Effects of Acute and Repeated Zolpidem Treatment on Pentylenetetrazole-Induced Seizure Threshold and on Locomotor Activity: Comparison with Diazepam

ARTICLE IN PRESS

Neuropharmacology xxx (2009) 1–7

Contents lists available at ScienceDirect

Neuropharmacology

journal homepage: www.elsevier.com/locate/neuropharm

Effects of acute and repeated zolpidem treatment on pentylenetetrazole-induced seizure threshold and on locomotor activity: Comparison with diazepam

Josipa Vlainic´, Danka Pericˇic´ *

Laboratory for Molecular Neuropharmacology, Division of Molecular Medicine, RuCer Bosˇkovic´ Institute, P.O. Box 180, 10002 Zagreb, Croatia article info abstract

Article history: Zolpidem and diazepam are widely used drugs acting via benzodiazepine binding sites on GABAA Received 31 July 2008 receptors. While diazepam is non-selective, zolpidem has a high afﬁnity for a1-, and no afﬁnity for a5- Received in revised form containing receptors. Several studies suggested that behavioral effects of zolpidem might be more similar 20 March 2009 to classical benzodiazepines than previously thought. To compare the sedative and anticonvulsant Accepted 23 March 2009 properties of these drugs and to evaluate the importance of GABAA receptor subunits for development of tolerance during chronic treatment, we tested the effects of acute and repeated administration of zol- Keywords: pidem and diazepam on ambulatory locomotor activity (a measure of sedation) and on the threshold for Zolpidem Diazepam myoclonic, clonic and tonic seizures in response to i.v. infusion of pentylenetetrazole (PTZ). Both drugs Pentylenetetrazole given acutely in doses 0.3, 1 and 3 mg/kg reduced locomotion, and in doses 1 and 3 mg/kg elevated the Seizure threshold threshold for PTZ-induced seizures. The effects of zolpidem and diazepam on the tonic seizure threshold Sedative activity were greater than on myoclonus and clonic seizure threshold. Diazepam and zolpidem (3 mg/kg), given Tolerance 18 or 42 h after repeated drug treatment (10 days, 5 mg/kg, twice daily), decreased the PTZ seizure threshold and increased the locomotor activity as compared to control mice, indicating development of tolerance to their anticonvulsant and sedative effects. After repeated treatment the PTZ seizure threshold was not different between the two drugs, while differences in sedation became larger than after the acute treatment. The results suggest that a5-containing GABAA receptors are not crucial for the development of sedative and anticonvulsant tolerance. Ó 2009 Published by Elsevier Ltd.

1. Introduction GABAA receptors (Arbilla et al., 1986; Graham et al., 1996; Atack et al., 1999; Crestani et al., 2000). Furthermore, unlike classical The imidazopyridine zolpidem is a widely used hypnotic drug benzodiazepines, which have a similar affinity and agonist effi- known to produce its effects via benzodiazepine binding sites on cacy for a1-3 and a5-containing GABAA receptors, zolpidem the GABAA receptors (Depoortere et al., 1986; Sanger, 2004). shows no affinity for receptors containing the a5 subunit Although zolpidem and benzodiazepines use the same binding (Pritchett and Seeburg, 1990; Graham et al., 1996; Atack et al., sites, the pharmacological profile of zolpidem is different from 1999). A recent study demonstrated that binding of zolpidem that of classical benzodiazepines. Unlike benzodiazepines, zolpi- requires a phenylalanine residue (Phe77) in the g2subunit(Wulff dem has very pronounced sedative and mild anxiolytic, anticon- et al., 2007). Crestani et al. (2000) suggested that the sedative vulsant and myorelaxant properties (Depoortere et al., 1986; action of zolpidem and its activity against pentylenetetrazole Sanger, 2004). Several groups of authors reported that unlike (PTZ)-induced tonic seizures are completely mediated by benzodiazepines, zolpidem does not produce either sedative a1-containing GABAA receptors. (Perrault et al., 1992; Scharf et al., 1994; Elliot and White, 2000)or Seizures induced by PTZ, a non-competitive antagonist of the anticonvulsant tolerance (Perrault et al., 1992) during its pro- GABAA receptors, are widely used in the screening of potential longed administration. Differences in the behavioral properties of antiepileptic drugs. They are considered to be a model of petit mal these positive allosteric modulators of GABAA receptor function or generalized absence seizures. PTZ seizures and human absence were explained by selective affinity of zolpidem for a1-containing seizures share behavioral and EEG similarity and are pharmaco- logically characterized by high anticonvulsant efficacy of trime- thadione, valproic acid, ethosuximide and benzodiazepines and by * Corresponding author. Tel.: þ385 1 4561126; fax: þ385 1 4561010. inefficacy of phenytoin and carbazepine (Lo¨scher and Schmidt, E-mail address: [email protected] (D. Pericˇic´). 1988; Snead, 1992; White, 2003). In rodents, these seizures are

0028-3908/$ – see front matter Ó 2009 Published by Elsevier Ltd. doi:10.1016/j.neuropharm.2009.03.010

Please cite this article in press as: Vlainic´, J., Pericˇic´, D., Effects of acute and repeated zolpidem treatment on pentylenetetrazole-induced seizure..., Neuropharmacology (2009), doi:10.1016/j.neuropharm.2009.03.010 ARTICLE IN PRESS

2 J. Vlainic´, D. Pericˇic´ / Neuropharmacology xxx (2009) 1–7 characterized by 7–9 Hz bilaterally synchronous spike-wave Activity Cage. The apparatus consists of an animal cage (with a transparent cover) discharges, which originate in thalamocortical circuitry (Snead, and an electronic unit. The activity detection relies on horizontal sensors, designed for the assessment of the ambulatory activity. The movements the animal makes 1995). According to Lo¨scher and Nolting (1991), one of the models inside the cage interrupt one or more infrared beam/s. The beam interruptions are that should be used in the laboratory evaluation of anticonvulsant counted and recorded by the electronic unit. Data related to horizontal ambulatory drugs is the i.v. PTZ seizure threshold model to avoid underesti- activity are printed in digital form at pre-set intervals. The activity was recorded for mation of anticonvulsant activity of test compounds. 10 min, starting after placing the animal into the test cage. The locomotor The aim of this study was to determine and compare in the same measurements were performed between 08:00 and 13:00 h in a quiet room under normal laboratory lighting. The observer left the room after placing the animal in the experiment the acute doses of zolpidem and diazepam producing apparatus. sedation and increasing the threshold for PTZ-induced myoclonic, clonic and tonic seizures. 2.4. Seizure threshold determination Our aim was also to compare the ability of zolpidem and diazepam, given in moderate doses, to induce sedative and anticon- The animal was observed throughout the infusion, and the time between the start of PTZ infusion (4.4 mg/min) and the onset of seizures (latency) was recorded, vulsant tolerance after prolonged treatment. If, as suggested by Van mainly as previously described (Pericˇic´ et al., 2001, 2008). The ﬁrst convulsive sign Rijnsoever et al. (2004), a5-GABAA receptors are required for the was myoclonus (a sudden involuntary muscle jerk, usually accompanied by a head development of sedative tolerance to diazepam, then drugs lacking twitch). Clonic seizures were characterized by whole-body clonus, including afﬁnity for a5 receptor subtype (such as zolpidem) should not running and explosive jumps, while tonic seizures were characterized by extreme develop sedative tolerance. rigidity, with forelimbs and hindlimbs extended caudally. The infusion was stopped after tonic seizures. For each animal, the threshold dose of convulsant (mg/kg of Sedative activity was assessed by measuring ambulatory loco- body weight) required to elicit myoclonic, clonic and tonic seizures was calculated motor activity. Anticonvulsant activity was evaluated by determi- from the time of infusion (latency), infusion rate, concentration of convulsant and nation of threshold for myoclonic, clonic and tonic seizures body weight. All experiments were carried out between 8:00 and 13:00 h. following an i.v. infusion of PTZ. 2.5. Statistical analysis

2. Materials and methods Statistical analysis and graphic presentation of results were carried out using SPSS version 13 and GraphPad Prism version 4.00 for Windows. Locomotor activity 2.1. Animals data were subjected to two-way, and the seizure threshold data to three-way ANOVA with seizure type as the repeated measure. Male CBA/HZgr mice (20–25 g) approximately 3 months old, raised in the For all signiﬁcant effects, we reported the partial eta-squared (h2) values and the breeding colony of the RuCer Bosˇkovic´ Institute, Zagreb, Croatia, were used. They p statistical power (Cohen, 1962). The partial eta squared is the proportion of the were housed at a constant temperature (22 C) and under a light cycle of 12-h light/ effect þ error variance that is attributable to the effect. It is computed as follows: 12-h darkness (lights on at 7:00 a.m.). Food and water were freely available. In order partial h2 ¼ SS /(SS þ SS ), where SS is the sum of squares for the to improve the accuracy in calculating the dosage of drugs as well as PTZ (expressed p effect effect error effect effect, and SS is the sum of squares for the error term associated with that effect. per kg of body weight) required to induce seizures, the animals were deprived of error Partial eta-squared values can sum to more than 1 because each effect size estimate food the night before the experiment. All animal care and experimental procedures uses a different denominator. Statistical power is deﬁned as 1 b, where b is the were carried out in accordance with the National Institute of Health, Guide for the probability of correctly rejecting a false null hypothesis (Type II error). Type I error Care and Use of Laboratory Animals (NIH publication No. 80-23) revised 1996, and probability was set to 0.05. Post hoc comparisons were conducted using Newman– with the Croatian law on animal welfare. Keuls multiple comparison test.

2.2. Drugs 3. Results Zolpidem hemitartrate (kindly provided by Pliva, Zagreb, Croatia), diazepam (Apaurin ampullae, Krka, Novo mesto, Slovenia) and pentylenetetrazole (PTZ, Sigma, 3.1. Acute effects of zolpidem and diazepam on the threshold for St Louis, MO) were used. Zolpidem and PTZ were dissolved in, and diazepam was PTZ-induced myoclonic, clonic and tonic seizures diluted with saline. Zolpidem and diazepam were administered i.p. in a volume of 1 ml per 100 g body weight. Control animals received i.p. injection of saline. In the acute experiment, doses of drugs were 0.3, 1 and 3 mg/kg. They were given 15 min As shown in Fig. 1A–C, diazepam and zolpidem (0.3, 1 and 3 before placing the animals in Ugo-Basile Activity Cage for measurement of ambu- mg/kg) produced a dose-dependent increase of the threshold for latory locomotor activity, i.e., 30 min before the PTZ seizure threshold determination PTZ-induced seizures. The three-way ANOVA (dose drug started. For the acute experiment, the timing of zolpidem administration was chosen seizure type) with seizure type as repeated measure, revealed based on previous studies (e.g., Depoortere et al., 1986; Fahey et al., 2006). In the experiment designed to study the development of tolerance, mice were a significant (p ¼ 0.000) within-subject effect of seizure type treated twice daily (at 8:00 a.m. and 5:00 p.m.) for 10 consecutive days with diaz- (F2,119 ¼ 1273.7), effect of seizure type by dose (F6,119 ¼ 154.0) and by epam (5 mg/kg i.p.), zolpidem (5 mg/kg i.p.) or saline. On day 11 or 12, i.e., 18 or 42 h drug dose interaction (F6,119 ¼ 6.24). The seizure type and the after termination of repeated treatment, diazepam treated mice received an addi- mentioned interactions explained 91.5, 79.5 and 13.6% of the tional i.p. injection of diazepam (3 mg/kg), and zolpidem treated mice an additional i.p. injection of zolpidem (3 mg/kg). Half of the animals in saline treated group were within-subjects variance in PTZ-induced seizure threshold, as evi- 2 given for the first time diazepam (3 mg/kg i.p.), and the other half were given zol- denced by their partial eta-squared values (hp ¼ 0.915, 0.795 and pidem (3 mg/kg i.p.). In the experiment performed 18 h after termination of 0.136, respectively). The interaction drug seizure type was repeated treatment, one part of saline treated mice were given an additional i.p. insignificant (F2,119 ¼ 0.106). There was a significant (p ¼ 0.000) injection of saline. between-subject effect of drug (F 105.3), of dose PTZ was given by constant i.v. infusion. A butterfly infusion needle (length 1,119 ¼ 20 mm, gauge 27) was inserted into the tail vein and correct placement was verified (F3,119 ¼ 407.5) and drug dose interaction (F3,119 ¼ 36.08), sug- by the appearance of blood in the infusion tubing. During the infusion the animal gesting that differences in effects between the drugs were not was held lightly by the tip of the tail. The concentration of PTZ was 4 mg/ml and the consistent across different doses. The effects described above, infusion rate controlled by a microinfusion pump was 1.1 ml/min. accounted for 46.9, 91.1 and 47.6% of the variance in individual As in the two previous studies (Fahey et al., 2006; Pericˇic´ et al., 2008), sedative 2 and anticonvulsant effects of zolpidem against PTZ-induced seizures were studied in differences in PTZ-induced seizure threshold (hp ¼ 0.469, 0.911 and the same animal. Sedative activity was assessed by measuring the decreases in 0.476, respectively). The observed statistical power, computed ambulatory locomotor activity. using alpha ¼ 0.05, was in all cases 1.00, suggesting that there was a zero probability of committing Type II error. 2.3. Locomotor activity Post hoc analysis indicated that for all seizure types the seizure threshold obtained with 1 and 3 mg/kg was significantly different Prior to the experiment, the animals were brought from the adjacent animal room to the laboratory and allowed to acclimatize for approximately 1 h. The (p ¼ 0.01–0.001) from that obtained in the saline treated group. The horizontal ambulatory activity in individual mice was registered by Ugo-Basile lowest dose of diazepam that produced a mild elevation (p ¼ 0.05)

J. Vlainic´, D. Pericˇic´ / Neuropharmacology xxx (2009) 1–7 3

of the threshold for myoclonic and clonic, but not tonic seizures, was 0.3 mg/kg.

3.2. Acute effects of zolpidem and diazepam on locomotor activity in mice

To assess the sedative activity of zolpidem and diazepam, we tested the effects of these drugs on ambulatory locomotor activity. As shown in Fig. 2, both zolpidem and diazepam given in doses 0.3, 1 and 3 mg/kg reduced markedly the locomotor activity in mice, as quantified by recording interruptions of infrared beams during 10 min (counts/10 min). Two-way ANOVA (dose drug) revealed a significant main effect of dose (F3,119 ¼ 321.3, p ¼ 0.0001) and drug (F1, 119 ¼ 9.03, p ¼ 0. 003). The observed statistical power was 1.000 and 0.846, respectively, and the partial eta-square values indicated that the effect of drug accounted only for a small (7%) and the effect of dose for a large proportion (89%) of the total variance in locomotor activity. A non-significant dose drug interaction (F3, 119 ¼ 1.68) suggested similar differences between the two drugs across different doses.

3.3. Effects of repeated treatment with zolpidem and diazepam on the threshold for PTZ-induced myoclonic, clonic and tonic seizures in mice

Challenge doses of both zolpidem and diazepam, given 18 h after repeated treatment (i.e., the 11th day from the start of the experiment), decreased the threshold for PTZ-induced myoclonic, clonic and tonic seizures (Fig. 3A–C). The three-way ANOVA (treatment drug seizure type) with seizure type as repeated measure, revealed a significant main effect of treatment (F1,30 ¼ 23.88, p ¼ 0.000) which accounted for 44.3% differences in the seizure threshold (observed power ¼ 0.997), confirming the development of tolerance to the anticonvulsant effects of drugs. There was also a significant effect of seizure type (F2,30 ¼ 152.66, 2 p ¼ 0.000, hp ¼ 0.836, power ¼ 1.000) and a less convincing 2 treatment seizure type (F2,30 ¼ 6.99, p ¼ 0.013, hp ¼ 0.189, power ¼ 0.725) interaction. There was no significant effect of either drug (F1,30 ¼ 0.015), drug seizure type (F2,30 ¼ 2.07) or drug treatment (F1,30 ¼ 0.434) interaction. Post hoc analysis revealed significant differences (p ¼ 0.05–0.001) between repeated (10 days) and acute treatment, and between the acute effects of zolpidem and diazepam on myoclonic (p ¼ 0.001) and clonic (p ¼ 0.05) seizure threshold. In the same experiment, the seizure threshold for PTZ-induced seizures in the group of mice which received saline after repeated saline, as evidenced in the legend to Fig. 3, was similar to that obtained in the acute experiment. Challenge doses of both zolpidem and diazepam, given 42 h after repeated drug treatment (i.e., the 12th day from the start of treatment), also, in comparison to acute doses of drugs given 42 h after repeated saline, decreased the threshold for PTZ-induced myoclonic, clonic and tonic seizures (Fig. 3D–F). Three-way ANOVA with seizure type as repeated measure revealed a significant 2 (p ¼ 0.000) effect of treatment (F1,29 ¼ 29.04, hp ¼ 0.500, observed power ¼ 0.999) indicating again a development of tolerance to the anticonvulsant effects of drugs. There was also a significant effect of

Fig. 1. Acute effects of zolpidem and diazepam on the threshold for myoclonic (A), clonic (B) and tonic (C) seizures induced in mice by i.v. infusion of PTZ. Different doses of drugs were administered 30 min before the infusion of PTZ was started. To test the locomotor activity, the same animals spent 10 min (15–25 min following drug administration) in Ugo-Basile Activity Cage. Points represent means SEM from at least 8 animals per group. *p ¼ 0.05–0.01, **p ¼ 0.001 versus control; jp ¼ 0.01, jjp ¼ 0.001 versus the corresponding diazepam treated group (three-way ANOVA and Newman–Keuls test).

4 J. Vlainic´, D. Pericˇic´ / Neuropharmacology xxx (2009) 1–7

4. Discussion

It is generally accepted that unlike classical benzodiazepines, the imidazopyridine zolpidem has very pronounced sedative and mild anxiolytic and anticonvulsant properties (Depoortere et al., 1986; Sanger, 2004). Several studies (Perrault et al., 1992; Scharf et al., 1994; Elliot and White, 2000) reported a lack of tolerance to zolpidem upon repeated administration. This study demonstrated that zolpidem given acutely is slightly more effective than diazepam in producing sedation and less potent as a convulsant. Although both drugs given acutely were more potent against tonic than against myoclonic and clonic seizures, this difference was dependent on the dose and it was greater after zolpidem than after diazepam administration. After repeated treatment, mice developed tolerance to the anticonvulsant and sedative effects of both drugs so that differences in their anticonvulsant activity were lost, while differences in the level of sedation were potentiated. Diazepam is known to be non-selective for GABAA receptor subtypes, while zolpidem has the highest affinity to a1- and no affinity to a5-containing GABAA receptors (Arbilla et al., 1986; Pritchett and Seeburg, 1990; Graham et al., 1996; Atack et al., 1999). Fig. 2. Acute effects of zolpidem and diazepam on ambulatory locomotor activity (LMA) GABAA receptors containing a1 subunit are supposed to be in mice. Zolpidem and diazepam were administered i.p. 15 min before placing the responsible for the sedative activity of zolpidem and benzodiaze- animals in the apparatus (Ugo-Basile Activity Cage) equipped with infrared sensors. The horizontal locomotor activity was measured for 10 min and quantified by interruptions pines (Perrault et al., 1988; Rudolph et al., 1999; McKernan et al., of infrared beams, which were counted and recorded automatically (counts/10 min). 2000; Kralic et al., 2002) and for the anticonvulsant effect of Points represent means SEM from at least 8 animals per group. *p ¼ 0.001 versus zolpidem against PTZ-induced tonic seizures (Crestani et al., 2000). control saline treated group (two-way ANOVA and Newman–Keuls test). Myoclonic, clonic and tonic seizures have different neuronal substrates. Myoclonic seizures originate in thalamocortical circuitry; clonic seizures have been shown to arise from the cere- 2 seizure type (F2,29 ¼ 172.12, hp ¼ 0.856, observed power ¼ 1.000) bral cortex and forebrain, whereas tonic seizures originate in the and treatment seizure type interaction (F2,29 ¼ 12.89, p ¼ 0.000, brain stem (cf. Andre´ et al., 1998). Immunocytochemical studies 2 hp ¼ 0.308, observed power ¼ 0.996) on the threshold for PTZ- demonstrated that the mentioned brain regions exhibit different induced seizures. There was no significant effect of drug distribution of GABAA receptor subtypes (Pirker et al., 2000). While (F1,29 ¼ 1.69), neither interaction between drug and treatment brain stem regions (medulla, pontine and cranial nerve nuclei) (F1,29 ¼ 1.31) nor between drug and seizure type (F2,29 ¼ 0.687). exhibit strong subunit a1-, b2- and g2-immunoreactivities, Post hoc comparisons revealed that repeated diazepam in a2-imunoreactivity was preferentially located in forebrain areas comparison with acute diazepam and repeated zolpidem in and a3- especially in reticular thalamic nucleus and inner layers of comparison with acute zolpidem produced a significant reduction the cerebral cortex. The strongest a5-subunit immunoreactivity of the threshold for all seizure types (p ¼ 0.05 or p ¼ 0.001). The was present in forebrain regions such as hippocampus, olfactory anticonvulsant effects of acute diazepam against PTZ-induced bulb and hypothalamus. myoclonic and clonic seizures were significantly (p ¼ 0.001) greater Fradley et al. (2007), who examined the ability of diazepam to than the effects of zolpidem. After repeated treatment, this differ- reduce PTZ-induced and maximal electroshock-induced seizures in ence in anticonvulsant potency between diazepam and zolpidem mice containing point mutations in single or multiple alpha was lost. subunits, suggested that GABAA receptors containing a5 subunit do not mediate the anticonvulsant effects of diazepam. Their results also suggested that a2-containing receptors play a larger role in 3.4. Effects of repeated treatment with zolpidem and diazepam on mediating the anticonvulsant properties of this drug than those locomotor activity in mice containing a1, although the efficacy at different subtypes of GABAA receptors may act synergistically. The ambulatory locomotor activity of mice treated with diazepam Based on the experiments demonstrating that b-CCT, a selective acutely or repeatedly (10 days) was increased in comparison to mice antagonist of BZ1 receptors i.e., of a1-containing GABAA receptors, treated with zolpidem, suggesting a higher level of sedation in zol- failed to affect the effect of diazepam and zolpidem on seizures pidem treated animals. In addition, mice repeatedly treated with produced by PTZ, Griebel et al. (1999) suggested that BZ2 receptors diazepam and zolpidem had an increased locomotor activity as may be primarily involved in the convulsant action of PTZ. Hence, compared to mice treated acutely with these drugs, suggesting a possible predominant involvement of a2-containing GABAA a development of tolerance to sedative effects of drugs (Fig. 4A and B). receptors in the convulsant action of PTZ (Griebel et al., 1999) and Two-way ANOVA confirmed (statistical power ¼ 1.000) a highly a different affinity of zolpidem and diazepam for different subtypes 2 significant (p ¼ 0.001) effect of drug: A (F1,30 ¼ 61.63, hp ¼ 0.673); B of GABAA receptors (primarily those containing a1 and a2 2 2 (F1,29 ¼ 78.81, hp ¼ 0.731) and treatment: A (F1,30 ¼ 11.72, hp ¼ 0.281, subunits), may explain the greater potency of zolpidem against 2 power ¼ 0.912); B (F1,29 ¼ 20.85, hp ¼ 0.418, power ¼ 0.993), a non- tonic than against myoclonic and clonic seizures as well as dose- significant drug treatment interaction for A (F1,30 ¼ 1.83) and dependent differences in the potency of zolpidem and diazepam 2 a significant interaction (p ¼ 0.005) for B (F1,29 ¼ 9.15, hp ¼ 0.240, against different PTZ-induced seizure types. power ¼ 0.877), indicating that in the latter case the differences It was already known that zolpidem has anticonvulsant activity between the two drugs were dependent on the treatment. against convulsions induced by isoniazid, PTZ and maximal

J. Vlainic´, D. Pericˇic´ / Neuropharmacology xxx (2009) 1–7 5

Fig. 3. Chronic effects of zolpidem and diazepam challenged with 3 mg/kg i.p. on the 11th (A–C) or 12th (D–F) day from the start of the repeated treatment (5 mg/kg i.p. twice daily for 10 days) on the threshold for myoclonic (A, D), clonic (B, E) and tonic (C, F) seizures induced in mice by i.v. infusion of PTZ. Drugs were administered 30 min before the infusion of PTZ was started. To test the locomotor activity, the same animals spent 10 min (15–25 min following drug administration) in Ugo-Basile Activity Cage. Points represent means SEM from 7 to 10 animals per group. *p ¼ 0.05–0.01, **p ¼ 0.001 versus the corresponding group receiving test dose of drugs after repeated saline; jp ¼ 0.05, jjp ¼ 0.001 versus the corresponding diazepam treated group (three-way ANOVA and Newman–Keuls test). In the control group (N ¼ 10), which received saline on the 11th day from the start of repeated treatment with saline, the threshold for myoclonic (28.9 0.63), clonic (32.1 0.65) and tonic (69.5 3.17) seizures was similar to that obtained in the acute experiment.

electroshock (Depoortere et al., 1986; Sanger et al., 1996), but early locomotor activity i.e., a sedative effect of zolpidem, but of diaz- studies reported that sedation with zolpidem occurs at doses 10 epam as well, was seen already after 0.3 mg/kg, a dose that failed to times lower than those inducing anticonvulsant effect (Depoortere increase the threshold for PTZ-induced seizures. The same dose of et al., 1986). In the present work, a signiﬁcant reduction of diazepam had a weak although signiﬁcant anticonvulsant effect

6 J. Vlainic´, D. Pericˇic´ / Neuropharmacology xxx (2009) 1–7

by lower seizure thresholds for PTZ-induced convulsions and an increased locomotor activity. Our results demonstrated that repeated zolpidem also induces both sedative and anticonvulsant tolerance. Mice which received diazepam (3 mg/kg) after repeated (10 days) i.p. injections of saline had a lower seizure threshold and a greater locomotor activity than mice receiving only one acute injection of diazepam. Because repeated i.p. injections are stressful for the animals, these results appear to be in line with those of Weizman et al. (1989), showing a decrease of benzodiazepine binding sites and the reduction in clonazepam’s anticonvulsant potency after repeated swim stress. Irrespective of whether the testing was done 42 or 18 h after the last of 20 repeated drug injections (i.e., whether the treatment was discontinued or not), decreases of the threshold for both diazepam and zolpidem were greater for tonic than for clonic and myoclonic seizures. The effect of diazepam against tonic seizures appeared to be more pronounced when mice were tested 42 than 18 h after repeated treatment. In both experiments, a significant treatment seizure type interaction, but not the effect of drug or drug treatment and drug seizure type interaction, suggested that the treatment effect changes with the seizure type. In contrast to this, testing of locomotor activity 18 or 42 h after termination of repeated treatment indicated that both treatment and drug had a significant effect on the development of tolerance to the sedative effects of drugs. It appears that mice withdrawn from diazepam for 42 h developed a greater level of tolerance to its sedative effects than mice receiving a test dose of diazepam in continuation of repeated treatment. The animals which were drug free for 42 h did not show visible withdrawal symptoms. A significant drug treatment interaction, obtained when analyzing this experiment, indicated that mice did not develop the same level of tolerance to sedative effects of zolpidem and diazepam. As reviewed by Lo¨scher and Schmidt (2006) there are two major types of tolerance, pharmacokinetic or metabolic and pharmacodynamic or functional tolerance. While the first type of tolerance is due to changes in the pharmacokinetics of drugs, the other is due to a reduced sensitivity of target tissue. Previous studies suggested that long-term treatment with diazepam (Rundfeldt et al., 1995; Auta et al., 2008) or zolpidem (Theńot et al., 1988; Holt et al., 1997) Fig. 4. Chronic effects of zolpidem and diazepam challenged with 3 mg/kg i.p. on the did not change the pharmacokinetics of drugs. Thus, we can 11th (A) or 12th (B) day from the start of repeated treatment (5 mg/kg i.p. twice daily presume that both drugs produce functional tolerance. for 10 days) on ambulatory locomotor activity (LMA) in mice. Drugs were administered The mechanisms involved in the development of tolerance and 15 min before placing the animals in the apparatus (Ugo-Basile). The horizontal dependence following chronic treatment with benzodiazepines locomotor activity was measured for 10 min. Points represent means SEM from 7 to 10 animals per group. *p ¼ 0.05–0.01, **p ¼ 0.001 versus the corresponding group are not clear (Costa et al., 2002; Wafford, 2005), although different receiving test dose of drugs after repeated saline; jjp ¼ 0.001 versus the corre- mechanisms such as allosteric uncoupling of GABA and benzodi- sponding diazepam treated group (two-way ANOVA and Newman–Keuls test). In the azepine binding sites (Gallager et al., 1984; Friedman et al., 1996) control group (N ¼ 10), which received saline on the 11th day from the start of have been suggested. Van Rijnsoever et al. (2004), who demon- repeated treatment with saline, the locomotor activity count (556 23) was slightly greater than that obtained in the acute experiment. strated that mice with point mutation in a5 subunits fail to manifest sedative tolerance, proposed that diazepam binding to a5-GABAA receptors is a key mechanism underlying the diminu- against myoclonic and clonic seizures, while 1 mg/kg was active tion of its sedative efficacy with chronic treatment. However, in also against tonic seizures. The lowest dose of zolpidem that this study, zolpidem, which does not have an affinity for receptors produced a significant enhancement of the PTZ threshold for containing the a5 subunit (Pritchett and Seeburg, 1990; Graham myoclonic, clonic and tonic seizures was 1 mg/kg. et al., 1996), produced sedative tolerance. The results of a recent Previous studies failed to show the anticonvulsant effectiveness study suggested the important role of the a1, but not of the a5 of zolpidem against PTZ-induced myoclonic seizures (Crestani subunit in the development of tolerance to the anticonvulsant et al., 2000; Fradley et al., 2007). The results of this study showing actions of positive allosteric modulators of benzodiazepine that zolpidem (1 mg/kg) enhanced the threshold for PTZ-induced binding sites (Auta et al., 2008). These authors demonstrated that myoclonic seizures, suggest that the determination of the seizure long-term treatment with imidazenil, which has a high affinity threshold in response to i.v. infusion of PTZ is a more sensitive and intrinsic efficacy for a5, and low for a1-containing GABAA method than the use of a lethal dose of PTZ. This is in accordance receptors, failed to induce anticonvulsant tolerance. with the results published previously (Lo¨scher and Nolting, 1991). In conclusion, our results demonstrate that in the acute study As expected, mice treated repeatedly with diazepam developed the lowest doses of zolpidem and diazepam producing sedation tolerance to its anticonvulsant and sedative activity, as evidenced were 0.3, while those elevating the threshold for PTZ-induced tonic

J. Vlainic´, D. Pericˇic´ / Neuropharmacology xxx (2009) 1–7 7 seizures were 1 mg/kg. Both diazepam and zolpidem were more benzodiazepine receptor ligands can be associated with activities at different effective against PTZ-induced tonic than against myoclonic and BZ (u) receptor subtypes. Psychopharmacology 146, 205–213. Holt, R.A., Bateson, A.N., Martin, I.L., 1997. Chronic zolpidem treatment alters GABAA clonic seizures. Mice treated repeatedly with zolpidem, like those receptor mRNA levels in the rat cortex. European Journal of Pharmacology 329, treated with diazepam, developed tolerance to its sedative and 129–132. anticonvulsant effects. Because zolpidem does not have an affinity Kralic, J.E., O’Buckley, T.K., Khisti, R.T., Hodge, C.W., Homanics, G.E., Morrow, A.L., 2002. GABAA receptor alpha-1 subunit deletion alters receptor subtype for a5-containing GABAA receptors, the results suggest that pro- assembly, pharmacological and behavioral responses to benzodiazepines and longed stimulation of this receptor subtype is not essential for the zolpidem. Neuropharmacology 43, 685–694. development of sedative and anticonvulsant tolerance. After Lo¨scher, W., Nolting, B., 1991. The role of technical, biological and pharmacological factors in the laboratory evaluation of anticonvulsant drugs. IV. Protective repeated treatment zolpidem and diazepam are approximately indices. Epilepsy Research 9, 1–10. equipotent anticonvulsants, at least against PTZ-induced convul- Lo¨scher, W., Schmidt, D., 1988. Which animal models should be used in the search sions in mice, but zolpidem remains more potent in producing for new antiepileptic drugs? A proposal based on experimental and clinical considerations. Epilepsy Research 2, 145–181. sedation. Lo¨scher, W., Schmidt, D., 2006. Experimental and clinical evidence for loss of effect (tolerance) during prolonged treatment with antiepileptic drugs. Epilepsia 47, Acknowledgements 1253–1284. McKernan, R.M., Rosahl, T.W., Reynolds, D.S., Sur, C., Wafford, K.A., Atack, J.R., Farrar, S., Myers, J., Cook, G., Ferris, P., Garrett, L., Bristow, L., Marshall, G., This study was supported by the Croatian Ministry of Science, Macaulay, A., Brown, N., Howell, O., Moore, K.W., Carling, R.W., Street, L.J., Castro, J.L., Ragan, C.I., Dawson, G.R., Whiting, P.J., 2000. Sedative but not Education and Sports (Project: Stress, GABAA receptors and mech- anxiolytic properties of benzodiazepines are mediated by the GABA receptor anisms of action of neuropsychoactive drugs). The authors express A a1 subtype. Nature Neuroscience 3, 587–592. their gratitude to Mr. Juraj Jerkovic´ and Mr. Ernest Mesˇtrovic´ from Pericˇic´, D., Jazvinsˇcák, M., Sˇvob, D., Mirkovic´, K., 2001. Swim stress alters the Pliva, Zagreb, Croatia, for their kind donation of zolpidem. The behavioural response of mice to GABA-related and some GABA-unrelated convulsants. Epilepsy Research 43, 145–152. skilful technical assistance of Mrs. Zlatica Tonsˇetic´ is gratefully Pericˇic´, D., Vlainic´, J., Sˇvob Sˇtrac, D., 2008. Sedative and anticonvulsant effects of acknowledged. zolpidem in adult and aged mice. Journal of Neural Transmission 115, 795–802. Perrault, G., Morel, E., Sanger, D.J., Zivkovic, B., 1988. The interaction between zolpidem and beta-CMC: a clue to the identification of receptor sites involved in References the sedative effect of zolpidem. European Journal of Pharmacology 156, 189–196. Andre´, V., Pineau, N., Motte, J.E., Marescaux, C., Nehlig, A., 1998. Mapping of Perrault, G., Morel, E., Sanger, D.J., Zivkovic, B., 1992. Lack of tolerance and physical neuronal networks underlying generalized seizures induced by increasing dependence upon repeated treatment with the novel hypnotic zolpidem. doses of pentylenetetrazol in the immature and adult rat: a c-Fos immuno- Journal of Pharmacology and Experimental Therapeutics 263, 298–303. histochemical study. European Journal of Neuroscience 10, 2094–2106. Pirker, S., Schwarzer, C., Wieselthaler, A., Sieghart, W., Sperk, G., 2000. GABA(A) Arbilla, S., Allen, J., Wick, A., Langer, S.Z., 1986. High affinity [3H]zolpidem binding in receptors: immunocytochemical distribution of 13 subunits in the adult rat the rat brain: an imidazopyridine with agonist properties at central benzodi- brain. Neuroscience 101, 815–850. azepine receptors. European Journal of Pharmacology 130, 257–263. Pritchett, D.B., Seeburg, P.H., 1990. Gamma-aminobutyric acidA receptor alpha 5- Atack, J.R., Smith, A.J., Emms, F., McKernan, R.M., 1999. Regional differences in the subunit creates novel type II benzodiazepine receptor pharmacology. Journal of inhibition of mouse in vivo [3H]Ro 15-1788 binding reflect selectivity for a1 Neurochemistry 54, 1802–1804. versus a2 and a3 subunit-containing GABAA receptors. Neuro- Rudolph, U., Crestani, F., Benke, D., Bru¨ nig, I., Benson, J.A., Fritschy, J.M., Martin, J.R., psychopharmacology 20, 255–262. Bluethmann, H., Mo¨hler, H., 1999. Benzodiazepine actions mediated by specific Auta, J., Impagnatiello, F., Kadriu, B., Guidotti, A., Costa, E., 2008. Imidazenil: a low g-aminobutyric acidA receptor subtypes. Nature 401, 796–800. efficacy agonist at a1- but high efficacy at a5-GABAA receptors fail to show Rundfeldt, C., Wlaz´, P., Ho¨nack, D., Lo¨scher, W., 1995. Anticonvulsant tolerance and anticonvulsant cross tolerance to diazepam or zolpidem. Neuropharmacology withdrawal characteristics of benzodiazepine receptor ligands in different 55, 148–153. seizure models in mice. Comparison of diazepam, bretazenil and abecarnil. Cohen, J., 1962. The statistical power of abnormal-social psychological research: Journal of Pharmacology and Experimental Therapeutics 275, 693–702. a review. Journal of Abnormal and Social Psychology 65, 145–153. Sanger, D.J., 2004. The pharmacology and mechanisms of action of new generation, Costa, E., Auta, J., Grayson, D.R., Matsumoto, K., Pappas, G.D., Zhang, X., Guidotti, A., non-benzodiazepine hypnotic agents. CNS Drugs 18 (Suppl. 1), 9–15. 2002. GABAA receptors and benzodiazepines: a role for dendritic resident Sanger, D.J., Morel, E., Perrault, G., 1996. Comparison of the pharmacological profiles subunit mRNAs. Neuropharmacology 43, 925–937. of the hypnotic drugs, zaleplon and zolpidem. European Journal of Pharma- Crestani, F., Martin, J.R., Mo¨hler, H., Rudolph, U., 2000. Mechanism of action of cology 313, 35–42. hypnotic zolpidem in vivo. British Journal of Pharmacology 131, 1251–1254. Scharf, M.B., Roth, T., Vogel, G.W., Walsh, J.K., 1994. A multicenter, placebo- Depoortere, H., Zivkovic, B., Lloyd, K.G., Sanger, D.J., Perrault, G., Langer, S.Z., controlled study evaluating zolpidem in the treatment of chronic insomnia. Bartholini, G., 1986. Zolpidem, a novel nonbenzodiazepine hypnotic. I. Neuro- Journal of Clinical Psychiatry 55, 192–199. pharmacological and behavioral effects. Journal of Pharmacology and Experi- Snead 3rd, O.C., 1992. Pharmacological models of generalized absence seizures in mental Therapeutics 237, 649–658. rodents. Journal of Neural Transmission 35 (Suppl.), 7–19. Elliot, E.E., White, J.M., 2000. Precipitated and spontaneous withdrawal following Snead 3rd, O.C., 1995. Basic mechanisms of generalized absence seizures. Annals of administration of lorazepam but not zolpidem. Pharmacology Biochemistry and Neurology 37, 146–157. Behavior 66, 361–369. Theńot, J.P., Hermann, P., Durand, A., Burke, J.T., Allen, J., Garrigou, D., Vajta, S., Fahey, J.M., Grassi, J.M., Reddi, J.M., Greenblatt, D.J., 2006. Acute zolpidem admin- Albin, H., The´bault, J.J., Olive, G., Warrington, S.J., 1988. Pharmacokinetics and istration produces pharmacodynamic and receptor occupancy changes at metabolism of zolpidem in various animal species and in humans. In: similar doses. Pharmacology Biochemistry and Behavior 83, 21–27. Sauvanet, J.P., Langer, S.Z., Morselli, P.L. (Eds.), Imidazopyridines in Sleep Fradley, R.L., Guscott, M.R., Bull, S., Hallett, D.J., Goodacre, S.C., Wafford, K.A., Disorders. Raven Press, New York, pp. 139–153. Garrett, E.M., Newman, R.J., O’Meara, G.F., Whiting, P.J., Rosahl, T.W., Van Rijnsoever, C., Taüber, M., Choulli, M.K., Keist, R., Rudolph, U., Mohler, H., Dawson, G.R., Reynolds, D.S., Atack, J.R., 2007. Differential contribution of Fritschy, J.M., Crestani, F., 2004. Requirement of a5-GABAA receptors for the GABAA receptor subtypes to the anticonvulsant efficacy of benzodiazepine site development of tolerance to the sedative action of diazepam in mice. Journal of ligands. Journal of Psychopharmacology 21, 384–391. Neuroscience 24, 6785–6790. Friedman, L.K., Gibbs, T.T., Farb, D.H., 1996. g-Aminobutyric acid A receptor regu- Wafford, K.A., 2005. GABAA receptor subtypes: any clues to the mechanism of lation: heterologous uncoupling of modulatory site interactions induced by benzodiazepine dependence? Current Opinion in Pharmacology 5, 47–52. chronic steroid, barbiturate, benzodiazepine, or GABA treatment in culture. Weizman, R., Weizman, A., Kook, K.A., Vocci, F., Deutsch, S.I., Paul, S.M., 1989. Brain Research 707, 100–109. Repeated swim stress alters brain benzodiazepine receptors measured in vivo. Gallager, D.W., Lakoski, J.M., Gonsalves, S.F., Rauch, S.L., 1984. Chronic benzodiaze- Journal of Pharmacology and Experimental Therapeutics 249, 701–707. pine treatment decreases postsynaptic GABA sensitivity. Nature 308, 74–77. White, H.S., 2003. Preclinical development of antiepileptic drugs: Past, present and Graham, D., Faure, C., Besnard, F., Langer, S.Z., 1996. Pharmacological profile of future directions. Epilepsia 44 (Suppl. 7), 2–8. benzodiazepine site ligands with recombinant GABAA receptor subtypes. Wulff, P., Goetz, T., Leppa¨, E., Linden, A.M., Renzi, M., Swinny, J.D., Vekovischeva, O.Y., European Neuropsychopharmacology 6, 119–125. Sieghart, W., Somogyi, P., Korpi, E.R., Farrant, M., Wisden, W., 2007. From Griebel, G., Perrault, G., Letang, V., Granger, P., Avenet, P., Schoemaker, H., synapse to behavior: rapid modulation of defined neuronal types with engi- Sanger, D.J., 1999. New evidence that the pharmacological effects of neered GABAA receptors. Nature Neuroscience 10, 923–929.