FIGURE 3 Rhythmic, spindle-shaped discharges induced by a low systemic dose of pentylenetetrazol (PTZ) in a Wistar rat. These discharges with a crescendo-decrescendo pattern were associated with freezing behavior (motionless stare). Electrocorticograms from RF, right frontal (sensorimo-tor) cortex; RO, right occipital (visual) cortex; LF, left frontal cortex; LO, left occipital cortex.
zures. Twitches and tonic-clonic seizures are recorded throughout development. However, there is limited occurrence of freezing behavior and clonic seizures during the first 2 postnatal weeks of the rat. PTZ-induced seizures can be easily scored according to the appropriate scoring table (see Chapter 48). Graded doses of PTZ in adult rats will induce specific seizures. Thus low doses can be titrated to induce only freezing with underlying EEG spindles (Figure 3), somewhat higher doses for kindling, even higher doses for clonic seizures, and finally doses over 100 mg/kg for tonic-clonic seizures. Using a constant PTZ dose, latency to onset of seizures is also age specific (de Casrilevitz et al., 1971; Velisek et al., 1992; Vernadakis and Woodbury, 1969a; Weller and Mostofsky, 1995). The whole spectrum of EEG changes can be observed after PTZ
administration (Schickerová et al., 1984; Zouhar et al., 1980); however these changes are developmentally regulated: EEG spindles occur from the third postnatal week and beyond (Marescaux et al., 1984; Ono et al., 1990; Schickerová et al., 1984). Metabolic [14C]2-deoxyglucose (2DG) studies demonstrate that there is an increased uptake (ergo metabolic activation) in the motor and limbic cortex after PTZ-induced status epilepticus during the first postnatal week and in the brainstem areas at postnatal day (PN)10. In PN17, PN21, and adult rats, there is a redistribution of glucose uptake from the cortex and hippocampus to the mid-brain, brainstem, hypothalamus, and septum (Ben-Ari et al., 1981a; Nehlig et al., 1992; Pereira de Vasconcelos et al.,
1992). In young rats (<PN21), after subconvulsive doses of PTZ, c-fos was expressed in the medial thalamus, cortex, and globus pallidus. Occurrence of clonic seizures induced c-fos immunoreactivity in the cortex, thalamus, hypothalamus, and brainstem. In rats older than PN21, c-fos labeling appeared in the thalamus and hypothalamus after subcon-vulsive PTZ doses. After clonic seizures, c-fos was found also in the striatum, nucleus accumbens, brainstem, and hippocampus (Andre et al., 1998). Neuronal injury was observed in immature (PN10) rats 24 hours after SE in the hippocampus, amygdale, and cerebral cortex. However, this injury was only transient and did not result in neuronal death (Pineau et al., 1999).
Freezing behavior with underlying EEG spindles cannot be easily assessed without EEG recordings (Ono et al., 1990; Schickerová et al., 1984). Clonic seizures cannot be reliably induced during the first 2 postnatal weeks (Velísek et al., 1992).
Although PTZ exerts is action mostly via the t-butyl-bicyclo-phosphorothionate (TBPS) site of the GABAa receptor (Olsen, 1981), additional actions of PTZ also have been reported, for example, nonspecific membrane effects (Swinyard et al., 1989). Ethosuximide (ETX), clonazepam, and valproic acid suppress EEG spindles (Brabcová et al.,
1993) and tonic-clonic seizures (Mares et al., 1981). Suppression of PTZ-induced EEG spindles has, therefore, substantial predictive value for discovery of antiabsence drugs, although PTZ-induced clonic seizures are routinely used for this screeening. Classic anticonvulsants (such as carba-mazepine and phenytoin (Mares et al., 1983)) and NMDA receptor antagonists diminish tonic-clonic seizures (Velíssek et al., 1990, 1991, 1997).
Bicuculline (Ludolph, 2000b) cannot be easily dissolved in water or saline. A weak acid (0.1 N HCl) is recommended as a solvent, followed by careful titration using a weak base (0.1N NaOH) up to the resulting pH of approximately 5.6 (de Feo et al., 1985; Velisek et al., 1995). Because of the low final pH, it is recommended that bicuculline solution be administered IP because the sub-Q injection is very painful and stressful. However, IP injection of bicuculline in mature rats may require significantly higher doses than young rats require because of the "first-pass" effect. This disadvantage can be avoided by using IV administration of bicuculline (Zouhar et al., 1989). For IP administration, 2- to 4-mg/kg doses are used for developing rats; 6- to 8-mg/kg doses are used for prepubertal and adult rats (Veliskova et al., 1990). However, if administered IV, a dose of 2mg/kg is sufficient for the adult rats (Zouhar et al., 1989). After these doses, seizures occur within 20 minutes. An additional bicuculline model is worth mentioning. A derivative, bicuculline methiodide, is almost freely soluble in saline. However, this compound does not cross the blood-brain barrier. Nevertheless, in developing rats, in which the blood-brain barrier is still imperfect, systemic administration of bicu-culline methiodide induces seizures. Required doses are between 2 and 20mg/kg IP for rats up to PN18 (Mares et al., 2000).
Bicuculline produces all behavioral phenomena described previously. The EEG pattern also fits the description (Figure 4). The only difference compared with other drugs is that clonic seizures induced by bicuculline begin to occur during the second postnatal week of the rat. Metabolic studies with 2DG identified changes after bicuculline seizures similar to those seen after PTZ seizures (Ben-Ari et al., 1981a). A study using 2DG in paralyzed, ventilated rats, however, did not show any increases in cerebral glucose metabolic rate (Evans and Meldrum, 1984).
Adult rat, 12 min after 6 mg/kg bicuculline i.p.
FIGURE 4 Example of several episodes of rhythmic, spindle-shaped discharges induced by intraperitoneal administration of bicuculline in a Wistar rat. Discharges were generalized over all cortical recording. Arrows mark these discharges in RF cortical area recording. Electrocorticograms from the RF, right frontal (sensorimotor) cortex; RO, right occipital (visual) cortex; LF, left frontal cortex; LO, left occipital cortex.
The difficulties in dissolving and the acidic pH of the solution, along with the significant first-pass effect somewhat limit the usefulness of systemically administered bicu-culline for seizure models. EEG spindles can vary not only as a function of age but also as a function of sex (Matejovska et al., 1998).
The mechanisms of action of bicuculine are well understood (competitive antagonism at GABA binding site of the GABAA receptor), although additional effects of bicuculline, such as prolongation of Ca2+ action potential and blockade of K+ channels, have been reported (Seutin and Johnson, 1999). Classic anticonvulsant drugs have good effects against bicuculline-induced seizures with superb activity of GABAA receptor-acting drugs, such as benzodiazepines (De Deyn et al., 1992). Except for the well-known mechanism of action, there is no other advantage to using bicuculline over PTZ for screening for anticonvulsant drugs.
In contrast to bicuculline, picrotoxin (Ludolph and Spencer, 2000) can be dissolved in saline (Veliskova et al., 1990, 1993), although some groups prefer dissolving in 10% dimethylsulfoxide (Hiscock et al., 1996). Route of administration can be sub-Q, IP, or IV. Doses for IP bolus administration range between 3 and 6mg/kg through all developmental stages of the rat. Development of seizures is slower than in PTZ and bicuculline. Seizures usually occur within 30 to 40 minutes. Continuous infusion of picrotoxin solution in the jugular vein has been used to characterize regional EEG patterns and the EEG power spectra (Mackenzie et al., 2002). The addition of a small dose of picrotoxin to pilocarpine has been used to intensify development of pilocarpine-induced SE, thus helping to decrease the pilo-carpine dose (Hamani and Mello, 1997, 2002).
Behavioral and EEG patterns after picrotoxin resemble to those found after PTZ and bicuculline administration (Mackenzie et al., 2002). After a single picrotoxin seizure, early gene c-fos expression was found in the frontal and parietal cortex as well as in the piriform and entorhinal cortex. Neurons with c-fos expression were mostly calbindin D-28K-positive cells and many of the calcium-binding, protein-unlabeled neurons, which were spiny and presumed excitatory, (Hiscock et al., 1996).
Picrotoxin seizures develop at a slower pace than PTZ and bicuculline seizures, and they also occur less reliably
(Velisek et al., 1995). However, the very specific mechanism of action (even compared with bicuculline) and the very good solubility provide a small advantage over bicuculline-induced seizures.
Picrotoxin seizures are well defined mechanistically. They arise from GABAA receptor chloride channel blockade. Available data on anticonvulsant drugs suggest a similar efficacy as against bicuculline-induced seizures (Veliskova et al., 1990; De Deyn et al., 1992; Veliskova et al., 1993). The usefulness for screening of putative antiepileptic drugs is similar as in bicuculline-induced seizures.
Glutamic Acid Decarboxylase (GAD) Inhibitors
The GAD inhibitors (isonicotinehydrazide, 3-mercapto-propionic acid, and allylglycine), which inhibit GABA synthesis, can be potent convulsants. Isonicotinehydrazide can be dissolved in saline in the concentration of 100mg/ml. Doses used for IP induction of seizures range between 150 and 400 mg/kg throughout development in the rat (Maress and Trojan, 1991). Seizures usually occur within 40 to 60 minutes. 3-Mercaptopropionic acid is a liquid and doses between 30 and 60mg/kg can be used for IP seizure induction throughout development (Mares et al., 1993). In this model, seizures occur within 10 to 15 minutes at all developmental stages except for the first postnatal week (within 25 to 40 minutes). Finally, allyglycine can be dissolved in saline; and 100 to 250mg/kg can produce seizures. The solution should be used within 2 hours, though (Horton and Meldrum, 1973; Ashton and Wauquier, 1979, 1981; de Feo et al., 1985; Thomas and Yang, 1991). After this dose, seizures develop between 60 and 80 minutes after administration.
All drugs invariably induce a sequence of myoclonic twitches as well as clonic and tonic-clonic seizures. In the isonicotinehydrazide model, in younger rats, tonic-clonic seizures continuously follow after clonic seizures (Maress and Trojan, 1991). In the EEG, all drugs induce spikes (sharp waves in immature animals), polyspikes, and spike and wave patterns. In the 3-mercaptopropionic acid model, the correlation between behavioral seizures and EEG (recorded in the neocortex) is poor even in adult rats (Maress et al., 1993). Tonic extension may be pronounced in seizures induced with allylglycine (Thomas and Yang, 1991). During L-allylglycine seizures, there is significantly increased glucose cerebral metabolic rate determined by the 2DG method in the cortex, caudate/putamen, amygdala, hippocampus, thalamus, and inferior colliculus (Evans and
Meldrum, 1984); however, this study was carried out on paralyzed and ventilated rats.
The slow development of isonicotinehydrazide and allylglycine-induced seizures may become inconvenient. Reliability of seizure production is very good for all three drugs.
Figure 5 shows a simplified scheme of GABA synthesis and the effects of main GAD inhibitors. Isonicotine-hydrazide (Schröder, 2000) acts most likely via inhibition of the GAD coenzyme pyridoxine (Williams and Bain, 1961; Woodbury, 1980b). 3-Mercaptopropionic acid (Spencer, 2000b) and allylglycine (Ludolph, 2000a) are direct inhibitors of GAD (Lamar, 1970; Horton and Meldrum, 1973; Loscher, 1979; Meldrum et al., 1987). As a result, GABA production is decreased and both GABAa and GABAb receptors are less activated. Use of these models for screening purposes does not provide advantages other than the known mechanisms of action at a particular stage of GABA synthesis.
Was this article helpful?