Time course of changes in the concentration of kynurenic acid in the brain of pentylenetetrazol-kindled rats
The time response of changes in the brain concentration of kynurenic acid (KYNA) was examined in rats subjected to the pentylenetetrazol (PTZ)-induced kindling of seizures (n = 32). The development of seizures was accompanied by a progressive decrease in KYNA concentration in the caudate putamen, entorhinal cortex, piriform cortex, amygdala and hippocampus. A single injection of PTZ (35 mg/kg i.p.—the dose used in the kindling experiment, n = 7) caused a much less pronounced KYNA depletion, with different structures affected: the nucleus accumbens, piriform cortex and amygdala. The comparison of KYNA con- centration in rats subjected to the kindling of seizures with that in animals given a single, proconvulsive, dose of PTZ (55 mg/kg, n = 7) showed that the kindling itself, rather than the occurrence of a fit of seizures, was responsible for the depletion of KYNA in the hippocampus and caudate putamen. Another control experiment showed that neither single nor repeated saline injections caused significant changes in KYNA concentration. The data indicate that changes in the brain concentration of an endogenous inhibitory neu- rotransmitter, KYNA, undergo selective modulation in the course of a kindling of seizures. This suggests that the depletion of KYNA within the hippocampus may be directly related to the development of kindled seizures in this model of epilepsy.
1. Introduction
The kindling of seizures is one of the most widely used animal models of epilepsy. It models a process of the progressive decrease of the threshold of seizures, leading to tonic–clonic convulsions. A kindling of seizures can be elicited by repeated electric stimulation of certain brain areas, such as the amygdala or the hippocampus, or by the repetitive administration of initially subthreshold doses of a proconvulsive agent [27].
It is well known that epileptic seizures can be produced by an imbalance between the processes of neuronal excitation and inhi- bition, wherein aminoacids, monoamines and neuropeptides play important roles [39,40,45]. In this context, it has been hypothe- sized that the phenomenon of epileptogenesis leading to seizures may be the consequence of an overactivation of excitatory systems, or a decrease in activity of inhibitory pathways [25,27].
Kynurenic acid (KYNA) is thought to be one of the most impor- tant endogenous inhibitory neuroactive metabolite [41,42]. It is one of the tryptophan metabolites produced by astrocytes and neurons via the kynurenine pathway [19]. In the central nervous system, kynurenic acid is synthesized through the transamina- tion of its precursor, L-kynurenine, by two enzymes: kynurenine aminotransferases (KATs) I and II. These enzymes are found in glial and neuronal cells [8,10,16,28,29]. At the high, non-physiological concentration, KYNA is a competitive antagonist at NMDA, AMPA and kainate receptors, but at lower concentrations KYNA acts as an antagonist at the glycine-binding site of the NMDA receptor [8,18,30,38]. KYNA can also antagonize, in a non-competitive way, the activity of presynaptic α7 nicotinic acetylcholine receptors [12]. The activation of NMDA receptors appears to play a major role in neuronal development, long-term potentiation, learning and mem- ory. Enhanced glutamatergic transmission, mediated through both the NMDA and other glutamatergic ionotropic receptors, has been suggested to significantly contribute to the kindling phenomenon [25]. It is well known that the activation of glutamate receptors is necessary for the initiation and spreading of epileptiform activ- ity, and that their antagonists are effective anticonvulsants. Many glutamate antagonist agents tested to date have been proved to antagonize seizures evoked by a variety of stimuli [25]. There are data indicating that KYNA and GABA belong to the most important inhibitory neuroactive metabolites, controlling the kindling and fits of seizures [43].
There is also some evidence that the brain concentration of KYNA is altered in epilepsy, though the nature and time response of changes in the local concentration of KYNA in various brain structures in the course of kindling is not well recognized [11,14]. Particularly important is the question about the role of changes occurring in the cortical and limbic brain structures. Taking into account the important role of the hippocampal formation, amyg- dala, caudate putamen, entorhinal, prefrontal and piriform cortex in the process of kindling, we have decided to examine changes in the KYNA levels in all these brain structures at various stages of pentylenetetrazole-induced kindling of seizures. Pentylenete- trazole (PTZ) is a non-competitive GABA receptor antagonist, characterized by an extremely short latency of action, due to the high bioavailability, and rapid distribution to all organs [33]. The model of the PTZ-induced kindling of seizures is commonly used to study the pathomechanism of epilepsy [9,27].
2. Materials and methods
2.1. Animals
Adult male Wistar rats, 12 weeks old, and weighing 200 ± 20 g at the beginning of the experiment, were used in the study. The total number of animals was 81. The rats were housed two per cage in standard laboratory conditions under 12 h cycle (lights on at 7:00 a.m.) in a controlled temperature (20 ± 4 ◦C) and 70% humidity. The rats were given free access to food and water. The experiments were performed between 9:00 a.m. and 3:00 p.m. The study was carried out in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC), and approved by the Committee for Animal Care and Use at the Medical University in Warsaw.
2.2. Drugs
In the experiment, pentylenetetrazole (PTZ) (Sigma–Aldrich, Poland) was used.PTZ was dissolved in 0.9% NaCl solution.
2.3. Kindling procedure
The animals received repeated injections of PTZ (kindled rats) or saline (con- trol rats). The drug was given in a volume of 1 ml/kg of saline. PTZ was injected intraperitoneally at a subconvulsive dose of 35 mg/kg, three times a week (Monday, Wednesday, and Friday). After each injection, the rats were placed singly in Plexi- glass cages (30 cm × 30 cm × 50 cm) and were observed for 30 min. The intensity of seizures was classified according to the Racine’s [32] scale: stage 1 – immobility, eye closed, facial clonus; stage 2 – head nodding, more severe facial clonus; stage 3 – clonus of one forelimb; stage 4 – rearing, with bilateral forelimb clonus; stage 5 – generalized clonic–tonic seizures [32]. Control rats received injections of saline and were kept isolated in the same cages as PTZ-kindled rats for 30 min. Kindling stimulation was continued to the various stages of seizure: a single PTZ dose, stages 1 and 2 seizures, stage 3, stage 4, or two consecutive stage 5 seizures. 1.5 h after the appropriate last injection of PTZ, the kindled animals and control animals (with the same number of saline injection) were sacrificed, their brains were removed, frozen at −70 ◦C and cut into slices. The following anatomical structures were iso- lated to analyze of KYNA level: prefrontal cortex, piriform cortex, entorhinal cortex, amygdala, hippocampus, caudate putamen, and nucleus accumbens. In this way the changes in KYNA concentration in particular brain structures could be observed at different stages of a development of kindled seizures. In order to better control the influence of unspecific factors on the KYNA levels, and to properly control the influence of kindling alone on KYNA, the following additional, control, groups were examined: naive animals without any injection, animals after a single injection of saline, animals after a single injection of 35 mg/kg PTZ (i.e. the dose used for kindling procedure), a group of kindled animals after a single injection of 55 mg/kg dose of PTZ (i.e. the dose inducing acutely a fit of seizures (tonic–clonic convulsions)).
2.4. Determination of kynurenic acid in homogenates
For determination of kynurenic acid (KYNA), each tissue sample was weighted, placed in a dry-cooled polypropylene vial, and homogenized in 20 volumes of an ice-cold 2% perchloric acid (30 s, 4 ◦C). The homogenates were then centrifuged at 26,880 × g for 8 min at 4 ◦ C. After centrifugation, the supernatants were collected and filtered through 0.45 µm filter (Millipore). Samples were immediately frozen and kept at −70 ◦C until assay.
Kynurenic acid was measured according to the modified method of Wu et al. [47]. Determination of KYNA was performed using a high-pressure liquid chromatog- raphy (HPLC) with fluorescence detection. The HPLC system used for analysis of kynurenic acid consisted of the following elements: a pump (Shimadzu, LC-10AD VP), fluorescence detector (Shimadzu, RF-10 XL) set at an excitation wavelength of 344 nm and an emission wavelength of 398 nm. Samples were injected manually via Rheodyne 7725i injection valve with a 20 µl sample loop. KYNA was separated on Phenomenex Luna C18 (150 mm × 3 mm) column, with Phenomenex KJO-4286 precolumn at a flow rate of 0.4 ml/min, run at room temperature.
The mobile phase (isocratic system) consisted of 50 mM sodium acetate, 250 mM zinc acetate and 4% acetonitrile (adjusted with acetic acid to the pH 6.2). The mobile phase was degassed for 15 min. Chromatogram registration and analysis was done using ChromaX 2004 software. The retention time of KYNA under these conditions was ∼10.5 min, and the sensitivity limit of HPLC was 0.1 nM. The fluorescence
method was evaluated by injection of a standard of kynurenic acid, in concentrations from 0.5 to 100 nM. The concentration of KYNA was calculated as pmol/g of tissue.
2.5. Statistics
Differences between KYNA levels in animals on different stages of seizures, and the effects of different doses of PTZ were tested for statistical significance using the Kruskal–Wallis analysis of variance by rank followed by the Mann–Whitney U-test. Mann–Whitney U-test was used when the effect of a single saline injection was com- pared to the naive animals without injection, and animals chronically treated with vehicle. The confidence limit of p < 0.05 was considered as statistically significant. 3. Results 3.1. Effects of PTZ-induced kindling of seizures on the concentrations of KYNA in brain structures Kruskal–Wallis analysis of variance by rank displayed a signif- icant effect of the kindling stage on KYNA concentration in the caudate putamen (H = 19.43; p < 0.01), entorhinal cortex (H = 12.84; p < 0.05), prefrontal cortex (H = 16.52; p < 0.01), piriform cortex (H = 16.58; p < 0.01), amygdala (H = 10.05; p < 0.05) and hippocam- pus (H = 12.73; p < 0.05) (Fig. 1). Mann–Whitney U-test, revealed that at the stages 1–2 of seizures there appeared an increase in KYNA concentration in the piriform cortex (U = 93.0; p < 0.05). At the stage 3 a significant decrease in KYNA concentration was found in the prefrontal cor- tex only (U = 51.0; p < 0.05). In the animals with stage 4 of seizures a decreased concentration of KYNA was present in the caudate puta- men (U = 16.0; p < 0.05). Most prominent changes in KYNA levels were found in animals with stage 5 of seizures, i.e. a decreased concentration of KYNA in the: caudate putamen (U = 9.0; p < 0.01), entorhinal cortex (U = 17.0; p < 0.05), prefrontal cortex (U = 16.0; p < 0.05), piriform cortex (U = 10.0; p < 0.01), amygdala (U = 11.0; p < 0.01) and hippocampus (U = 7.0; p < 0.01). 3.2. Control analysis The influence of injection protocol on the local concentration of KYNA was also examined. Mann–Whitney U-test revealed no significant changes in KYNA concentration in all examined brain structures after single saline administration (p > 0.05), in compari- son to the naive control group, without injection (Table 1). Similarly, concentrations of KYNA did not differ after single or repeated injec- tions of saline (p > 0.05) (Table 2).
To check for the influence of PTZ alone on KYNA concentra- tion, we have performed the comparison between the effects of PTZ administered at the dose of 35 mg/kg, i.e. the dose used in the course of kindling, and at the dose of 55 mg/kg, i.e. the dose sufficient to evoke acutely the fit of tonic–clonic seizures, vs. control animals, i.e. a single saline injected group (Fig. 2). Kruskal–Wallis analysis of variance by rank revealed a significant effect of PTZ injections in the nucleus accumbens (H = 8.81; p < 0.05), piriform cortex (H = 7.79; p < 0.05), and amygdala (H = 6.70; p < 0.05). Mann–Whitney U-test showed a significant reduction of KYNA concentration after injection of 35 mg/kg PTZ in the nucleus accumbens and piriform cortex (U = 5.0; p < 0.05 and U = 3.0; p < 0.05, respectively). A single injection of pentylentetrazole at a dose of 55 mg/kg significantly decreased KYNA concentration in the nucleus accumbens (U = 0.0; p < 0.01), piriform cortex (U = 2.0; p < 0.05), and amygdala (U = 1.0; p < 0.01) (Fig. 2). In order to check for the effects of seizures alone on KYNA concentration we have compared the group of animals which reached stage 5 of seizures (after a final injection of 35 mg/kg PTZ), with the group of animals after a single injection of PTZ (at the dose 35 mg/kg), and with the group of animals after a single injection of PTZ (at the dose of 55 mg/kg, evoking acute seizures). Kruskal–Wallis analysis, showed significant differences among studied groups in the caudate putamen (H = 9.11; p < 0.05), and in the hippocampus (H = 11.2; p < 0.01). After administration of PTZ in doses of 35 and 55 mg/kg a higher KYNA concentration was observed in the caudate putamen (U = 1.0; p < 0.01 and U = 5.0; p < 0.05, respectively), and in the hippocampus (U = 1.0; p < 0.01 and U = 0.0; p < 0.01, respectively), in comparison to kindled animals with the stage 5 of seizures (Fig. 3). 4. Discussion Our study demonstrated a fluctuation of KYNA concentrations in the course of the kindling of seizures. The primary finding was that the changes in KYNA levels during kindling were structure- dependent and correlated with the stage of seizures. Biochemical analysis indicated that, in animals with stages 1 and 2 seizures, kindling selectively elicited an increase in KYNA levels in the piriform cortex. Subsequently, during the course of the kindling, there was a gradual decrease in KYNA concentrations, with a maximum drop at stage 5 seizures. This phenomenon was only found in the piriform cortex. Taking into consideration that the piriform cortex is among the structures activated early during the development of kindling, and the fact that KYNA is a NMDA recep- tor antagonist, its elevation during the early stages of kindling may represent a defense mechanism activated to prevent the excessive neuronal activation by glutamate. These results also indicate that the piriform cortex may play a particular role in the development of PTZ-induced kindling [4,7,13,23,36]. On the other hand, at stage 5 seizures a significant reduction of KYNA levels was observed in the caudate putamen, piriform cortex, hippocampus and amygdala when the seizures were fully developed. Our study indicates that PTZ kindling, particularly at stage 5 seizures, causes a general reduction in KYNA concentration. KYNA production is modulated by several distinct factors and mecha- nisms. Its synthesis, for example, can be significantly reduced in response to hypoglycemic or depolarizing conditions [8]. Taking into consideration that KYNA concentration was decreased after exposing the animals to the depolarizing agent (PTZ), or glutamate, a significant reduction of KYNA levels observed in this study might be due to such a depolarizing effect [46]. It is possible that the depletion of KYNA could subsequently facilitate the spread of the depolarization wave throughout the brain [34]. Accordingly, Kamin´ ski et al. suggested that the selective deficit of endogenous KYNA in WAG/Rij rats, a genetic model of absence epilepsy, may account for increased excitability to the depolarizing stimuli [14]. Similarly in humans, KYNA levels were found to be decreased in the absence epilepsy, and infantile spasms (West syndrome), as well as in the neurodegenerative disorders such as Parkinson’s disease [31,38,49,50]. However, this decrease in KYNA concen- tration is not a universal phenomenon in neurodegenerative disorders. For example, in Alzheimer’s disease, Down’s syndrome, or schizophrenia, brain KYNA concentrations have been shown to be significantly elevated [1,2,6,38]. The role of KYNA in the process of epileptogenesis is still a matter of debate. For example, some authors have observed an increase in KYNA levels in the brain structures of electrically kindled animals. The concentration of kynurenic acid was raised in the hippocam- pus of kindled rats, probably as an adaptive reaction, counteracting the kindling-induced over excitability of neurons [48]. However, this discrepancy may be a consequence of different animal models used (electrical vs. chemical), different stages of seizures, and differ- ent experimental paradigms (analysis in vivo vs. in vitro). Moreover, in the paper by Löscher et al., besides nucleus accumbens, no sig- nificant changes were observed in other brain structures analyzed, such as hippocampus, cortex, striatum and cerebellum [24]. All this suggests that changes in KYNA levels in the course of kindling are complex and depend on many different variables. A growing body of evidence suggests that KYNA is an important endogenous neuroprotective agent preventing neuronal loss following excitotoxic or ischemia-induced neuronal injuries [3,35]. The mechanism of KYNA’s neuroprotective potency arises from its antagonist actions at the glutamate receptors in the brain regions sensitive to neuronal damage and characterized by enhanced NMDA receptor excitability (e.g. hippocampus). Besides the block- ade of postsynaptic NMDA receptors, KYNA-related antagonism of presynaptic nicotinic acetylcholine receptors may also contribute to the inhibitory effects on glutamate release [12]. Thus, the reduction of KYNA concentrations in kindled animals observed in this study may be considered a primary reason of exac- erbated vulnerability to the excitotoxic or ischemic insults [37]. In other words, it appears that changes in KYNA levels may be a consequence of the process of shifting the balance between the inhibitory and excitatory systems, represented by GABA and glu- tamate, in favor of the latter. The contribution of GABA systems to the control of epileptiform-like firing of neurons is well recog- nized [27]. It is also known that the function of the GABA system can be profoundly modified by seizures. The available experimen- tal evidence indicates a variety of seizure-induced alterations in the function of the GABAergic system; both stimulation and inhibition of its activity in response to kindling has been reported [15]. It is conceivable that a kindling-induced, transient increase in the inhibitory processes (increased KYNA levels) observed in the piriform cortex, may reflect a compensative reaction to the exces- sive neuronal activation. Such a reaction was also observed in the activity of the GABA system. For example, GABA release was increased in the hippocampal and amygdala slices in the kindled rats [15,17,22]. Hippocampal microdialysis studies demonstrated an elevation of extracellular concentrations of GABA during kindled seizures [5,26,44]. However, it is also conceivable that an enhance- ment in the GABA release may not be sufficient to protect against the increased glutamate release during the kindling of seizures. At the same time, changes in GABA activity may interfere with KYNA transmission. It was found that an enhanced GABA concentration dose-dependently diminished KYNA synthesis in the brain cortical slices [20,21]. These data suggest that GABA and GABA agonists can modulate KYNA synthesis and release. To check the unspecific effect of fits of seizures on KYNA levels, we have performed a comparison between animals with stage 5 seizures with animals after a single administration of 35 mg/kg of PTZ, and with animals after an acute fit of seizures evoked by a sin- gle injection of 55 mg/kg of PTZ. It was found that in almost all the analyzed structures the KYNA levels were diminished in fully kin- dled (stage 5) animals, in comparison with rats with a single injec- tion of PTZ, regardless of the dose (though the statistically signifi- cant level was achieved in the caudate putamen and hippocampus only, Fig. 3). These results indicate that kindling-induced changes in KYNA concentrations are not directly related to the biochemi- cal effect evoked by a single episode of tonic–clonic convulsions. Another control experiment showed that a single injection of PTZ, at the dose of 35 or 55 mg/kg, caused a different pattern of KYNA depletion in the brain structures. Single injection of PTZ produced a dose-dependent decrease in KYNA concentrations in the nucleus accumbens and piriform cortex, in comparison to rats receiving a single injection of saline. In the amygdala, only the dose of 55 mg/kg of PTZ decreased the KYNA concentration (Fig. 2). However in the case of the PTZ-induced kindling of seizures, the most potent KYNA depletion appeared in the caudate putamen, piriform cortex, entorhinal cortex, hippocampus, prefrontal cortex and amygdala. These changes were, therefore, much deeper and concerned brain structures more directly linked to the process of epileptogenesis (Fig. 1). Additional control experiments showed also that a single or repeated injections of saline did not change the concentration of KYNA in the examined brain structures (Tables 1 and 2). The present data, however, do not explain the mechanism of alterations in KYNA levels (the role of modification in its synthesis, degradation or release). Further studies of changes in the activity of kynurenine converting or KYNA degrading enzymes are needed to clarify the issue of an intrinsic mechanism of the reported phenomenon. Summing up, this study underlines the role of KYNA in the process of the PTZ-induced kindling of seizures, especially in the neuronal circuits more prone to depolarization (e.g. the hippocam- pus and entorhinal cortex). The results add more arguments to the accumulating evidence suggesting an important contribution of KYNA to the epileptogenesis, and the possible role of KYNA receptor ligands in the control of seizures.