Trazodone reduces the anticonvulsant action of certain classical antiepileptics in the mouse maximal electroshock model

(1)

Trazodone reduces the anticonvulsant action of certain classical antiepileptics in the mouse maximal electroshock model

Kinga K. Borowicz¹, Elwira Gurdziel¹, Stanis³aw J. Czuczwar^2,3

1Independent Unit of Experimental Neuropathophysiology,²Department of Pathophysiology, Medical University, Jaczewskiego 8, PL 20-090 Lublin, Poland

3Department of Physiopathology, Institute of Agricultural Medicine, Jaczewskiego 2, PL 20-950 Lublin, Poland

Correspondence: Kinga K. Borowicz, e-mail: kinga.borowicz@umlub.pl

Abstract:

Background: The aim of the study was to examine effects of an acute and chronic treatment with trazodone, a serotonin antagonist and reuptake inhibitor (SARI), on the protective activity of four classical antiepileptic drugs provided in the maximal electroshock test in mice.

Methods: Electroconvulsions were produced in mice by means of an alternating current (50 Hz, 25 mA, 0.2 s) and delivered via ear- clip electrodes. Motor impairment in animals were assessed in the chimney test, and long-term memory deficits were quantified in the passive-avoidance task. Brain concentrations of antiepileptic drugs were analyzed by fluorescence polarization immunoassay.

Results: The obtained results showed that a single administration of trazodone (up to 40 mg/kg) did not influence the electroconvul- sive threshold. In contrast, chronic treatment with the antidepressant (40 mg/kg) significantly increased this parameter. Furthermore, both single and chronic administration of trazodone reduced the anticonvulsant effect of phenytoin and carbamazepine against the maximal electroshock. However, the antidepressant remained without effect on the anticonvulsant action of valproate and phenobarbital. Some interactions between trazodone and antiepileptic drugs may have a pharmacodynamic background. Both, acute and chronic treatment with the antidepressant diminished the brain concentration of phenytoin. Chronic trazodone lowered the brain levels of carbamazepine and phenobarbital. Moreover, acute and chronic trazodone increased the valproate concentration in the brain.

As regards undesired effects, acute and chronic trazodone (40 mg/kg), alone and in combination with phenytoin, significantly impaired long-term memory in tested animals, evaluated in the passive avoidance task. Acute trazodone (40 mg/kg) alone and combined with phenytoin produced also significant motor deficits in mice, as measured in the chimney test.

Conclusion: The obtained results allow to conclude that trazodone is not a good candidate for an antidepressant drug in epileptic patients.

Key words:

trazodone, antiepileptic drugs, electroshock maximal, pharmacokinetic interaction

Introduction

It is widely accepted that the proper treatment of depression co-existing with epilepsy may improve the outcome of both disorders [29]. Therefore, the choice

of drugs used for a combination should be properly considered. Almost all antidepressants in overdose may induce seizures in both experimental and clinical conditions. Selective serotonin reuptake inhibitors (SSRIs) and serotonin/norepinephrine reuptake in-

(2)

hibitors (SNRIs), exhibit the lowest proconvulsive potential and are considered as drugs of choice in patients with concomitant epilepsy and depression [29].

In fact, numerous experimental studies confirmed that enhancement of noradrenergic and/or serotonergic neurotransmission might increase both antidepressant and anticonvulsant effects [20]. This would suggest that antidepressants with such a mechanism of action should even improve the course of epilepsy. In vivo electrophysiological study in the rat brain showed that sustained 14-day trazodone administration enhanced serotonergic neurotransmission. The antidepressant reduced serotonin reuptake, increased serotonin re- lease by decreasing the inhibitory function of 5-HT_1B autoreceptors, and activated postsynaptic 5-HT_1Are- ceptors [14, 21]. Trazodone shows also moderate po- tency at blocking 5-HT_2Areceptors [14, 28], which, at least theoretically, increases the binding of serotonin to 5-HT_1Areceptors. All these effects may contribute to the antidepressant action of trazodone. The antide- pressant acts also as an antagonist of a₁and H₁receptors. This blockade, in turn, facilitates sleep. It is worth stressing that clinical effects of trazodone applied at the dose range of 25–150 mg/kg result mainly from its antagonistic action on 5-HT_2A, a₁and H₁receptors [36].

Effects of antidepressant drugs on seizure phenomena, in contrast to antidepressant effects, can be observed even after the first injection. In experimental studies, the electroconvulsive threshold was elevated by several tricyclic antidepressants [22], SSRIs, SNRIs, and mianserin [3–5, 7, 29]. Moreover, some of mentioned antidepressants may also potentiate the anticonvulsant effect of several conventional antiepileptics [3–5, 7]. Effects of chronic administration of antidepressant drugs are more challenging and less predictable. However, such experimental protocols correspond more to clinical practice, thus providing a valuable complement to procedures based on the single drug application. For instance, although acute fluoxetine increased the threshold for electroconvulsions, chronic treatment decreased [27] or failed to affect this parameter [7]. Similarly, acute milnacipran showed an anticonvulsant action [4], but chronic application of this antidepressant exhibited proconvulsive properties and diminished the action of valproate and phenytoin against the maximal electroshock test [3].

In this study, we decided to continue our research and examine the influence of trazodone on the an- tielectroshock action of conventional antiepileptics.

From a theoretical point of view, enhancement of the serotonergic neurotransmission induced by this drug may be beneficial in the treatment of depression co- existing with epilepsy. At present, trazodone, considered as a safe and well tolerated antidepressant, is in- creasingly being used in the treatment of depression, panic attacks, aggressive behavior, agoraphobia, in- somnia and cocaine withdrawal [17]. Results of the present study may help to determine whether trazodone would be advantageous in the combined treatment with classical antiepileptic drugs in epileptic patients with depression.

Materials and Methods

Animals

Experiments were carried out on male Swiss mice weighing 20–25 g. The animals were housed in col- ony cages with free access to food (chow pellets) and tap water. The experiments started after 7-day accli- matization to standardized laboratory conditions (temperature 21 ± 1°C, a natural light-dark cycle).

The tested groups, consisting of eight animals, were randomly assigned. All experiments were performed in spring months (from March to June) between 9:00 a.m. and 2:00 p.m. Each mouse was used only once. The Local Ethical Committee of Lublin Medi- cal University approved all experimental procedures of this study (license number 45/2008).

Drugs

The following drugs were used in this study: trazodone, carbamazepine, phenytoin (all three drugs from Sigma-Aldrich, St. Louis, MO, USA), valproate mag- nesium (a generous gift from ICN-Polfa, Rzeszów, Poland), and phenobarbital (Polfa, Kraków, Poland).

Valproate was dissolved in distilled water, while trazodone, carbamazepine, phenytoin, and phenobarbital were suspended in a 1% solution of Tween 80 (Sigma-Aldrich, St. Louis, MO, USA). All drugs were prepared each day as fresh solutions or suspen- sions and administered intraperitoneally (ip) in a vol- ume of 0.01 ml/g body weight. Trazodone was administered in a single injection 30 min before tests (acute protocol), or it was given for 14 days every 24 h, on the last day – 30 min before tests (chronic protocol).

(3)

Antiepileptic drugs were injected only once, phenytoin – 120 min, phenobarbital – 60 min, while valproate and carbamazepine – 30 min before electroconvulsions and behavioral tests. The dose range of trazodone was based on doses used by Silvestrini et al. [33]. Time of trazodone injection was determined experimentally as time of the maximal effect against electroconvulsions. Dose ranges and times of administration of antiepileptic drugs were established experimentally in our previous studies [3–7].

Electroconvulsive threshold and maximal electroshock seizure test

Electrically-induced seizures in rodents are a well- known animal model of tonic-clonic convulsions [16].

Electroconvulsions were produced by a Hugo Sachs generator (Rodent Shocker, type 221, Freiburg, Germany). An alternating current (50 Hz, fixed current intensity of 25 mA, maximum stimulation volt- age of 500 V) was delivered via ear-clip electrodes.

The stimulus duration was 0.2 s. Tonic hindlimb extension (the hindlimbs of animals outstretched 180° to the plain of the body axis) was considered as the endpoint.

The electroconvulsive threshold was evaluated as CS₅₀, which is the current strength (expressed in mA) necessary to induce tonic convulsions in 50% of animals. To estimate the electroconvulsive threshold, at least four groups of mice (eight animals per group) were challenged with currents of various intensities (4–12 mA). CS₅₀values were calculated by computer from the equation of an intensity-response curve based on the percentage of convulsing animals [24].

The protective efficacy of antiepileptic drugs was determined as their ability to protect 50% of animals against the maximal electroshock-induced tonic hindlimb extension and expressed as respective values of the median effective dose (ED₅₀). To evaluate each ED₅₀ value (in mg/kg), at least four groups of mice received progressive doses of an antiepileptic drug and were challenged with the maximal electroshock test. ED₅₀values were calculated by computer from the equation of a dose-response curve based on the percentage of convulsing animals [24].

Chimney test

The effect of antiepileptic drugs, trazodone, and combinations of trazodone with antiepileptics on motor coordination was quantified in the chimney test [2]. In this test, animals had to climb backward up the plastic

tube (25 cm length, 3 cm inner diameter). Motor impairment was indicated by the inability of mice to per- form this test within 60 s.

Step-through passive-avoidance task

The effect of antiepileptic drugs, trazodone, and trazodone/antiepileptics combinations on time of retention was assessed in the step-through passive-avoidance that may be recognized as a measure of long-term memory [43]. The drug-treated mice were placed in an illuminated box (10 × 13× 15 cm) connected to a large dark box (25 × 20 × 15 cm), which was equipped with an electric grid floor. Entrance of the animals to the dark box was punished by an electric foot shock (0.6 mA for 2 s; facilitation of acquisition). The mice that did not enter the dark compartment within 60 s were excluded from the experiment. On the next day (24 h later), the same animals, without any treatment, were put into the illuminated box and observed up to 180 s. The median time to enter the dark box was subsequently calculated.

The control (vehicle-treated animals) did not enter the dark box within the observation time limit. The results were shown as medians with 25^thand 75^thpercentiles.

Measurement of brain concentrations of antiepileptic drugs

Mice were administered one of the conventional antiepileptic drugs + vehicle or the respective antiepileptic drug + trazodone. The antidepressant was applied in a single injection or chronically for 14 days.

Animals were killed by decapitation at times respective to those scheduled for the maximal electroshock test. Brains were removed from skulls, weighed, and homogenized using Abbott buffer (Abbott Laborato- ries, North Chicago, IL, USA; 2:1 vol/weight) in an Ultra-Turrax T8 homogenizer (IKA-Werke, Stauffen, Germany). The homogenates were centrifuged at 10,000 × g for 10 min. The supernatant samples (75 µl) were analyzed by fluorescence polarization immunoassay for phenytoin, carbamazepine, valproate, or phenobarbital content using a TDx analyzer and reagents exactly as described by the manufacturer (Abbott Laboratories, North Chicago, IL, USA). All concentrations of antiepileptic drugs are expressed in µg/ml of brain supernatants as the means ± standard deviation (SD) of at least eight determinations. More details upon this method may be found in Borowicz et al. [7].

(4)

Fig. 2. Effect of the acute treatment with trazodone (TRA) on the anticonvulsant action of carbamazepine – CBZ (A), valproate – VPA (B), phenytoin – PHT (C), and phenobarbital – PB (D) against maximal electroshock-induced seizures in mice. Data are presented as median effec- tive doses (ED₅₀s with SEM values), at which antiepileptic drugs protected 50% of animals against seizures. All drugs were administered ip:

trazodone, valproate, and carbamazepine – 30 min, phenobarbital – 60 min, and phenytoin – 120 min before electroconvulsions; * p < 0.05 vs.

control (animals treated with an antiepileptic plus saline)

Fig. 1. Effect of the acute (A) and chronic (B) treatment with trazodone (TRA) on the electroconvulsive threshold in mice. Data are presented as median current strength (CS₅₀with SEM) producing tonic convulsions in 50% of animals. Acute TRA was injected ip 30 min before the test.

Chronic TRA was applied for 14 days, the last time was 30 min before electroconvulsions; *** p < 0.001 vs. control (animals treated with saline)

(5)

Statistics

ED₅₀ values with their respective 95% confidence limits were estimated using computer log-probit analysis according to Litchfield and Wilcoxon [24].

Subsequently, standard error (SEM) of the mean values were calculated and compared by the Student’s t-test [6].

Qualitative variables from the chimney test were compared by the Fisher’s exact probability test, whereas the results obtained in the step-through passive-avoidance task were statistically evaluated using the Kruskal-Wallis nonparametric analysis of variance (ANOVA) followed by post-hoc Dunn’s test.

Total brain concentrations of antiepileptic drugs were evaluated with the unpaired Student’s t-test.

The significance level was set at p £ 0.05.

Results

Electroconvulsive threshold test

Acute trazodone, applied at doses of 10–40 mg/kg, did not affect the electroconvulsive threshold evaluated as 5.1 ± 0.40 mA. Similarly, chronic trazodone administered at lower doses (10 and 20 mg/kg) was not effective in the electroconvulsive test in mice. However, the antidepressant administered chronically at the dose of 40 mg/kg significantly increased the threshold from 5.1

± 0.55 mA to 6.7 ± 0.37 mA (Fig. 1).

Maximal electroshock test

Both acute and chronic trazodone, administered alone, was ineffective in the maximal electroshock

* ** **

0 2 4 6 8 10 12 14 16 18 20

CBZ+vehicle CBZ+TRA (10) CBZ+TRA (20) CBZ+TRA (40) ED50(mg/kg)

A

0 50 100 150 200 250 300

VPA+vehicle VPA+TRA (10) VPA+TRA (20) VPA+TRA (40) ED50(mg/kg)

B

*

** ***

0 5 10 15 20 25

PHT+vehicle PHT+TRA (10) PHT+TRA (20) PHT+TRA (40) ED50(mg/kg)

C

0 5 10 15 20 25 30

PB+vehicle PB+TRA (10) PB+TRA (20) PB+TRA (40) ED50(mg/kg)

D

Fig. 3. Effect of the chronic treatment with trazodone (TRA) on the anticonvulsant action of carbamazepine – CBZ (A), valproate – VPA (B), phenytoin – PHT (C), and phenobarbital – PB (D) against maximal electroshock-induced seizures in mice. Data are presented as median effec- tive doses (ED₅₀s with SEM values), at which antiepileptic drugs protected 50% of animals against seizures. All drugs were administered ip, trazodone for 14 days and antiepileptic drugs in a single injection: valproate and carbamazepine – 30 min, phenobarbital – 60 min, and pheny- toin – 120 min before electroconvulsions; * p < 0.05, ** p < 0.01, *** p < 0.001 vs. control (animals treated with an antiepileptic plus saline)

(6)

test in mice. Single administration of the antidepressant (10–40 mg/kg) failed to influence the protective action of valproate (ED₅₀= 234.0 ± 7.50 mg/kg) and phenobarbital (ED₅₀ = 17.9 ± 1.54 mg/kg). On the other hand, acute trazodone applied at 20 and 40 mg/kg significantly diminished the antiseizure action of phenytoin, increasing its ED₅₀value from 8.2 ± 0.81 to 12.2 ± 0.79, and 12.5 ± 1.65 mg/kg, respectively.

Single injection of trazodone (40 mg/kg) also attenu- ated the action of carbamazepine, elevating its ED₅₀ value from 11.4 ± 1.13 to 15.4 ± 1.13 mg/kg (Fig. 2).

Chronic trazodone, applied at the subprotective doses of 10 and 20 mg/kg, decreased the antiseizure action of phenytoin and carbamazepine. The respective ED₅₀values of phenytoin were changed from 9.9

± 0.86 to 14.3 ± 1.62, and 16.2 ± 1.65 mg/kg, respectively. ED₅₀ values of carbamazepine elevated from 12.5 ± 0.70 to 16.9 ± 1.37 and 17.2 ± 1.28 mg/kg, respectively. Chronic administration of trazodone did not significantly influence control ED₅₀values of val-

proate (245.0 ± 10.36 mg/kg) and phenobarbital (25.0

± 1.06 mg/kg; Fig. 3).

Chimney test and step-through passive- avoidance task

Trazodone (applied acutely or chronically at doses of 20 and 40 mg/kg) and conventional antiepileptic drugs administered alone (at doses equal to their ED₅₀ values) did not cause any motor impairment in mice.

Among all combinations of trazodone with antiepileptic drugs, only this with phenytoin (12.5 mg/kg) and acute trazodone (administered at the highest dose of 40 mg/kg) resulted in a significant motor deficit.

Single administration of trazodone (20–40 mg/kg) and classical antiepileptics alone did not impair long- term memory in mice. However, the combined treatment with acute trazodone (20 and 40 mg/kg) and phenytoin (12.2 and 12.5 mg/kg, respectively) led to significant memory deficits. Similarly, chronic trazo-

Tab. 2. Effect of chronic trazodone, conventional antiepileptic drugs alone and in combination with trazodone on motor performance and long-term memory

Treatment (mg/kg)

Animals impaired (%)

Retention time (s)

Vehicle 0 180 (180, 180)

TRA (10) 10 180 (100, 180)

TRA (20) 0 180 (180, 180)

TRA (40) 0 127.5 (53, 180)*

CBZ (12.6) – ED₅₀ 0 140 (110, 180)

CBZ (16.9) 0 153.5 (40, 180)

CBZ (16.9) + TRA (10) 0 135 (48, 180)

CBZ (17.2) – ED₅₀ 10 180 (56, 180)

CBZ (17.2) + TRA (20) 0 180 (180, 180) CBZ (17.2) + TRA (40) 10 99 (81, 180)

PHT (9.9) – ED₅₀ 10 180 (180, 180)

PHT (14.3) 0 180 (180, 180)

PHT (14.3) + TRA (10) 0 175 (51, 180)

PHT (16.2) 0 175 (105, 180)

PHT (16.2) + TRA (20) 0 180 (142, 180)

PHT (18.1) 0 170 (100, 180)

PHT (18.1) + TRA (40) 0 67.5 (44, 91)**

* p < 0.05, ** p < 0.01 vs. control. For details see the legend to Table 1 Tab. 1. Effect of acute trazodone, conventional antiepileptic drugs

alone and in combination with trazodone on motor performance and long-term memory in mice

Treatment (mg/kg)

Animals impaired (%)

Retention time (s)

Vehicle 0 180 (180, 180)

TRA (20) 0 180 (136, 180)

TRA (40) 0 172 (53, 180)

CBZ (11.4) – ED₅₀ 0 180 (180, 180)

CBZ (15.4) 10 180 (180, 180)

CBZ (15.4) + TRA (40) 0 180 (79, 180) PHT (8.2) – ED₅₀ 0 180 (170, 180)

PHT (12.2) 0 180 (180, 180)

PHT (12.2) + TRA (20) 0 116 (83, 148)**^,##

PHT (12.5) 10 180 (180, 180)

PHT (12.5) + TRA (40) 60*^,# 76.5 (27, 148)***^,###

Results are expressed as percentage of animals that failed to per- form the chimney test and as median retention time (with 25^thand 75^thpercentiles), during which the animals avoided the dark compartment in the passive avoidance task. Statistical analysis of data from the chimney test was performed by using the Fisher’s exact probability test, whereas the results from the step-through passive avoidance task were analyzed using the nonparametric Kruskal- Wallis ANOVA test followed by Dunn’s post-hoc test. TRA, trazodone;

CBZ, carbamazepine; PHT, phenytoin; * p < 0.05, ** p < 0.01, *** p <

0.001 vs. control;^#p < 0.05,^##p < 0.01,^###p < 0.001 vs. respective antiepileptics alone

(7)

done (40 mg/kg), applied either alone or in combination with phenytoin (18.1 mg/kg) markedly diminished long-term memory in mice (Tab. 1, 2).

Influence of trazodone on total brain concentra- tions of antiepileptic drugs

Acute trazodone (40 mg/kg) significantly decreased the brain concentration of phenytoin (12.5 mg/kg), and increased that of valproate (208.5 mg/kg).

Chronic trazodone (40 mg/kg) markedly reduced the brain concentrations of phenytoin (8.5 mg/kg), carbamazepine (17.2 mg/kg) and phenobarbital (25.2 mg/kg). Chronic administration of the antidepressant increased, however, the brain level of valproate (223.9 mg/kg) (Tab. 3, 4).

Discussion

Results of this study suggest that only chronic treatment with trazodone leads to the protective effect in the electroconvulsive threshold test in mice. Either acute or chronic trazodone reduces, however, the an- tielectroshock properties of phenytoin and carbamazepine, remaining ineffective on the action of valproate and phenobarbital. Revealed interactions may be, at least partially, explained by pharmacokinetic

events, since both acute and chronic trazodone significantly lowered the brain concentration of phenytoin. Similarly, chronic, but not acute trazodone, diminished the brain level of carbamazepine. Neverthe- less, pharmacokinetic interactions were also noticed between the antidepressant and the two remaining antiepileptic drugs in spite of the lack of any change of their antiseizure activity. In details, acute and chronic trazodone increased the brain concentration of valproate. This suggests that a possible antagonistic interaction between the two medicines may be masked by pharmacokinetic phenomena. Moreover, chronic administration of trazodone decreased the brain level of phenobarbital, which may, in turn, mask a probable synergistic interaction. It should be stressed that the existing data on the action of trazodone in other animal seizure models are very scarce. The only available study was published more than 40 years ago.

In this report, acute treatment with AF 1161 (up to 80 mg/kg), a tool substance corresponding to trazodone, failed to affect pentylenetetrazole-, strychnine-, and maximal electroshock-induced convulsions in mice. Furthermore, the antidepressant administered at a higher dose of 100 mg/kg appeared to have some proconvulsant effects [34]. Results obtained in the pentylenetetrazole seizure model were confirmed in our study. On the other hand, proconvulsant action of trazodone found by Silvestrini et al. [34] can be at least partially involved (apart from pharmacokinetic events) in the trazodone-induced reduction of the anti-

Tab. 4. Effect of chronic trazodone on the brain concentrations of conventional antiepileptics in mice

Treatment (mg/kg) Brain concentration (µg/ml) VPA (223.9) + vehicle 58.52 ± 8.56 VPA (223.9) + TRA (40) 73.48 ± 16.08*

CBZ (17.2) + vehicle 4.65 ± 0.70 CBZ (17.2) + TRA (40) 3.18 ± 0.89**

PB (25.2) + vehicle 9.02 ± 1.13

PB (25.2) + TRA (40) 7.68 ± 1.28*

PHT (18.1) + vehicle 2.99 ± 0.71 PHT (223.9) + TRA (40) 2.14 ± 0.64*

Data are presented as the means ± SD of at least eight determinations. Statistical analysis of the brain concentrations of antiepileptic drugs was performed using the unpaired Student’s t-test. VLF, venla- faxine; CBZ, carbamazepine; PB, phenobarbital; PHT, phenytoin;

VPA, valproate; * p < 0.05, ** p < 0.01 vs. control group

Tab. 3. Effect of acute trazodone on the brain concentrations of con- ventional antiepileptics in mice

Treatment (mg/kg) Brain concentration (µg/ml) VPA (208.5) + vehicle 50.50 ± 5.64 VPA (208.5) + TRA (40) 66.94 ± 6.70***

CBZ (15.4) + vehicle 3.14 ± 0.93 CBZ (15.4) + TRA (40) 3.86 ± 0.77 PB (13.3) + vehicle 11.57 ± 2.49 PB (13.3) + TRA (40) 11.26 ± 1.37 PHT (12.5) + vehicle 2.30 ± 0.65 PHT (12.5) + TRA (40) 1.53 ± 0.26**

Data are presented as the means ± SD of at least eight determinations. Statistical analysis of the brain concentrations of antiepileptic drugs was performed using the unpaired Student’s t-test. VLF, venla- faxine; CBZ, carbamazepine; PB, phenobarbital; PHT, phenytoin;

VPA, valproate; ** p < 0.05, *** p < 0.001 vs. control group

(8)

convulsant effects provided by phenytoin and carbamazepine.

According to Duncan and Taylor [12], trazodone may be recommended, along with representatives of SSRIs and SNRIs, as the first-line treatment of co- existing depression and epilepsy. It should be remembered, however, that there are single reports on the proconvulsant action of trazodone in clinical conditions. The drug applied at the therapeutic dose-range induced generalized seizures in two non-epileptic patients [8, 23].

Most literature data relate to SSRIs. Representa- tives of this group did not lower the seizure threshold in patients with epilepsy [33]. In experimental conditions, acute [7], but not chronic fluoxetine [5] raised the electroconvulsive threshold in mice. This can suggest that initial anticonvulsant properties of the drug may change over time. In the pentylenetetrazole- induced clonic convulsions, fluoxetine showed either anticonvulsant [40] or proconvulsant [1, 27] action.

Some authors observed no significant effects of SSRIs on experimental convulsions [9].

Although venlafaxine and milnacipran belong to the same group of SNRIs, they showed different profile of the anticonvulsant action against electroconvulsions [6]. In other experimental studies, acute venlafaxine showed proconvulsant [32], anticonvulsant [30] or no significant action on seizures [1]. However, most experimental data argues for the anticonvulsant effect of venlafaxine. Therefore, it can be supposed that this antidepressant can be more advantageous than SSRIs in patients with co-existing epilepsy and depression. In clinical practice, venlafaxine induced seizures in overdose (1,300 mg), but it seems to be safe at therapeutic doses [45].

Among antidepressants with receptor mechanism of action, mirtazapine, blocking a₂, 5HT₂ and 5HT₃ receptors, did not affect either pentylenetetrazole- or maximal electroshock-induced convulsions in mice [47]. Mianserin, blocking a₁, a₂, and 5HT₂receptors, increased the electroconvulsive threshold after acute administration, but chronic treatment significantly decreased this parameter [3].

The exact mechanism of anticonvulsant action of antidepressant drugs is usually explained by an increase in serotonergic and/or noradrenergic neurotransmission [18, 19, 37, 44]. From this point of view, all representatives of SSRIs, SNRIs and trazodone, should be almost equally effective in all seizure models. In fact, available experimental data are rather

equivocal. Moreover, tianeptine, a drug decreasing serotonin concentration in the synaptic cleft, showed anticonvulsant activity against pentylenetetrazole- induced convulsions in mice [9]. This strongly suggests that other mechanisms must contribute to the anticonvulsant action of antidepressant drugs. In fact, purinergic, glutamatergic, and nitrergic mechanisms may be involved in the anticonvulsant properties of tianeptine [41]. Uzbay et al. [42] documented that the beneficial effect of tianeptine in pentylenetetrazole- induced seizures in mice is mediated by adenosine A₁ receptors. Recent studies showed that the antidepressant-like action of tianeptine and 3-[(2-methyl-4- thiazol-4-yl)ethynyl]-pyridine (MTEP), a selective antagonist of the metabotropic glutamate receptor subtype mGluR5, may be regulated by NMDA [25] or NMDA and AMPA [46] subtypes of ionotropic glutamatergic receptors. It should be remembered that both these receptors are strongly involved in seizure phenomena [29]. The anticonvulsant action of fluoxetine may be mediated by allopregnanolone, the potent inhibitory neurosteroid [40]. Other SSRIs can reduce seizure propagation by blocking T-, N-, and L-type Ca²⁺channels [10].

Choice of the proper antidepressant drug for epileptic patients should be based not only on its isolated activity in a battery of seizure models, but also on interactions with antiepileptic drugs. Since clinical data are very limited, our knowledge about such interactions comes from experimental studies, conducted mainly in the model of maximal electroshock in mice.

Antidepressant drugs with own anticonvulsant properties usually, but not always, potentiated the action of antiepileptic drugs. For instance, acute amitriptyline, desipramine, and imipramine enhanced the protective action of valproate [22]. Acute fluoxetine potentiated the anticonvulsant action of valproate, carbamazepine, phenobarbital, and phenytoin, while chronic fluoxetine enhanced that of valproate, carbamazepine, and phenytoin. However, most of these interactions had a pharmacokinetic character [5, 7].

Considering SNRIs, acute milnacipran enhanced the protective action of carbamazepine and phenobarbital, but chronic administration failed to affect the anticonvulsant efficacy of antiepileptics. The pharmacokinetic contribution to observed interactions was un- likely. Furthermore, both single and chronic application of venlafaxine potentiated the action of valproate without affecting brain concentrations of this antiepileptic.

(9)

Finally, although acute treatment with mianserin increased the effectiveness of valproate, phenytoin, and carbamazepine, chronic administration reduced the anticonvulsant action of these antiepileptics. In this case, all revealed interactions appeared to be pharmacodynamic [3]. In light of these observations, trazodone seems to possess quite different profile of action against electrically-induced seizures than other antidepressant drugs. It appears that the most compa- rable drug to trazodone is mianserin. It is worth stressing that both antidepressants act antagonistically towards 5HT₂, a₁and a₂ receptors. Both antidepressants administered chronically are able to decrease the protective activity of some antiepileptic drugs. In the case of mianserin, such properties may directly result from the proconvulsant action of the antidepressant.

In contrast, trazodone administered chronically showed per se the anticonvulsant action, but it signifi- cantly decreased the protective efficacy of phenytoin and carbamazepine. This strongly suggests that the observed effect of antidepressants in seizure models may be mediated by quite different mechanisms than in their interactions with antiepileptic drugs.

There are no available data on pharmacokinetics of trazodone in animals. In humans, the antidepressant is metabolized by CYP3A4 isozyme. Therefore, the serum concentration of trazodone is increased by CYP3A4 blockers (ritonavir, ketoconazole) and decreased by CYP3A4 inductors (carbamazepine) [11, 24]. In two case reports, it was indicated that trazodone increased the serum level of phenytoin [35]. In another one, the authors showed that trazodone raised the serum concentration of carbamazepine [31]. Re- sults of our study remain in conflict with these findings. Nevertheless, brain drug concentrations (the present study) and plasma drug levels (the cited case reports) may not be always similar. Also, it should be kept in mind that pharmacokinetic interactions in animals often do not coincide with those in humans.

Most of the previously tested antidepressant drugs, administered alone or in combination with antiepileptic drugs, did not produce motor and long-term memory deficits in mice [3–7]. Trazodone was not so beneficial in this respect. Chronic application of this drug impaired long-term memory, while acute and chronic trazodone co-administered with phenytoin induced both motor and memory deficits in tested animals.

Many experimental data on antidepressant drugs has been confirmed in clinical conditions. Several

drugs increasing the risk of seizures in humans [15]

showed also proconvulsant activity in animal seizure models. This applies, for instance, to bupropion [39], mianserin [3], and maprotiline [38]. On the other hand, fluoxetine, with an established anticonvulsant activity in several seizure models [5], improved also seizure control in humans [13].

In summary, despite anticonvulsant effects in the electroconvulsive threshold test, trazodone reduced the protective properties of phenytoin and carbamazepine against the maximal electroshock in mice and potentiated the adverse effects of phenytoin in the chimney and passive-avoidance tests. Additionally, the antidepressant easily interacted pharmacokineti- cally with antiepileptic drugs.

Conclusion

Taking into account all these findings, trazodone does not seem to be a good candidate for clinical trials in patients with epilepsy co-existing with depressive disorders.

Acknowledgment:

This study was supported by a grant from the Medical University of Lublin, Poland.

References:

1.Ahern TH, Javors MA, Eagles DA, Martillotti J, Mitchell HA, Liles LC, Weinshenker D: The effects of chronic norepinephrine transporter inactivation on seizure susceptibility. Neuropsychopharmacology, 2006, 31, 730–738.

2.Boissier JR, Tardy J, Diverres JC: Une nouvelle methode simple pour explorer l’action tranquilisante: le test de la cheminee. Med Exp (Basel), 1960, 3, 81–84.

3.Borowicz KK, Banach M, Zarczuk R, £ukasik D,

£uszczki JJ, Czuczwar SJ: Acute and chronic treatment with mianserin differentially affects the anticonvulsant activity of conventional antiepileptic drugs in the mouse maximal electroshock model. Psychopharmacology, 2007, 195, 167–174.

4.Borowicz KK, Furmanek-Karwowska K, Morawska M,

£uszczki JJ, Czuczwar SJ: Effect of acute and chronic treatment with milnacipran potentiates the anticonvulsant activity of conventional antiepileptic drugs in the maximal electroshock-induced seizures in mice. Psychophar- macology, 2010, 207, 661–669.

(10)

5.Borowicz KK, Furmanek-Karwowska K, Sawicka K,

£uszczki JJ, Czuczwar SJ: Chronically administered fluoxetine enhances the anticonvulsant activity of conventional antiepileptic drugs in the mouse maximal electroshock model. Eur J Pharmacol, 2007, 567, 77–82.

6.Borowicz KK, Go³yska D, £uszczki JJ, Czuczwar SJ:

Effect of acutely and chronically administered venlafaxine on the anticonvulsant action of classical antiepileptic drugs in the mouse maximal electroshock model. Eur J Pharmacol, 2011, 670, 114–120.

7.Borowicz KK, Stêpieñ K, Czuczwar SJ: Fluoxetine enhances the anticonvulsant effects of conventional antiepileptic drugs in maximal electroshock seizures in mice.

Pharmacol Rep, 2006, 58, 83–90.

8.Bowdan ND: Seizure possibly caused by trazodone HCl.

Am J Psychiatry, 1983, 140, 642–642.

9.Ceyhan M, Kayir H, Uzbay IT: Investigation of the effects of tianeptine and fluoxetine on pentylenetetrazole-induced seizures in rats. J Psychiatr Res, 2005, 39, 191–196.

10.Deak F, Lasztoczi B, Pacher P, Petheo GL, Kecskemeti V, Spat A: Inhibition of voltage-gated calcium channels by fluoxetine in rat hippocampal pyramidal cells. Neuro- pharmacology, 2000, 39, 1029–1036.

11.Dorn JM: A case of phenytoin toxicity possibly precipi- tated by trazodone. J Clin Psychiatry, 1986, 47, 89–90.

12.Duncan D, Taylor D: Which is the safest antidepressant to use in epilepsy? Psychiatric Bull, 1995, 19, 355–357.

13.Favale E, Rubino V, Mainardi P, Lunardi G, Albano C:

Anticonvulsant effect of fluoxetine in humans. Neurol- ogy, 1995, 45, 1926–1927.

14.Ghanbari R, El Mansari M, Blier P: Sustained administration of trazodone enhances serotonergic neurotransmission: in vivo electrophysiological study in the rat brain. J Pharmacol Exp Ther, 2010, 335, 197–206.

15.Harden CL, Goldstein MA: Mood disorders in patients with epilepsy: epidemiology and management. CNS Drugs, 2002, 16, 291–302.

16.Holmes GL: Animal model studies application to human patients. Neurology, 2007, 69, Suppl 3, S28–S32.

17.Janiri I, Hadjichristos A, Buonanno R: Adjuvant trazodone in the treatment of alcoholism: an open study.

Alcohol Alcohol, 1998, 33, 362–365.

18.Jobe PC: Common pathogenic mechanisms between depression and epilepsy: an experimental perspective.

Epilepsy Behav, 2003, 4, S14–S24.

19.Jobe CJ, Browning RA: The serotonergic and noradrenergic effects of antidepressant drugs are anticonvulsant, not proconvulsant. Epilepsy Behav, 2005, 7, 602–619.

20.Jobe PC, Dailey JW, Wernicke JF: A noradrenergic and serotonergic hypothesis of the linkage between epilepsy and affective disorders. Crit Rev Neurobiol, 1999, 13, 317–356.

21.Kinney GG, Griffith JC, Hudzik TJ: Antidepressant-like effects of 5-hydroxytryptamine 1A receptor agonists on operant responding under a response duration differentia- tion schedule. Behav Pharmacol, 1998, 9, 309–318.

22.Kleinrok Z, Gustaw J, Czuczwar SJ: Influence of antidepressant drugs on seizure susceptibility and the anticonvulsant activity of valproate in mice. J Neural Transm Suppl, 1991, 34, 85–90.

23.Lefkowitz D, Kilgo G, Lee S: Seizures and trazodone therapy. Arch Gen Psychiatry, 1985, 42, 523–523.

24.Litchfield JT, Wilcoxon F: A simplified method of evalu- ating dose-effect experiments. J Pharmacol Exp Ther, 1949, 96, 99–113.

25.Pomierny-Chamio³o L, Poleszak E, Pilc A, Nowak G:

NMDA but not AMPA glutamatergic receptors are involved in the antidepressant-like activity of MTEP during the forced swim test in mice. Pharmacol Rep, 2010, 62, 1186–1190.

26.Prapotnik M, Waschgler R, Konig P, Moll W, Conca A:

Therapeutic drug monitoring of trazodone: are there pharmacokinetic interactions involving citalopram and fluoxetine? Int J Clin Pharmacol Ther, 2004, 42, 120–124.

27.Raju SS, Noor AR, Gurthu S, Giriyappanavar CR, Acharya SB, Low HC, Quah SH: Effect of fluoxetine on maximal electroshock seizures in mice: acute vs chronic administration. Pharmacol Res, 1999, 39, 451–454.

28.Richelson E: Pharmacology of antidepressants – charac- teristics of the ideal drug. Mayo Clin Proc, 1994, 69, 1069–1081.

29.Salzberg MR, Vajda FJE: Epilepsy, depression and antidepressant drugs. J Clin Neurosci, 2001, 8, 209–215.

30.Saðlam E, Uzbay IT, Kayir H, Celik T, Beyazyürek M:

Effects of venlafaxine on ethanol withdrawal syndrome in rats. Fundam Clin Pharmacol, 2004, 18, 693–698.

31.Sánchez-Romero A, Mayordomo-Aranda A, García- Delgado R, Durán-Quintana JA: Probable interaction between trazodone and carbamazepine. Pharmacopsychia- try, 2011, 44, 158–159.

32.Santos JG Jr, Do Monte FH, Russi M, Augustine PE, Lanziotti VM: Proconvulsant effects of high doses of venlafaxine in pentylenetetrazole-convulsive rats.

Braz J Med Biol Res, 2002, 35, 469–472.

33.Seethalakshmi R, Krishnamoorthy ES: Depression in epilepsy: phenomenology, diagnosis and management.

Epileptic Disord, 2007, 9, 1–10.

34.Silvestrini B, Cioli V, Burberi S, Catanese B: Pharma- cological properties of AF 1161, a new psychotropic drug. Int J Neuropharmacol, 1968, 7, 587–599.

35.Spina E, Perucca E: Clinical significance of pharmacokinetic interactions between antiepileptic and psychotropic drugs. Epilepsia, 2002, 43, 37–44.

36.Stahl SM: Mechanism of action of trazodone: a multi- functional drug. CNS Spectr, 2009, 14, 536–546.

37.Statnick MA, Maring-Smith ML, Clough RW, Wang C, Dailey JW, Jobe PC, Browning RA: Effect of 5,7- dihydroxytryptamine on audiogenic seizures in geneti- cally epilepsy-prone rats. Life Sci, 1996, 69, 1763–1771.

38.Trimble MR, Meldrum BS, Anlezark G: Effect of nomifensine on brain amines and epilepsy in photosensi- tive baboons. Br J Clin Pharmacol, 1977, 4, Suppl 2, 101S–107S.

39.Tutka P, Mróz T, Klucha K, Piekarczyk M, Wielosz M:

Bupropion-induced convulsions: preclinical evaluation of antiepileptic drugs. Epilepsy Res, 2005, 64, 13–22.

40.Ugale RR, Mitkal N, Hirani K, Chopde CT: Essentiality of central GABAergic neuroactive steroid allopregnanolone for anticonvulsant action of fluoxetine against

(11)

pentylenetetrazole-induced seizures in mice. Brain Res, 2004, 1023, 102–111.

41.Uzbay IT: Serotonergic anti-depressants and ethanol withdrawal syndrome: a review. Alcohol Alcohol, 2008, 43, 15–25.

42.Uzbay TI, Kayir H, Ceyhan M: Effects of tianeptine on onset time of pentylenetetrazole-induced seizures in mice: possible role of adenosine A1 receptors. Neuro- psychopharmacology, 2007, 32, 412–416.

43.Venault P, Chapouthier G, De Carvalho LP, Simiand J, Morre M, Dodd RH, Rossier J: Benzodiazepine impairs and b-carboline enhances performance in learning and memory tasks. Nature, 1986, 321, 864–866.

44.Wang C, Mishra PK, Dailey JW, Jobe PC, Browning RA:

Noradrenergic terminal fields as determinants of seizure predisposition in GEPR-3s: a neuroanatomic assessment with intracerebral microinjections of 6-hydroxydopa- mine. Epilepsy Res, 1994, 18, 1–9.

45.White CM, Gailey RA, Levin GM, Smith T: Seizure re- sulting from a venlafaxine overdose. Ann Pharmacother, 1997, 31, 178–180.

46.WlaŸ P, Kasperek R, WlaŸ A, Szumi³o M, Wróbel A, Nowak G, Poleszak E: NMDA and AMPA receptors are involved in the antidepressant-like activity of tianeptine in the forced swim test in mice. Pharmacol Rep, 2011, 63, 1526–1532.

47.Yilmaz I, Sezer Z, Kayir H, Uzbay TI: Mirtazapine does not affect pentylenetetrazole- and maximal electroconvulsive shock-induced seizures in mice. Epilepsy Beh, 2007, 11, 1–5.

Received: October 16, 2011; in the revised form: April 14, 2012;

accepted: June 8, 2012.