NEUROTRANSMITTER EVALUATION IN THE HIPPOCAMPUS OF RATS AFTER INTRACEREBRAL INJECTION OF TsTX SCORPION TOXIN

TsTX is an α-type sodium channel toxin that stimulates the discharge of neurotransmitters from neurons. In the present study we investigated which neurotransmitters are released in the hippocampus after TsTX injection and if they are responsible for electrographic or histopathological effects. Microdialysis revealed that the toxin increased glutamate extracellular levels in the hippocampus; however, levels of gamma-aminobutyric acid (GABA), glycine, 5-hydroxyindoleacetic acid (5HIAA), homovanillic acid (HVA) and 3,4-dihydroxyphenylacetic acid (DOPAC) were not significantly altered. Neurodegeneration in pyramidal cells of hippocampus and electroencephalographic alterations caused by the toxin were blocked by pretreatment with riluzole, a glutamate release inhibitor. The present results suggest a specific activity of TsTX in the hippocampus which affects only glutamate release.


INTRODUCTION
The toxicity of scorpion venoms is due to neurotoxins that are composed of basic polypeptides with molecular weights of approximately 7,000 Da. These toxins affect cell permeability to ions including sodium, potassium, chloride and calcium (1)(2)(3)(4).
Sodium channel neurotoxins affect sodium conductance in various excitable tissues, thus serving as important pharmacological tools for the study of excitability and sodium channel structure (5).
Scorpion neurotoxins that act on sodium channels have been divided into two groups, α and β toxins, according to the ligand-binding sites in the channel. The αscorpion toxins were the first to be studied. Their primary effect is interference in channel inactivation upon prolonged depolarization, increasing sodium permeability and consequently extending the duration of the action potential (6). One of the main consequences is the augmented release of neurotransmitters from neuronal endings (7). Gomez and Diniz (8) were the first to isolate toxic polypeptides from the venom of the Brazilian scorpion Tityus serrulatus. According to Arantes et al. (9), TsTX was first isolated by Coutinho Netto in 1975. This α-type sodium channel toxin increases the release of glutamate, acetylcholine and GABA from different preparations of nervous tissue (10)(11)(12)(13). When injected in the dorsal hippocampus of rats, it induces behavioral alterations and epileptic discharges besides neurodegeneration (14).
The aim of the present study was to investigate which neurotransmitters are released after TsTX intrahippocampal injection. To do this, microdialysis analysis was performed to evaluate extracellular levels of some cerebral neurotransmitters.
In a previous study, we observed an increase in extracellular levels of glutamate only until three hours after toxin injection (11). Based on the toxin kinetic and on the glutamate release profile observed in that study, we decided to lengthen the time of collection (15).
Moreover, since the previous work by Nencioni et al. (11) had indicated the involvement of glutamate in the convulsive/neurodegenerative effect of the toxin; in the present study, we examined the result of pretreatment with riluzole (2-amino-6trifluoromethoxy benzothiazole, RP54274), a drug that inhibits glutamate release and presents neuroprotective, anticonvulsant, anxiolytic and anesthetic properties (16). Nencioni

Subjects
Male Wistar rats (200 to 250 g), obtained from an established colony kept by the Central Animal House Service at the Butantan Institute, were used. Upon their arrival in the laboratory (seven days before the experiments), the animals were individually housed in wire mesh cages and kept at constant temperature (22 ± 1°C), in a 12hour light/12-hour dark cycle (lights on at 0700 h), with food and water provided ad libitum.
The animals employed in the current study were maintained in accordance with the policies of the Ethics Committee on Use of Laboratory Animals, Butantan Institute, Brazil.

Surgery
Rats were anesthetized with an intraperitoneal (IP) injection of 3 mL/kg of a mixture of pentobarbitone (1 g) and chloral hydrate (4 g) diluted in 100 mL of 0.9% NaCl.
Animals were placed in a stereotaxic frame and a microdialysis guide cannula was implanted unilaterally in their dorsal hippocampus. Guide cannulas were fixed to the skulls with stainless steel jeweler screws and dental acrylate. The guide cannulas for microdialysis were implanted in rats according to the following coordinates: 5.3 mm posterior to bregma, 3.2 mm lateral to midline, and 2 mm below dura mater (17). The cannulas were used for both toxin injection and dialysate collection. In rats undergoing behavioral studies and EEG recording, the coordinates were 4.8 mm posterior to bregma, 3.2 mm lateral to midline, and 2.8 mm below dura mater. In these rats, bipolar twisted electrodes for depth recordings were implanted contralaterally at the same coordinates and anchored to the skull as were the cannulas. Jeweler screws were inserted bilaterally in the skull over the occipital cortex for surface recordings. A screw placed in the frontal sinus was employed as reference (indifferent electrode). The surgery was conducted using aseptic techniques and animals with surgical complications were excluded. Subsequently, rats were individually housed and allowed to recover for one or two days. µg intrahippocampal injection of TsTX. Six rats from this latter group were employed for EEG recording and histology, while the remainder were utilized for amino acid analysis.

Microdialysis
Microdialysis was performed in freely-moving rats between 24 and 48 hours after

Measurement of Neurotransmitter Concentration
Extracellular neurotransmitter levels in the dialysate were expressed as μg/μL or μg/mL. The first four samples from each animal were used as control and compared with post-treatment samples of the same individual. The sample was mixed with 50 µL phenylisothiocyanate, and distilled water was added to a 350-µL volume. Levels of glutamate, glycine and GABA from the dialysate were determined by means of HPLC with UV detection at 254 nm.

Electroencephalographic Recordings and Behavioral Observations
Electroencephalographic recordings and behavioral observations were carried out in a glass compartment placed in a Faraday cage. Animals were connected to a PowerLab® recording apparatus (ADInstruments, Canada) and allowed to settle down for 15 minutes. Subsequently, basal electroencephalographic trace was recorded for 15 minutes. In rats treated with riluzole, the drug basal trace was analyzed for additional 15 minutes after its injection. Afterwards, an intracerebral injection of the toxin was administered through a 5 µL Hamilton® microsyringe (Hamilton Company, USA) connected with polyethylene tubing to an injection needle.
Then, EEG was recorded and the behavior was observed for extra four hours.

Histology
The location of implanted electrodes and guide cannulas was histologically examined. Animals that presented missing target areas were excluded. Seven days after the injection, animals were completely anesthetized with CO 2 and received a cardiac perfusion (left ventricle) with phosphate-buffered saline (PBS) solution followed by 10% formalin solution. The brains were removed, stored in formalin for at Nencioni  least one week, and embedded in Paraplast® (Oxford Labware, USA). Coronal brain sections of 10 µm were cut on microtome from a 700 µm brain block containing the cannula track. Every sixth tissue slice was mounted on a glass slide and stained with cresyl violet. And the other five slices from each animal had their hippocampal field analyzed. The number of cells in the CA1, CA3 and CA4 hippocampal areas was examined through a light microscope at 400x magnification. A two-dimensional cell counting was performed using a 100 x 100 µm reticulum. Only pyramidal neurons located in the area of the reticulum that had a visible nucleus and nucleolus were considered intact. ANOVA followed by Tukey's test were employed for statistical analysis (p < 0.05).

Monoamine Levels
TsTX did not significantly affect levels of homovanillic acid ( Error bars show standard deviation p < 0.05 compared to pre-injection level, one-way repeated measures ANOVA and Tukey's test. All groups n = 5.

Extracellular Amino Acid Levels
Injection of TsTX toxin into the hippocampus augmented glutamate concentration in extracellular fluid (Figure 2), which reached statistical significance three hours after the injection. TsTX did not significantly affect levels of GABA and glycine.

Electroencephalographic Recordings and Behavioral Observations
Ringer solution injection into CA1 hippocampal area caused no alterations in behavior or in electroencephalographic records. Results of TsTX intrahippocampal injection were previously described and included long electroencephalographic epileptic-like discharges, immobility, orofacial movements and "wet dog shakes" in all animals (14). The effects started five to ten minutes after the injection and persisted for the entire recording period.
Rats treated with riluzole or with riluzole and toxin did not show epileptiform activity in electroencephalographic records or behavioral signs of convulsion.

Histology
Injection of Ringer solution into the hippocampus did not influence the number of cells in this area neither in the ipsilateral or contralateral regions (Figure 3).
According to previous descriptions, outcomes provoked by intracerebral injection of the toxin were degeneration of pyramidal cells in ipsilateral and contralateral CA1 regions, and in CA3 and CA4 fields of the ipsilateral hippocampus (11,14). Many dark-stained neurons were observed, which indicated the degeneration. Pretreatment

DISCUSSION
Among scorpion neurotoxins, the α-group is the most studied and the more useful in the functional mapping of the sodium channel structure (18).
The present work aimed to assess the profile of in vivo neurotransmitter release in the hippocampus after TsTX injection. Based on the toxin kinetics and on the glutamate release profile registered in a previous study, in the current study the collection period was prolonged up to six hours after the injection and other neurotransmitters were analyzed (11,15). It was observed that the toxin injected into the hippocampus did not modify extracellular levels of 5-HIAA, HVA, DOPAC, glycine or GABA. Nevertheless, extracellular levels of glutamate were significantly increased.
These findings agree with Nencioni et al. (11) who observed enhanced levels of glutamate for three hours after the toxin injection and no alterations in GABA or glycine levels.
According to Nunan et al. (15), TsTX can be found in rat brain after subcutaneous (SC) injections. In adult animals the toxin increase is smaller and slower than in young ones. The highest brain concentration of the toxin was observed one hour after the inoculation in young rats and three hours later in adult ones, in which the levels were augmented up to 12 hours. These results were in accordance with ours, which revealed continuous enhancing in glutamate release that persisted until the end of the collection and probably after (six hours or more). Moreover, considering that in our study the toxin was injected directly into the brain, there was a large quantity of glutamate in the hippocampus.
Similar results had been obtained by Dorce and Sandoval (25) with crude venom, when DA, HVA, NE and 5-HIAA levels as well as glutamic acid decarboxylase (GAD) activity were determined after intravenous (IV) or intrastriatal injection of Tityus serrulatus crude venom. Subsequently, an increase in HVA levels and a decrease in GAD activity in the striatum and hypothalamus were observed. On the other hand, no alteration was registered in neurotransmitter levels in the hippocampus.
Previous data provided evidences that glutamate is the main responsible for TsTX neurotoxic effects since glutamate receptor antagonists totally or partially block the toxin effects (11). This hypothesis was corroborated by the utilization of riluzole, a substance that belongs to a class of anticonvulsant and neuroprotective agents which selectively inhibits glutamate release over release of other neurotransmitters (26,27). Zona et al. (28) concluded that riluzole modulates Na + currents and the late anticonvulsant and neuroprotective properties of this compound. A decrease of the voltage-activated Na + current certainly reduces neuronal excitability and enhances excitatory amino acid release (28). Under experimental conditions, the drug prevented audiogenic convulsion in DBA/2 mice, maximal electroshock and amygdala-kindled seizures in rats, as well as delayed the appearing of seizures and reduced the duration of afterdischarges (29,30).
Riluzole injection 15 minutes before the toxin inoculation abolished all electrographic alterations as well as neuronal losses. Riluzole, when intraperitoneally injected, achieves the entire brain, including ipsi and contralateral areas of hippocampus, preventing the excitotoxic effects of the toxin. Animals did not show behavioral alterations either. They slept throughout the collection period, which is consistent with the finding that riluzole easily crosses the blood brain barrier and enhances slowwave and rapid eye movement sleep (31). The level of extracellular glutamate remained elevated and reached statistical significance in 2 points in rats pretreated with riluzole; however, it is not enough to provoke epileptiform activity or neuronal loss.
According to Meldrum (32), enlarged release of glutamate may worsen or prolong preexisting seizure activity, while GABA release may be a compensatory inhibitory mechanism that limits the progression and spread of seizure activity (33). However, hippocampal augmentation of extracellular glutamate and GABA levels are directly related to seizure activity rather than to the convulsant agent (34). Meurs et al. (34) described that maximal glutamate increases occurred in the early stages of seizure development, and preceded maximal GABA elevation. Nevertheless, in our results, the electrographic convulsive activity did not keep a relation with the maximal extracellular glutamate level. Seizure activity onset occurred (approximately 15 minutes after toxin injection) before glutamate extracellular level reach a significant increase (approximately three hours after the injection), indicating that it is not necessary the maximal increase in glutamate extracellular levels to initiate convulsions. The convulsive activity enhanced by TsTX is not so intense to increase GABA levels. Rats treated with riluzole and TsTX maintained elevated levels of glutamate, although, they did not present behavioral or electrographic signs of convulsion neither neuronal loss. The glutamate level in these animals was not as high as in the group treated only with the toxin. It appears that riluzole can block In hippocampal neurons, voltage-gated sodium channels appear to be the primary, if not the only, target responsible for synaptic effects of riluzole at low micromolar concentrations. Riluzole's voltage-dependent blockade of sodium channels enhances the depression of excitatory postsynaptic currents, which does not apply to inhibitory postsynaptic currents determined by GABA. Inhibition of sodium channels account for the preferentially anti-glutamate effect of this drug. Therefore, the preferential depression of glutamate release is explained by a direct effect of riluzole on glutamatergic cells (35).
These findings constitute an evidence of a putative specific activity of TsTX, at least in the hippocampus, that affects only glutamate release.
In our opinion, outcomes of TsTX in the hippocampus may be due to its interaction with a specific subtype of sodium channel. Different actions on sodium channels produce diverse consequences on neurotransmitter release that involve distinct presynaptic calcium channels, which supports the idea that sodium channels may modulate neurotransmitter release (22). Depending on the rate of increase in channel conductance, the outcome in terms of neurotransmitter release and calcium channel type coupled to that event is different (36).
A large number of biological toxins exert their effects by modifying sodium channels properties (37). Gilles et al. (38) elucidated, for the first time, how different toxins affect mammalian central and peripheral excitable cells. They interact selectively with sodium channel subtypes in a discrete subcellular region. The study by Gilles et al. (39) revealed unexpected subtype specificity of toxins that interact with receptor site 3 (α-type toxins). Additionally, it was found that multiple sodium channel subtypes in mammalian brain can be pharmacologically distinguished by their sensitivity into certain toxins, such as ScαTxs (scorpion α-toxins) and αLTxs (spider α-latrotoxins) (39).
Scorpion α-toxins provide a unique instrument for the identification of sodium channel subtypes due to their affinity for different channel subtypes (47). Thus, TsTX may be