125ICB2 specific binding
employed, under the following incubation conditions (Table 1): B- was noticed, following a polynomial pattern (Fig. 1). This
[CA – 125ICB2 + CNF + S] for 2 h; C-[CA – 125ICB2 + S] (1 h) plus CNF
paradoxical effect of the subunit CB has been also verified
(1 h); D-[CA – 125ICB2 + CNF] (1 h) plus S (1 h). A control sample (A in
Table 1) was prepared by incubating [CA – 125ICB2 + CA – CB2 + S] for 2 h. before [21,24,25].
Thirty Ag of synaptosomes were used per test sample. Once the biological system was validated, CNF was
Data were analyzed using non-linear regression of the sigmoidal dose – included in the experiments. The homogeneity of the CNF
response curve, according to the following equation preparation to be used was checked by SDS-PAGE (Fig. 2).
ðTop  BottomÞ Two components of molecular masses 23 and 18 kDa,
Y ¼ ðBottomÞ þ
1 þ 10ðXLogIC50Þ:HillSlope respectively, were detected. Further treatment with PNGase F
The potency of the competitors was expressed as the concentration of inhibitor clearly demonstrated that the 18-kDa component corresponded
that competes for half the specific binding (IC50). to the deglycosylated form of the CNF monomer. CNF was,

2.7. Saturation binding experiments

Specific binding at equilibrium was measured at concentrations of

CA – 125ICB2 in the range 1011 to 109, in the presence of a 1000-fold
excess of CA – CB2 or CNF and 1 Ag of synaptosomes. The equilibrium
dissociation constants (K d) were determined after fitting the data to a
rectangular hyperbola using non-linear regression.

3. Results

The results for the biological system used in the present

work, that is, CA – 125ICB2 as the toxin and rat brain
synaptosomes as target receptors, were firstly validated by
comparison with saturation and competition binding data of
unlabeled Ctx and its subunits, already reported for analogous
systems. Typical association curves were obtained (data not
shown), in accordance to those previously reported for whole Fig. 1. Inhibition of the specific binding of CA – 125ICB2 to synaptosomes by
labeled 125ICtx and guinea pig synaptosomal membrane the unlabeled complex and its isolated components. CA – 125ICB2 was
fragments [21] or presynaptic membranes from Torpedo incubated with synaptosomes in the presence of increasing concentrations of
marmorata electric organ [24]. The inhibition of CA – 125ICB2 CA – CB2, CB2 or CA. B0 is the concentration of CA – 125ICB2 bound in the
absence of any competitor, whereas B is the concentration of CA – 125ICB2
binding to rat brain synaptosomes by unlabeled CA – CB2 and bound in the presence of a competitor at a given concentration. Symbols: (o)
its isolated subunits, CA or CB2, also followed the same pattern CA – CB2, (g) CA and (r) CB2. Values are expressed as mean T S.D. of
already published for the other two neuronal membranes triplicates.
30 R.M.M. dos Santos et al. / Biochimica et Biophysica Acta 1717 (2005) 27 – 33

Fig. 3. Inhibition of the specific binding of CA – 125ICB2 (10 10 M) to

synaptosomes (S) by CNF. B0 is the concentration of bound CA – 125ICB2 in the
Fig. 2. SDS-PAGE (12.5%) of CNF (2.5 Ag per well), before and after absence of any competitor, whereas B is the concentration of bound
deglycosylation with PNGase F. The gel was silver impregnated. The molecular CA – 125ICB2 in the presence of a given concentration of a competitor.
masses of standard proteins (prestained broad range markers, Bio-Rad Unlabeled CA – CB2 was used as control. Order of addition of reagents and
Laboratories, Inc.) are indicated on the left side. incubation times (in brackets) were as follows: (n) A. [CA – 125ICB2+ S+ CA –
CB2] (2 h)-reference curve; (o) B. [CA – 125ICB2+ S+ CNF] (2 h); (r) C.
[CA – 125ICB2+ S] (1 h) plus CNF (1 h). Inset: D. (q) [CA – 125ICB2+ CNF] (1
then, used in the experiments as the natural mixture of h) plus S (1 h). Values are expressed as mean T S.D. of triplicates.
glycosylated and deglycosylated monomers.
Competition binding assays with synaptosomes were on the fact that CNF binds preferentially to CB2 [8].
performed with a fixed concentration of CA – 125ICB2 and Nevertheless, our results should agree with those performed
increasing concentrations of CNF over six orders of magnitude. with mixtures of CB isoforms, considering that CB2 is the
The time scale of addition of reagents was varied as detailed in major isoform in every venom of C. d. terrificus analyzed in
Table 1 (B to D). Incubation periods of 2 h for one-step and 1 our laboratory, so far.
h each for two-step reactions were based on previous kinetic Although the activity of the Ctx complex was not directly
experiments, where equilibrium plateaus were reached after 40 assayed after radio labeling, it was shown before that Ctx’s
min of reaction, in typical hyperbolic curves for every case. structure and function were not greatly modified by the
Competition by unlabeled CA – CB2 was run in parallel to be iodination method used in the present work [21,24]. Besides,
taken as reference (Table 1A). Under all experimental condi- the ready replacement of CA – 125ICB2 by unlabeled Ctx
tions, a dose-dependent inhibition was obtained for CNF (Fig.
3). When an excess of CNF was simultaneously incubated with
CA – 125ICB2 and synaptosomes (Table 1B) the IC50 (100 nM)
was very close to that determined before for CA (97 nM,
Fig. 1). Pre-incubation of CA – 125ICB2 and CNF followed
by the addition of synaptosomes (Table 1D) reduced the
initial ratio B/B o to 60% (inset of Fig. 3). An IC50 of 25
nM was found. When the labeled toxin was already bound
to the synaptosomes (Table 1C), the displacement by CNF
reached a maximum of 45%, even with the highest dose of
CNF tested (1 AM) (Fig. 3). This percentage remained
unchanged even after prolonged incubation times up to 240
min (data not shown).
Saturation curves for the binding of labeled Ctx to
synaptosomes, in the presence of CA – CB2 or CNF as
competitors, are shown in Fig. 4. Equilibrium dissociation
constants (K d) of 1.2 T 0.3 nM and 3.2 T 1.2 nM were
determined for CA – CB2 and CNF, respectively.

4. Discussion Fig. 4. Saturation binding curves of CA – 125ICB2 to rat brain synaptosomes, in

the presence of CA – CB2 (g) and CNF (.) as competitors. The values are
mean of triplicates. Data fitting to a rectangular hyperbole gave equilibrium
The strategy of using a Ctx complex made of one single dissociation constants (K d) of 1.2 T 0.3 nM and 3.2 T 1.2 nM for CA – CB2 and
isoform of CB and total CA (CA –CB2) was employed based CNF, respectively.
 R.M.M. dos Santos et al. / Biochimica et Biophysica Acta 1717 (2005) 27 – 33 31

offered indirect evidence that the procedure did not compro- account the chaperone role of CA in the homologous
mise the binding properties of the toxin. competition. An additional consideration is that, when CNF is
The presence of calcium ions in binding studies of toxins the competitor, part of 125ICB2 is sequestered from CA – 125ICB2
with PLA2 activity has been a matter of discussion in the to give rise to the CNF – 125ICB2 complex (Table 1B). The
literature. The kinetic parameters for Ctx binding to guinea pig formation of this new complex, with CNF replacing CA in Ctx
synaptosomes were almost unaffected by the presence of [8], was more evident when CNF was pre-incubated with
calcium ions [21]. For Torpedo membranes, instead, concen- CA – 125ICB2, before the addition of synaptosomes (Table 1D).
trations between 1 and 10 mM CaCl2 have been considered Under this assay condition, the initial ratio B/B o was reduced to
ideal for the specific binding of Ctx [26]. However, an increase 60%, due to the formation of that new stable complex in the first
in the non-specific binding of Ctx was unexpectedly observed incubation step. The real amount of CA – 125ICB2 available to
after 10 min. This effect, absent when the calcium ions were synaptosomes in the second incubation period was, then, greatly
omitted, was attributed to the hydrolysis of the membranes due reduced (Table 1D, inset of Fig. 3).
to the addition of a large amount of PLA2 activity [26]. For rat Comparable IC50 values were found for CNF and CA as
brain synaptosomes (present study), the non-specific binding of competitors (100 and 97 nM, respectively). CA alone,
CA – 125ICB2 was stable up to 90 min in the presence of 1.8 obviously, does not play its chaperone role. CNF can be
mM CaCl2 (data not shown) and this concentration was considered, then, as potent a competitor as CA for the Ctx
maintained throughout. binding to rat brain synaptosomes. (Figs. 1 and 3).
In an extensive study on the inhibition of Ctx binding to Although unlabeled Ctx was taken as a reference for the
guinea pig synaptosomal membrane fragments [21], a series of assays with CNF, the molecular models being dealt with are
competitors have been classified as strong, moderate, weak, quite different. The reference model comprises two ligands
very weak and non-inhibitors, based on their strength of (labeled and unlabeled Ctx) competing for the same receptor
inhibition. Native Ctx was among the strongest inhibitors site (S) on membranes, that is, two ligands versus one
tested, with an IC50 around 10 nM [21,24]. A close value was receptor. Two new complexes, S-125ICB2 and S-CB2, will be
found here for rat brain synaptosomes (IC50 = 16 nM) formed. As of CNF, it will not be competing for the receptor
demonstrating an agreement in the results with both biological site (S) but for the ligand (CB) instead, through two different
preparations. As for CA, its potency seemed more dependent pathways. Firstly, the formation of the stable CNF – 125ICB2
on the biological membrane. CA was as potent as Ctx for complex (Table 1B) will sequester part of 125ICB2, reducing
guinea pig synaptosomal membrane fragments [21] but its IC50 the amounts of the labeled ligand available to binding to the
increased ten times for Torpedo membranes [24] and for rat synaptosomes. Secondly, CNF will act as an alternative
brain synaptosomes (present work). So, based on these IC50 receptor for 125ICB2 already bound to the synaptosomes.
values (around 100 nM), CA should not be considered as a These characteristics make the competition by CNF a unique
strong but a moderate inhibitor for the latter two target and complex model.
membranes. In the absence of a more adequate mathematical treatment
CNF was the first member of the g-type PLIs to be to describe this peculiar competition, the equation of the law
purified and characterized as an oligomer of polypeptide of mass action, commonly employed for analysis of
subunits of a single type, in its native form [8]. Later on, radioligand binding experiments, was applied to our data.
when a g-PLI from Laticauda fasciata snakes was described The equilibrium dissociation constant (K d) of S-125ICB2 was
as composed of two different types of polypeptide subunits, determined as 1.2 T 0.3 nM in the presence of CA – CB2 as
the homomeric nature of CNF was questioned [27]. At that competitor (Fig. 4, Table 1A). This value is very close to that
time, the conclusion was that a SDS-PAGE band of 20 kDa,
that appeared in the CNF preparation originally reported [7],
was most probably the second subunit of CNF. Presently, the
SDS-PAGE profiles obtained before and after deglycosylation
of CNF (Fig. 2), leave no doubt about the homomeric nature
of CNF. The positioning of CNF in the subclass II of g-type
PLIs can, thus, be no longer contested. Nevertheless, the
oligomeric aggregates of CNF are formed by a mixture of
glycosylated and deglycosylated monomers, the ratio of
which will vary among specimens of C. d. terrificus snakes
(unpublished results).
A direct competition between the target receptor and CNF
for 125ICB2 could be followed with the simultaneous incuba-
tion of CNF, CA – 125ICB2 and synaptosomes (Table 1B, Fig.
3). Comparison of the IC50 value for CA – CB2 and CNF
indicates that CA –CB2 is about ten times more potent as a Fig. 5. Schematic representation of the interactions between Ctx, its target
competitor than CNF, under our experimental conditions (Table receptor (TR) on pre-synaptic membranes and CNF [8], updated with the
1A and B). This finding would be expected, though, taking into inclusion of a new equilibrium step between CNF – CB and TR – CB.
32 R.M.M. dos Santos et al. / Biochimica et Biophysica Acta 1717 (2005) 27 – 33

reported for Ctx and guinea pig synaptosomes (K d = 2 nM) [5] G. Faure, C. Bon, Crotoxin, a phospholipase A2 neurotoxin from
[21]. When CA – CB2 was replaced by CNF, a K d of 3.2 T 1.2 the South American rattlesnake Crotalus durissus terrificus: purifi-
cation of several isoforms and comparison of their molecular
nM was found. This increased value of K d does not mean structure and of their biological activities, Biochemistry 27 (1988)
that CNF is a better competitor than Ctx because the 730 – 738.
mechanisms of competition in each case are different, as [6] C.L. Fortes-Dias, C.R. Diniz, E. Kochva, Neutralization of Crotalus
already discussed. This K d reflects an apparent increase in the durissus terrificus (South American rattlesnake) venom and crotoxin by
homologous plasma, Cienc. Cult. 42 (1990) 501 – 506.
dissociation of S-125ICB2 due to a decrease in the real
[7] C.L. Fortes-Dias, B.C.B. Fonseca, E. Kochva, C.R. Diniz, Purification
concentration of 125ICB2 available for the membrane receptor and properties of an antivenom factor from the plasma of the South
associated to a competition of CNF for 125ICB2 bound to the American rattlesnake (Crotalus durissus terrificus), Toxicon 29 (1991)
synaptosomes. 997 – 1008.
The paradoxical concentration-dependent effect of CB for [8] C.L. Fortes-Dias, Y. Lin, J. Ewell, C.R. Diniz, T.Y. Liu, A phospholipase
guinea pig synaptosomal fragment [21] and for Torpedo A2 inhibitor from the plasma of the South American rattlesnake (Crotalus
durissus terrificus). Protein structure, genomic structure, and mechanism
membranes [24] was confirmed for rat brain synaptosomes of action, J. Biol. Chem. 269 (1994) 15646 – 15651.
(Fig. 1). The potentiation effect for concentrations of CB2 [9] J. Perales, C. Villela, G.B. Domont, V. Choumet, B. Saliou, H.
above 106 M could be possibly explained by its tendency to Moussatche, C. Bon, G. Faure, Molecular structure and mechanism of
form aggregates due to its highly basic character [28], its action of the crotoxin inhibitor from Crotalus durissus terrificus serum,
binding to nonspecific sites on the membranes or both [25]. In Eur. J. Biochem. 227 (1995) 19 – 26.
[10] R.D. Dunn, K.W. Broady, Snake inhibitors of phospholipase A2 enzymes,
either case, the final result will be a decrease in the Biochim. Biophys. Acta 1533 (2001) 29 – 37.
concentration of free CB competing for the membrane receptor [11] C.L. Fortes-Dias, Endogenous inhibitors of snake venom phospholipases
and the induction of an increase in the ‘‘specific binding’’ of A2 in the blood plasma of snakes, Toxicon 40 (2002) 481 – 484.
labeled Ctx, as already suggested [24]. [12] S. Lizano, G. Domont, J. Perales, Natural phospholipase A2 myotoxin
In conclusion, our results demonstrate that CNF is able to inhibitor proteins from snakes, mammals and plants, Toxicon 42 (2003)
963 – 977.
totally inhibit the pharmacological action of Ctx and that this [13] N. Ohkura, H. Okuhara, S. Inoue, K. Ikeda, K. Hayashi, Purification and
action can be partially extended to the toxin already bound to characterization of three distinct types of phospholipase A2 inhibitors
synaptosomes. The schematic model for the interaction of Ctx from the blood plasma of the Chinese mamushi, Agkistrodon blomhoffii
with its target receptor and CNF proposed before [8] can thus siniticus, Biochem. J. 15 (1997) 527 – 531.
be updated with the inclusion of an additional equilibrium [14] C.L. Fortes-Dias, M.L.D. Jannotti, F.J.L. Franco, A. Magalhães, C.R.
Diniz, Studies on the specificity of CNF, a phospholipase A2 inhibitor
step between CNF – CB and TR – CB, representing the isolated from the blood plasma of the South American rattlesnake
competition between TR and CNF for CB (Fig. 5). CNF, in (Crotalus durissus terrificus): I. Interaction with PLA2 from Lachesis
its natural form or modified, may constitute a useful tool to muta muta snake venom, in the blood plasma of snakes, Toxicon 37
investigate the mechanism of action and intoxication by Ctx (1999) 1747 – 1759.
[15] G. Faure, C. Villela, J. Perales, C. Bon, Interaction of the neurotoxic and
and other PLA2-toxins on targets in the nervous system, as
nontoxic secretory phospholipases A2 with the crotoxin inhibitor from
well as in the development of a future therapeutics for snake Crotalus serum, Eur. J. Biochem. 267 (2000) 4799 – 4808.
bites. [16] U.K. Laemmli, Cleavage of structural proteins during the assembly of the
head of bacteriophage T4, Nature 227 (1970) 680 – 685.
Acknowledgements [17] K. Slotta, H. Fraenkel-Conrat, Snake venoms II. Nature of the sulphur
union, Ber. Dtsch. Chem. Ges. 71B (1938) 1076 – 1081.
[18] C. Seki, J.C. Vidal, A. Barrio, Purification of gyroxin from a South
We thank Ms. A.C. Valentim for technical assistance, Dr.
American rattlesnake (Crotalus durissus terrificus), Toxicon 18 (1980)
L.G.D. Heneine and Mrs. P. Cotta for their help in the 235 – 247.
preparation of the affinity column and Dr. M. Richardson for [19] F.C. Greenwood, W.M. Hunter, J.S. Glover, The preparation of 131I-
language review. This work was granted by FAPEMIG (EDT labelled human growth hormone of high specific radioactivity, Biochem.
24000/01; CBB 492/03) and by CNPq with fellowships to J. 89 (1963) 114 – 123.
R.M.M. dos Santos (141781/2000-9) and to C.L. Fortes-Dias [20] A. Johnstone, R. Thorpe, Immunochemistry in Practice, Blackwell
Scientific Publications, Oxford, 1982, pp. 107 – 109.
(300767/2003-0). [21] L.L. Degn, C.S. Seebart, I.I. Kaiser, Specific binding of crotoxin to
brain synaptosomes and synaptosomal membranes, Toxicon 29 (1991)
References 973 – 988.
[22] P.R. Dunkley, J.W. Heath, S.M. Harrison, P.E. Jarvie, P.J. Glenfield, J.A.
[1] R.A. Hendon, H. Fraenkel-Conrat, Biological roles of the two components Rostas, A rapid Percoll gradient procedure for isolation of synaptosomes
of crotoxin, Proc. Natl. Acad. Sci. U. S. A. 68 (1971) 1560 – 1563. directly from an S1 fraction: homogeneity and morphology of subcellular
[2] K. Rübsamen, H. Breithaupt, E. Habermann, Biochemistry and pharma- fractions, Brain Res. 441 (1998) 59 – 71.
cology of the crotoxin complex, Naunyn-Schmiedeberg’s Arch. Pharma- [23] O.H. Lowry, N.J. Rosebrough, A.L. Farral, R.J. Randall, Protein
col. 270 (1971) 274 – 288. measurement with the Folin phenol reagent, J. Biol. Chem. 193 (1951)
[3] C. Bon, C. Bouchier, V. Choumet, G. Faure, M.S. Jiang, M.P. 265 – 275.
Lambezat, F. Radvanyi, B. Saliou, B, Crotoxin, half-century of [24] I. Kri”aj, G. Faure, F. Gubenek, C. Bon, Neurotoxic phospholipases A2
investigations on a phospholipase A2 neurotoxin, Acta Physiol. Latinoam. ammodytoxin and crotoxin bind to distinct high-affinity protein
39 (1989) 439 – 448. acceptors in Torpedo marmorata electric organ, Biochemistry 36
[4] G. Faure, C. Bon, Several isoforms of crotoxin are present in individual (1997) 2779 – 2787.
venoms from the South American rattlesnake Crotalus durissus terrificus, [25] I. Kri”aj, G. Faure, F. Gubenxek, C. Bon, Re-examination of crotoxin –
Toxicon 25 (1987) 229 – 234. membrane interactions, Toxicon 34 (1996) 1003 – 1009.
 R.M.M. dos Santos et al. / Biochimica et Biophysica Acta 1717 (2005) 27 – 33 33

[26] E. Délot, C. Bon, Model for the interaction of crotoxin, a phospholipase plasma of the Sea Krait, Laticauda semifasciata, J. Biochem. 125 (1999)
A2 neurotoxin, with presynaptic membranes, Biochemistry 32 (1993) 375 – 382.
10708 – 10713. [28] E. Habermann, H. Breithaupt, Mini-review—The crotoxin complex, an
[27] N. Ohkura, Y. Kitahara, S. Inoue, K. Ikeda, K. Hayashi, Isolation and example of biochemical and pharmacological protein complementation,
amino acid sequence of a phospholipase A2 inhibitor from the blood Toxicon 16 (1978) 19 – 30.