- Email: [email protected]
Diltiazen enhances and flunarizine inhibits nimodipine's antiseizure effects
European Journal of Pharmacology, 163 (1989) 299-307
299
Elsevier EJP 50741
Diltiazem enhances and flunarizine inhibits nimodipine's antiseizure effects M.A. M o r 6 n *, C.W. Stevens a n d T.L. Y a k s h Departments of Pharmacology and Neurologic Surgery, Mayo Clinic, Rochester, MN 55905, U.S.A. Received 3 January 1989, accepted 17 January 1989
The dihydropyridine calcium channel antagonist, nimodipine has antiepileptic and anticonvulsive properties that are thought to be mediated through neuronal calcium channel blockade. The dihydropyridine binding site can be positively and negatively allosterically regulated by the benzothiazepines and the phenylalkylamines/piperazines, respectively. We investigated this binding interaction at the physiologic level by examining the effects of diltiazem (a benzothiazepine) and flunarizine (a piperazine) on the antiseizure activity of nimodipine. Seizures were induced with pentylenetetrazole in awake rats with chronically implanted EEG electrodes. Calcium channel antagonists were administered intracerebroventricularly 30 min after pentylenetetrazole at doses given at 15 min intervals. Diltiazem and flunarizine alone lacked antiseizure properties. The calculated EDs0 values for nimodipine were: nimodipine alone = 135 fig; nimodipine + diltiazem (100 #g) = 67 t~g. Nimodipine + flunarizine (10 t~g) completely suppressed nimodipine's antiseizure activity. These findings may reflect the interaction observed among these agents at binding sites associated with the calcium channel and supports the idea that dihydropyridines mediate their antiseizure actions through neuronal calcium channel antagonism. Nimodipine; Flunarizine; Diltiazem; Ca 2+ channel antagonists; Seizures; Dihydropyridines
1. Introduction
The antiepileptic and anticonvulsive properties of organic compounds known as calcium channel antagonists have been demonstrated in various seizure models (Walden et al., 1985; Meyer et al., 1986; Morocutti et al., 1986; Ots et al., 1987). There are four classes of calcium channel antagonists: (1) the dihydropyridines (nimodipine, nitrendipine); (2) the benzothiazepines (diltiazem); (3) the phenylalkylamines (verapamil, D600); and (4) the piperazines (flunarizine, cinnarizine). Currently it is thought that these agents mediate their antiseizure effects through neuronal calcium channel blockade; however, no evidence substantiating this mechanism has been reported. For the calcium
* To whom all correspondence should be addressed: Mayo Clinic, Rochester, MN 55905, U.S.A.
channel antagonists, an ordering of antiseizure activity similar to that provided by binding studies would provide corollary evidence that the calcium channel antagonists exert their antiseizure activity by binding to sites that regulate neuronal calcium channels. Binding studies have identified three distinct allosterically linked binding sites for the first three classes of calcium channel antagonists listed above (Glossmann et al., 1985). The fourth class, the piperazines, utilize the phenylalkylamine binding site (Murphy et al., 1984). The allosteric interactions between these sites are such that the benzothiazepines increase dihydropyridine binding affinity (see Murphy et al., 1984; Schoemaker and Langer, 1985). This relationship is stereospecific for d-cis-diltiazem, the active isomer for calcium channel blockade (DePover et al., 1982; Ferry and Glossmann, 1983; Goll et al., 1983). A reflection of this binding interaction at the physiologic level has been re-
0014-2999/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
300
ported in cardiac tissue (DePover et al., 1983) in which nimodipine's negative ionotropic effects upon perfused rat hearts was potentiated by diltiazem. In contrast, the allosteric interaction between the dihydropyridines and the phenylalkylamines or piperazines is a negative allosteric interaction (Ehlert et al., 1982a,b; Bolger et al., 1983; Janis et al., 1984; Murphy et al., 1984; Zobrist et al., 1986). This negative binding interaction has been best demonstrated in cardiac and nervous tissue (Ehlert et al., 1982a,b), however functional antagonism of dihydropyridines actions by either the phenylalkylamines or the piperazines has not been previously reported. In this investigation we examined at the physiologic level the hypothesis that the dihydropyridine calcium channel antagonists exert their antiseizure actions via neuronal calcium channel blockade by examining the positive and negative dihydropyridine binding interactions discussed above. The calcium channel antagonists utilized in this investigation are nimodipine (a dihydropyridine), diltiazem (a benzothiazepine) and flunarizine (a piperazine). Nimodipine and flunarizine have been shown to have antiseizure activity (nimodipine: Meyer et al., 1986; flunarizine: Desmedt et al., 1975) while diltiazem lacks antiseizure activity (Wauquier et al., 1985). From this investigation we report that centrally administered diltiazem enhances and flunarizine inhibits the antiseizure action of nimodipine suggesting that this result may be a reflection of allosteric binding interactions occurring at neuronal calcium channel binding sites.
2. Materials and methods
2.1. Animal preparation Male Sprague-Dawley rats (250-350 g) were surgically i m p l a n t e d u n d e r 2% h a l o t h a n e anesthesia with epidural electrodes and an intraventricular (i.c.v.) cannula. Following anesthetic induction the animal was placed in a stereotaxic head holder and the head shaved and swabbed with betadine. The skull was exposed and the underlying periosteum scraped away.
Cranial burr holes were made for three epidural E E G electrodes in the following locations: 2 mm rostral to the coronal suture and 2 mm lateral to midline, one on each side (hemispheric leads). The third burr hole was placed 2 mm rostral to the lambdoid suture and 2 mm lateral to the sagittal suture on the right side of the skull (cerebellar lead). The burr hole for the i.c.v, cannula was made 1.5 mm caudal to the coronal suture and 1.5 mm lateral to the midline on the right side. Three stainless steel screws (self-tapping screws 00 × 3 / 1 6 inches; Small Parts, Inc., Miami, FL) were placed in a tripod arrangement followed by EEG electrode implacement. The screws provided anchorage of the electrode housing to the skull surface. A 21 gauge bevel-tipped cannula (1.2 cm in length) was stereotaxically placed into the fight lateral cerebroventricle (5 mm in from the dura). After use, each rat was killed and cannula iraplacement verified by i.c.v, injection of trypan blue dye. The epidural E E G recordings were obtained from electrodes made from stainless steel wire (28 gauge). Construction of the EEG electrodes is as follows. Stainless steel wires were soldered into female crimp gold plated pins. The pin portion of the wire-pin combination was inserted into a microminiature strip connector (Allied Electronics, For Worth, TX). The strip connector was cut such that three electrode leads could be accommodated. The tip of the wire was bent so a foot process was created. The exposed stainless steel wire was covered with acrylic, except for the last mm of the foot. The electrode array was secured to the skull with dental cement placed around the EEG electrode housing and the i.c.v, cannula. Dental cement adhesion between the EEG housing and the stainless steel screws provided additional stabilization of the electrode array. Each animal was allowed at least two days recovery before entering a study group. Implanted rats were housed individually, allowed free access to food and water, and placed on a 12 h light-dark cycle.
2.2. Drugs The calcium channel antagonists were administered i.c.v. (total volume = 10 ffl). The calcium
301 channel antagonists, except diltiazem, were dissolved in a vehicle consisting of 25% ethanol and 75% polyethylene glycol (PEG 400) by volume. Diltiazem (30, 100 and 300 /~g; Sigma Chemical Co., St. Louis, MO) was dissolved in normal saline. The other two calcium channel antagonists were nimodipine (30, 100 and 300/~g; gift from Dr. A. Scriabine, Miles Pharmaceutical Co., New Haven, CT) and flunarizine (1, 30 and 100 ~tg; Sigma Chemical Co., St. Louis, MO). Because nimodipine is light sensitive, care was taken to prevent its exposure to light by wrapping the solution vials in foil and darkening the room. Seizures were induced by intraperitoneal (i.p.) administration of pentylenetetrazole (35 m g / k g i.p.; Sigma Chemical Co., St. Louis, MO) dissolved in normal saline.
over the cerebellum. Rats were placed in 40 cm diameter, clear plexiglass cylinders that allowed free movement without undue stress on the EEG wires. The E E G was monitored via an oscillograph ( T y p e R564B Storage Oscilliscope, Tectronix) after amplification with an AC amplifier (EEG amplifier, Grass model 7P511; 10 and 100 Hz low and high pass filter settings; 200 /~V/cm). EEG activity was quantitated by electronic spike counting. EEG spikes were defined as waveforms with a frequency greater than 10 Hz and an amplitude greater than three times that of background (preseizure) activity. These spikes were fed into a data acquisition board and stored as spikes per minute in an electronic buffer until the end of the experiment. Permanent storage was made by downloading the buffers onto floppy discs via an IBM-PCXT computer.
2.3. Treatment protocol 2.5. Data analysis Each rat was employed only once in these studies. The animals were administered pentylenetetrazole (35 m g / k g i.p.) and seizure and convulsive activity allowed to progress over the ensuing 30 rain. Typically, at 30 min post-pentylenetetrazole, a consistent spiking pattern was observed on the EEG. For each drug three log-spaced doses were given in a cumulative dosing regimen. The lowest dose of calcium channel antagonist was administered i.c.v. 30 min after pentylenetetrazole, followed 15 min later by the next dose of calcium channel antagonist and at 30 rain the highest dose of calcium channel antagonist. After the last dose the animals were observed both visually and via the electroencephalogram for the subsequent 60 min. This method of drug administration allowed for the generation of a dose-response curve for each animal. In two separate groups, a constant repeated dose of diltiazem (100 ~tg) or flunarizine (10 /~g) was administered with increasing doses of nimodipine. Repeated vehicle injections over a similar time course were found to have no effect on seizure activity.
Data analysis was facilitated by using a basic program (SPIKE.BAS, available from author, C.W. Stevens) which calculated the following: the average baseline spikes per minute, determined from the 10 min period preceding the first calcium channel antagonist administration; and the average spikes per minute for the 10 min periods following all three doses of calcium channel antagonist, these latter values were compared to the baseline value and reported as percent inhibition of spiking. For statistical analysis the percent inhibition values for each rat were grouped and compared to the baseline values by one-way ANOVA. Cumulative dose-response curves were analyzed to give EDs0 ( + 95 % confidence limits), slopes and regression values via a software package (Tallarida and Murray, 1986). Significance was set at P < 0.05.
3. Results
3.1. Vehicle injection 2.4. EEG monitoring EEG activity was monitored bipolar between the two hemispheres with respect to the third lead
As a control group (n = 14) three vehicle i.c.v. injections (10 /~l/injection) were given at 30, 45 and 60 min after pentylenetetrazole. Multiple
302 Time After PTZ
PTZ BL
3O
35
V
50
V
65
V
V
Vehicle
Time
i,.,,,,.,,i
.................................
iI
NM dose
,,,,,
li ......
' .............
i,J ........
30
; ...........
100
V
..........
I,,,
iIiiIh
300
V
V
r
NM -t- FLZ (10)
i
r
~
Fig. 1. EEG activity (2 min segments). Top set of tracings show the EEG of pentylenetetrazole-treated rats receiving multiple vehicle injections via i.c.v, administration at 35, 50 and 65 min after pentylenetetrazole. Nimodipine alone (2nd set of tracings) reduces EEG spiking at 100 and 300/~g i.c.v. Nimodipine + diltiazem (100 fig) (3rd set of tracings) causes a reduction in EEG spiking for all three doses of nimodipine. The bottom set of tracings show flunarizine's (10 fig) ability to completely suppress nimodipine's antiseizure activity at all doses of nimodipine examined. BL: baseline, pre-pentylenetetrazole; V: vehicle/drug injection).
Spikes/minute
Spikes/minute *
1501
150-
NM Dose (/~g)
30
2001
200.
:
100 300
NM injection
*
*
Time(rnin)
32
45 65
-k
100. 50-
0-
0
2s
so
7s
loo
12s
1so
Time ( m i n ) Fig. 2. Multiple vehicle i.c.v, injections. EEG spikes per minute versus time for a typical single rat. Multiple vehicle injections (36, 49 and 63 rain injection times in this example) had no effect on pentylenetetrazole-induced spiking activity (* vehicle injection).
0
25
50
75 100 Time (minutes)
125
150
Fig. 3. Nimodipine alone, EEG spikes per minute vs. time. Increasing doses of nimodipine (30, 100 and 300 fig i.c.v.) injected at 32, 45 and 65 min. The two highest nimodipine doses significantly reduced (P < 0.05) spiking activity (* nimodipine injection).
303
vehicle injections did not alter pentylenetetrazoleinduced seizure activity time course as noted in the EEG (fig. 1, top tracing) and spikes per minute values (fig. 2). The average baseline spikes per minute value (determined from the 10 rain period preceding the first vehicle injection) was not statistically different from the average spikes per minute values for the 10 rain period following each vehicle administration.
Spikes/minute
NM Dose (~g)
Time (min)
30
200 "Jr: NM injection 150
39
100
55
300
68
"Jr
OO
50.
3.2. Single calcium channel antagonists O'
I.c.v. administered nimodipine at three doses (30, 100 and 300 #g) produced a dose-dependent reduction in seizure activity (fig. 1, 2nd tracing) and a reduction in spikes per minute (fig. 3) from which a dose-response curve was generate (fig. 4). The EDs0 and 95% confidence limits calculated for nimodipine was 135.3 /zg (108.7-168.4). The other two calcium channel antagonists, diltiazem and flunarizine, did not cause a reduction in seizure activity when administered alone (fig. 4). Diltiazero was examined at doses of 30, 100 and 300 t~g i.c.v.; while flunarizine was examined at 1, 30 and 100 #g i.c.v. Higher doses of flunarizine were not examined because of toxicity. Thus, flunarizine at 300 t~g i.c.v, in three rats to death within 30 min preceded by respiratory symptoms that included gasping and frothing.
Inhibition of Spiking 1000--0 NM A--A
0
50
25
75
1O0
125
150
Time (minutes) Fig. 5. l.c.v, nimodipine in combination with a constant dose of diltiazem (100 #g). Spikes per minute versus time. A low dose of nimodipine in combination with diltiazem transiently reduces spiking activity, an effect not seen when nimodipine is administered alone at this dose. Higher doses of nimodipine in combination with diltiazem suppress spiking activity to levels observed with nimodipine alone. In this example nimodipine was administered at 39, 55 and 69 min (* injection).
3.3. Coadministration of diltiazem and flunarizine with increasing nimodipine concentrations A constant dose of diltiazem (100 /zg) was coadministered with increasing doses of nimodipine ( n = 12). This combination resulted in a greater reduction of EEG seizure activity (fig. 1, 3rd tracing) than nimodipine alone. This was especially prominent at the lowest dose of nimodipine TABLE 1
DL I ~~
Summary of the dose-response analysis of the antiseizure effects of nimodipine, dihiazem and flunarizine alone and in combination given i.c.v.
II--II fkZ 50-
Agent
Slope of dose-response
EDs0 (95% CL)
N
curve
-50 10
100
1000
Dose (/zcj i.c.v) Fig. 4. Dose-response curve for nimodipine, diltiazem and flunarizine. Percent inhibition of spiking activity versus dose. Administered alone, only nimodipine reduced spiking in a dose-dependent fashion. Each curve represents mean + S.E. for 8 or 12 animals per drug.
Nimodipine Nimodipine + diltiazem a Nimodipine +flunarizine a Diltiazem Flunarizine
119.2+13.4
135.3 (109-168)
66.8 + 12.5
66.9 (52-87)
_ b _ b
> 300 c > 300 c
_
•
b
300
c
12 12 12 8 8
a Dose-response curves for nimodipine in combination with either 100 ~ g diltiazem or 10 /~g flunarizine, b Slope not different from zero. c EDs0 not calculated due to zero slope.
304
(30 ~g) at which nimodipine alone was ineffective. This EEG seizure reduction was reflected in the spike counting (fig. 5). The ED50 and 95% confident limit values for nimodipine in combination with a constant dose of diltiazem was 67/~g (52-87), which is significantly less (P < 0.05) than nimodipine alone (see table 1). In a separate group of rats (n = 12), a constant flunarizine dose (10 ~g i.c.v.) was coadministered with increasing doses of nimodipine. This combination resulted in a complete inhibition of nimodipine's antiseizure activity as reflected on the EEG (fig. 1, bottom tracing) and the spikes per minute.
3.4. Behavior An advantage of this seizure model (pentylenetetrazole-induced seizures in conscious rats with chronically implanted EEG electrodes) is that behavior can be monitored. The dose or pentylenetetrazole (35 m g / k g i.p.) utilized in this study produced several tonic-clinic fits occurring only within the first 20 min after pentylenetetrazole administration. Subsequently, while EEG activity displayed frequent repetitive bursting, motor seizures were not observed. However, in all rats receiving pentylenetetrazole, characteristic in-
Inhibition of Spiking 100&--& • --"
so
-50
0
+ DIL(IO0) ~ - / NM + F L Z ( I ~ / / T ~ NM
T~ ~ "
1O0 Dose (/~g ICV)
1000
Fig. 6. Dose-response curve for nimodipine alone and in combination with diltiazem and flunarizine. Nimodipine + diltiazem (100 ~g) significantly shifts (P < 0.05) the low dose portion of the curve to the left, reflecting the EEG observations. Flunarizine completely suppresses nimodipine inhibition of spiking. Each curve represents mean _+S.E. of 12 animals per drug.
creases in levels of motor activity and augmented startled responses to auditory and tactile stimuli were observed. Though not systematically quantified, administration of diltiazem and flunarizine alone did not alter these behavior changes induced by pentylenetetrazole. Administration of nimodipine at 100 and 300 /xg i.c.v, reversed pentylenetetrazole-induced excitatory behavior and led to a state of relative behavioral depression, i.e. decreased motor activity and diminished response to auditory and tactile stimuli. When administered alone, the lowest dose of nimodipine (30 ~g) examined did not produce behavioral depression, however when this low dose is coadministered with diltiazem behavioral depression resulted (fig. 6). Coadministration of flunarizine with nimodipine resulted in behavioral depression only at the two highest doses of nimodipine (100 and 300/~g i.c.v.), a finding similar to administration of nimodipine alone at these two doses.
4. Discussion
4.1. Dihydropyridine effects following i.c.v, administration." modulation by diltiazem and flunarizine The dihydropyridine calcium channel antagonists bind to specific and high affinity sites in nervous tissue thought to be associated with calcium channels (Miller and Freedman, 1984). This dihydropyridine binding can be positively and negatively allosterically modulated by the benzothiazepines (diltiazem) and the phenylalkylamines/piperazines (verapamil/flunarizine), respectively. The dihydropyridine class of calcium channel antagonists are thought to mediate their antiseizure effects by binding to the calcium channel associated binding sites. An attempt to associate these binding interactions to a physiological correlate was made in this investigation by coadministering diltiazem and flunarizine with nimodipine to determine their effects on nimodipine's antiseizure capabilities against pentylenetetrazoleinduced seizures in the conscious rat. Nimodipine alone was shown to reduce seizure activity in a dose-dependent fashion (fig. 4) and this activity was significantly enhanced through coadministra-
305 tion with diltiazem (fig. 6). If the dihydropyridines do indeed mediate their antiseizure actions through binding to highly specific sites, the affinity for which is positively enhanced by diltiazem, then this finding of enhanced antiseizure activity may be a reflection of this binding interaction. Conversely, flunarizine coadministration completely suppressed nimodipine's antiseizure activity, and is thus consistent with the negative interaction of flunafizine on nimodipine binding. These related findings suggest that the dihydropyridine calcium channel antagonists mediate their antiseizure activity through interaction with a site similar to that defined by the binding studies. It should be stressed that although these studies provide evidence that the dihydropyridines mediate their antiseizure actions through neuronal calcium channel blockade, they do not rule out the possibility that these agents may be acting through more than one mechanism. The dihydropyridine calcium channel antagonists have been reported to possess actions other than calcium channel blockade, these include inhibition of adenosine uptake (Murphy and Snyder, 1982; Wu et al., 1983; Marangos et al., 1984; Phillis et al., 1984), and inhibition of the calcium dependent regulatory protein calmodulin (Bostrom et al., 1981; Johnson et al., 1982). Inhibition of adenosine reuptake would cause increased extracellular adenosine. Adenosine is a potent anticonvulsant (Dunwiddie and Worth, 1981; Barraco et al., 1984) and is thought by some to be the brain's endogenous anticonvulsant (Dragunow, 1985). 4.2. Considerations of the pentylenetetrazole i.c.v. model The present studies employed the i.c.v, route of administration to systematically assess drug effects. Repeated vehicle injections were without effect on pentylenetetrazole seizures, indicating the lack of a cumulative injection artifact. The use of the i.c.v, route assumes that the site of drug action is sufficiently close to the ventricular lumen for the drug to reach. The present studies clearly display a dose-dependent effect of nimodipine. Similar results have been recently observed with this model for a wide range of dihydropyridine
analogues (Moron et al., 1988). The failure of diltiazem or flunarizine to exert antiseizure activity would not appear to be due to simply drug access. Indeed, an i.c.v, route of administration diminished the role played by peripheral metabolism, plasma protein binding and drug brain penetration. Moreover, as the drugs were coadministered, and the dose-response curves reflect the maximum effect and not seizure duration, it is unlikely that the pronounced effects observed with these agents could be accounted for by differential clearance. 4.3. Diltiazem's and flunarizine's lack of antiseizure activity In our investigations diltiazem alone was found to lack antiseizure activity although when coadministered with nimodipine it enhanced nimodipine's antiseizure activity. Wauquier et al. (1985) reported diltiazem's lack of protection against bicuculline induced convulsions in rats. Additionally, Ascioti et al. (1986) reported only weak antiseizure activity of diltiazem against audiogenic induced seizures in D B A / 2 mice. Both of these previous investigations utilized systemic routes of administration of diltiazem; and both investigators suggested that diltiazem's lack of antiseizure activity may be due to poor CNS penetrance. However, in this investigation we report the lack of antiseizure activity for diltiazem after central (i.c.v.) administration. Though flunarizine has been reported to be active after systemic administration in a variety of seizure models (Desmedt et al., 1975; Ashton and Wauquier, 1979; Wauquier et al., 1979; Lee et al., 1986) and in man (Declerck and Wauquier, 1983; Overweg et al., 1984), we failed to observe activity even at doses which were toxic. We do not know the reason for this apparent lack of effect. It is possible that, when administered systematically, flunarizine exerted its actions at sites different from those acted upon by nimodipine, lying more distant from the ventricles. Another possible explanation for the lack of antiseizure activity by flunarizine in this model may be due to its slow onset of action. Others (Desmedt et al., 1975; Wauquier et al., 1979; Drago et al., 1986) have noted this delayed onset of action
306 r e q u i r i n g 2 h or m o r e a f t e r a d m i n i s t r a t i o n b e f o r e a n t i s e i z u r e a c t i v i t y is n o t e d . I n these studies t h e total e x p e r i m e n t a l p e r i o d l a s t e d less t h a n 2 h, t h u s s u g g e s t i n g a n e x p l a n a t i o n for l a c k of effect. F l u n a r i z i n e ' s d e l a y e d o n s e t of a c t i o n is a s s o c i a t e d w i t h its l o n g d u r a t i o n of action, D e s m e d t et al. (1975) n o t e d seizure p r o t e c t i o n u p to 72 h a f t e r a single f l u n a r i z i n e dose.
4.4. Conclusion I n s u m m a r y , t h e effects of a g e n t s t h a t p o s i tively and n e g a t i v e l y r e g u l a t e n i m o d i p i n e b i n d i n g w e r e e x a m i n e d for effects u p o n n i m o d i p i n e ' s a n t i seizure a c t i v i t y d u r i n g c o a d m i n i s t r a t i o n w i t h nimodipine. Diltiazem, which increases nimodipine binding, enhanced nimodipine's antiseizure activity. I n c o n t r a s t , f l u n a r i z i n e , w h i c h r e d u c e s n i m o d i p i n e b i n d i n g , s u p p r e s s e s n i m o d i p i n e ' s antiseizure activity. T h e s e f i n d i n g s are c o n s i s t e n t w i t h the h y p o t h e s i s t h a t n i m o d i p i n e (a d i h y d r o p y r i d i n e c a l c i u m c h a n n e l a n t a g o n i s t ) exerts its a n t i s e i z u r e action, at least in part, t h r o u g h i n t e r a c t i o n s w i t h sites classified b y b i n d i n g studies to b e a s s o c i a t e d with a calcium channel.
Acknowledgement Supported in part by Grant R01 NS24329.
References Ascioti, C., G.B. De Sarro, B.S. Meldrum and G. Nistico, 1986, Calcium entry blockers as anticonvulsant drugs in DBA/2 mice, Br. J. Pharmacol. 88, 379p. Ashton, D. and A. Wauquier, 1979, Behavioural analysis of the effects of 15 anticonvulsants in the amygdaloid kindled rat, Psychopharmacology 65, 7. Barraco, R.A., T.H. Swanson, J.W. Phillis and R.E. Berman, 1984, Anticonvulsive effects of adenosine analogues on amygdaloid-kindled seizures in rats, Neurosci. Lett. 46, 317. Bolger, G.T., P. Gengo, R. Klockowski, E. Luchowski, H. Siegel, R.A. Janis, A.M. Triggle and D.J. Triggle, 1983, Characterization of binding of the C a 2 + channel antagonists (3H)nitrendipine, to guinea pig ileal smooth muscle, J. Pharmacol. Exp. Ther. 225, 291. Bostrom, S.L., B, Ljung, S. Mardh, S. Forsen and E. Thulin, 1981, Interactions of antihypertensive drug felodipine with calmodulin, Nature 292, 777.
Declerck, A.C. and A. Wauquier, 1983, Double blind study of the effectiveness of flunarizine in therapy resistant epilepsy in mentally retarded children, Boll. Chim. Farm. 122, 9. DePover, A., I.L. Grupp, G. Grupp and A. Schwartz, 1983, Diltiazem potentiates the negative inotropic action of nimodipine in the heart, Biochem. Biophys. Res. Commun. 114, 922. DePover, A., M.A. Matlib, S.W. Lee, G.P. Dube, I.L. Grupp, G. Grupp and A. Schwartz, 1982, Specific binding of (3H)nitrendipine to membranes from coronary arteries and heart in relation to pharmacological effects. Paradoxical stimulation by diltiazem, Biochem. Biophys. Res. Commun. 108, 110. Desmedt, L.K.C., C.J.E. Niemegeers and P.A.J. Janssen, 1975, Anticonvulsant properties of cinnarizine and flunarizine in rats and mice, Arzneim. Forsch. (Drug Res.) 25, 1408. Drago, F., C. Valerio, G. Clementi and U. Scapagini, 1986, Effects of flunarizine on experimentally induced convulsions in animals, Funct. Neurol. 1,529. Dragunow, M., 1985, Is adenosine an endogenous anticonvulsant?, Epilepsia 26, 480. Dunwiddie, T.V. and T. Worth, 1982, Sedative and anticonvulsive effects of adenosine analogues in the mouse and rat, J. Pharmacol. Exp. Ther. 220, 70. Ehlert, F.J., E. Itago, W.R. Roeske and H.I. Yamamura, 1982a, The interaction of (3H)nitrendipine with receptors for calcium antagonists in the cerebral cortex and heart and rats, Biochem. Biophys. Res. Commun. 104, 937. Ehlert, F.J., W.R. Roeske, E. Itagon and H.I. Yamamura, 1982b, The binding of (3H)nitrendipine to receptors for calcium channel antagonists in the heart, cerebral cortex, and ileum of rats, Life Sci. 30, 2191. Ferry, D.R. and H. Glossmann, 1983, Tissue-specific regulation of (3 H)nimodipine binding to putative calcium channels by the biologically active isomer of diltiazem, Br. J. Pharmacol. 78, 81p. Glossmann, H., D.R. Ferry, A. Goll, J. Stiessing and M. Schober, 1985, Calcium channels: basic properties revealed by radioligand binding studies, J. Cardiol. Pharmacol. 7 (Suppl. 16), $20. Goll, A., D.R. Ferry and H. Glossmann, 1983, Target size analysis of skeletal muscle channels: positive allosteric heterotrophic regulation by d-cis-diltiazem is associated with apparent channel oligmer dissociation, FEBS Lett. 157, 63. Janis, R.A., J.G. Sarmiento, S.G. Maurer, G.T. Bolger and D.J. Triggle, 1984, Characteristics of the binding of (3H)nitrendipine to rabbit ventricular membranes: modification by other calcium channel antagonists and by the Ca ++ channel agonist Bay k 8644, J. Pharmacol. Exp. Ther. 231, 8. Johnson, J.D., P.L. Vaghy, T.H. Crouch, J.D. Potter and A. Schwartz, 1983, A hypothesis for the mechanism of action of some of the Ca ++ antagonistic drugs: calmodulin as a receptor, in: Advances in Pharmacology and Therapeutics II, Vol. 3, eds. H. Yoshida, Y. Hagihara and S. Ebashi (Pergamon, Oxford) p. 121. Lee, S., H. Kawawaki, O. Matsuoka and R. Murata, 1986, Effect of Ca-antagonist (flunarizine) on kindling seizures in suckling rats, No To Hattatsu 18, 292.
307 Marangos, P.J., M.S. Finkel, A. Verma, M.F. Maturi, J. Patel and R.E. Patterson, 1984, Adenosine uptake sites in dog heart and brain: interaction with calcium channel antagonists, Life Sci. 35, 1109. Meyer, F.B., R.E. Anderson, T.M. Sundt, Jr. and F.W. Sharbrough, 1986, Selective central nervous system calcium channel blockers - a new class of anticonvulsant agents, Mayo Clin. Proc. 61, 239. Miller, R.J. and S.B. Freedman, 1984, Are dihydropyridine binding sites voltage sensitive calcium channels, Life Sci. 34, 1205. Morocutti, C., F. Pierelli, L. Sanarelli, E. Stefano, A. Peppe and G.L. Mattioli, 1986, Antiepileptic effects of calcium antagonist (nimodipine) on cefazolin-induced epileptogenic foci in rabbits, Epilepsia 27, 498. Mor6n, M.A., T.L. Yaksh and C.W. Stevens, 1988, The antiepileptic activity of eight dihydropyridine calcium channel antagonists: mechanism of action, Pharmacologist 30, A120. Murphy, K.M.M., R.J. Gould and S.H. Snyder, 1984, Regulation of (3H)nitrendipine binding: a single allosteric site for verapamil, diltiazem and prenylamine, in: Nitrendipine, eds. A. Scriabine, S. Vanov and K. Deck (Urban and Schwarzenberg, Baltimore) p. 107. Murphy, K.M.M. and S.H. Snyder, 1982, Adenosine receptor binding and specific receptors for calcium channel drugs, in: Calcium Entry Blockers, Adenosine, and Neurohumors: Advances in Coronary Vascular and Cardiac Control, eds. G.F. Merrill and H.R. Weiss (Urban and Schwarzenberg, Baltimore) p. 295. Ots, M.E., T.L. Yaksh, R.E. Anderson and T.M. Sundt, Jr., 1987, Effect of dihydropyridines and diphenylalkylamines on pentylenetetrazole-induced seizures and cerebral blood flow in cats, J. Neurosurg. 67, 406.
Overweg, J., C.D. Binnie, J.W.A. Meijer, H. Meinardi, S.T.M. Nuijten, S. Schmaltz and A. Wauquier, 1984, Double-blind placebo controlled trial of flunarizine as add on therapy in epilepsy, Epilepsia 25, 217. Phillis, J.W., T.H. Swanson and R.A. Barraco, 1984. Interactions between adenosine and nifedipine in the rat cerebral cortex. Neurochem. Int. 6, 693. Schoemaker, H. and S.Z. Langer, 1985, [3H]Diltiazem binding to calcium channel antagonists recognition sites in the rat cerebral cortex, European J. Pharmacol. 111. 273. Tallarida R.J. and R.B. Murray, 1984, Graded dose-response, in: Manual of Pharmacological Calculations with Computer Programs (Springer-Verlag, New York, NY) p. 82. Walden, J., E.J. Speckmann and O.W. Witte, 1985, Suppression of focal epileptiform discharges by intraventricular perfusion of a calcium antagonist, Electroencephalogr. Clin. Neurophysiol. 61,299. Wauquier, A., D. Ashton and W. Melis, 1979, Behavioural analysis of amygdaloid kindling in beagle dogs and the effects of clonazepam, diazepam, phenobarbital, diphenylhyndantoin and flunarizine on seizure manifestation, Exp. Neurol. 64, 579. Wauquier, A., D. Ashton, G. Clincke and J. Fransen, 1985, Calcium entry blockers as cerebral protecting agents: comparative activity in tests of hypoxia and hyperexcitability, Jap. J. Pharmacol. 38, 1. Wu, J.H., J.W. Phillis and V.L. Coffin, 1983, Calmodulin antagonists inhibit adenosine uptake by rat brain cortical synaptosomes, Neurosci. Lett. 37, 187. Zobrist, R.H., K.M. Giacomini, W.L. Nelson and J.C. Giacomini, 1986, The interaction of phenylalkylamine calcium channel blockers with the 1,4-dihydropyridine binding site, J. Mol. Cell. Cardiol. 18, 963.
299
Elsevier EJP 50741
Diltiazem enhances and flunarizine inhibits nimodipine's antiseizure effects M.A. M o r 6 n *, C.W. Stevens a n d T.L. Y a k s h Departments of Pharmacology and Neurologic Surgery, Mayo Clinic, Rochester, MN 55905, U.S.A. Received 3 January 1989, accepted 17 January 1989
The dihydropyridine calcium channel antagonist, nimodipine has antiepileptic and anticonvulsive properties that are thought to be mediated through neuronal calcium channel blockade. The dihydropyridine binding site can be positively and negatively allosterically regulated by the benzothiazepines and the phenylalkylamines/piperazines, respectively. We investigated this binding interaction at the physiologic level by examining the effects of diltiazem (a benzothiazepine) and flunarizine (a piperazine) on the antiseizure activity of nimodipine. Seizures were induced with pentylenetetrazole in awake rats with chronically implanted EEG electrodes. Calcium channel antagonists were administered intracerebroventricularly 30 min after pentylenetetrazole at doses given at 15 min intervals. Diltiazem and flunarizine alone lacked antiseizure properties. The calculated EDs0 values for nimodipine were: nimodipine alone = 135 fig; nimodipine + diltiazem (100 #g) = 67 t~g. Nimodipine + flunarizine (10 t~g) completely suppressed nimodipine's antiseizure activity. These findings may reflect the interaction observed among these agents at binding sites associated with the calcium channel and supports the idea that dihydropyridines mediate their antiseizure actions through neuronal calcium channel antagonism. Nimodipine; Flunarizine; Diltiazem; Ca 2+ channel antagonists; Seizures; Dihydropyridines
1. Introduction
The antiepileptic and anticonvulsive properties of organic compounds known as calcium channel antagonists have been demonstrated in various seizure models (Walden et al., 1985; Meyer et al., 1986; Morocutti et al., 1986; Ots et al., 1987). There are four classes of calcium channel antagonists: (1) the dihydropyridines (nimodipine, nitrendipine); (2) the benzothiazepines (diltiazem); (3) the phenylalkylamines (verapamil, D600); and (4) the piperazines (flunarizine, cinnarizine). Currently it is thought that these agents mediate their antiseizure effects through neuronal calcium channel blockade; however, no evidence substantiating this mechanism has been reported. For the calcium
* To whom all correspondence should be addressed: Mayo Clinic, Rochester, MN 55905, U.S.A.
channel antagonists, an ordering of antiseizure activity similar to that provided by binding studies would provide corollary evidence that the calcium channel antagonists exert their antiseizure activity by binding to sites that regulate neuronal calcium channels. Binding studies have identified three distinct allosterically linked binding sites for the first three classes of calcium channel antagonists listed above (Glossmann et al., 1985). The fourth class, the piperazines, utilize the phenylalkylamine binding site (Murphy et al., 1984). The allosteric interactions between these sites are such that the benzothiazepines increase dihydropyridine binding affinity (see Murphy et al., 1984; Schoemaker and Langer, 1985). This relationship is stereospecific for d-cis-diltiazem, the active isomer for calcium channel blockade (DePover et al., 1982; Ferry and Glossmann, 1983; Goll et al., 1983). A reflection of this binding interaction at the physiologic level has been re-
0014-2999/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
300
ported in cardiac tissue (DePover et al., 1983) in which nimodipine's negative ionotropic effects upon perfused rat hearts was potentiated by diltiazem. In contrast, the allosteric interaction between the dihydropyridines and the phenylalkylamines or piperazines is a negative allosteric interaction (Ehlert et al., 1982a,b; Bolger et al., 1983; Janis et al., 1984; Murphy et al., 1984; Zobrist et al., 1986). This negative binding interaction has been best demonstrated in cardiac and nervous tissue (Ehlert et al., 1982a,b), however functional antagonism of dihydropyridines actions by either the phenylalkylamines or the piperazines has not been previously reported. In this investigation we examined at the physiologic level the hypothesis that the dihydropyridine calcium channel antagonists exert their antiseizure actions via neuronal calcium channel blockade by examining the positive and negative dihydropyridine binding interactions discussed above. The calcium channel antagonists utilized in this investigation are nimodipine (a dihydropyridine), diltiazem (a benzothiazepine) and flunarizine (a piperazine). Nimodipine and flunarizine have been shown to have antiseizure activity (nimodipine: Meyer et al., 1986; flunarizine: Desmedt et al., 1975) while diltiazem lacks antiseizure activity (Wauquier et al., 1985). From this investigation we report that centrally administered diltiazem enhances and flunarizine inhibits the antiseizure action of nimodipine suggesting that this result may be a reflection of allosteric binding interactions occurring at neuronal calcium channel binding sites.
2. Materials and methods
2.1. Animal preparation Male Sprague-Dawley rats (250-350 g) were surgically i m p l a n t e d u n d e r 2% h a l o t h a n e anesthesia with epidural electrodes and an intraventricular (i.c.v.) cannula. Following anesthetic induction the animal was placed in a stereotaxic head holder and the head shaved and swabbed with betadine. The skull was exposed and the underlying periosteum scraped away.
Cranial burr holes were made for three epidural E E G electrodes in the following locations: 2 mm rostral to the coronal suture and 2 mm lateral to midline, one on each side (hemispheric leads). The third burr hole was placed 2 mm rostral to the lambdoid suture and 2 mm lateral to the sagittal suture on the right side of the skull (cerebellar lead). The burr hole for the i.c.v, cannula was made 1.5 mm caudal to the coronal suture and 1.5 mm lateral to the midline on the right side. Three stainless steel screws (self-tapping screws 00 × 3 / 1 6 inches; Small Parts, Inc., Miami, FL) were placed in a tripod arrangement followed by EEG electrode implacement. The screws provided anchorage of the electrode housing to the skull surface. A 21 gauge bevel-tipped cannula (1.2 cm in length) was stereotaxically placed into the fight lateral cerebroventricle (5 mm in from the dura). After use, each rat was killed and cannula iraplacement verified by i.c.v, injection of trypan blue dye. The epidural E E G recordings were obtained from electrodes made from stainless steel wire (28 gauge). Construction of the EEG electrodes is as follows. Stainless steel wires were soldered into female crimp gold plated pins. The pin portion of the wire-pin combination was inserted into a microminiature strip connector (Allied Electronics, For Worth, TX). The strip connector was cut such that three electrode leads could be accommodated. The tip of the wire was bent so a foot process was created. The exposed stainless steel wire was covered with acrylic, except for the last mm of the foot. The electrode array was secured to the skull with dental cement placed around the EEG electrode housing and the i.c.v, cannula. Dental cement adhesion between the EEG housing and the stainless steel screws provided additional stabilization of the electrode array. Each animal was allowed at least two days recovery before entering a study group. Implanted rats were housed individually, allowed free access to food and water, and placed on a 12 h light-dark cycle.
2.2. Drugs The calcium channel antagonists were administered i.c.v. (total volume = 10 ffl). The calcium
301 channel antagonists, except diltiazem, were dissolved in a vehicle consisting of 25% ethanol and 75% polyethylene glycol (PEG 400) by volume. Diltiazem (30, 100 and 300 /~g; Sigma Chemical Co., St. Louis, MO) was dissolved in normal saline. The other two calcium channel antagonists were nimodipine (30, 100 and 300/~g; gift from Dr. A. Scriabine, Miles Pharmaceutical Co., New Haven, CT) and flunarizine (1, 30 and 100 ~tg; Sigma Chemical Co., St. Louis, MO). Because nimodipine is light sensitive, care was taken to prevent its exposure to light by wrapping the solution vials in foil and darkening the room. Seizures were induced by intraperitoneal (i.p.) administration of pentylenetetrazole (35 m g / k g i.p.; Sigma Chemical Co., St. Louis, MO) dissolved in normal saline.
over the cerebellum. Rats were placed in 40 cm diameter, clear plexiglass cylinders that allowed free movement without undue stress on the EEG wires. The E E G was monitored via an oscillograph ( T y p e R564B Storage Oscilliscope, Tectronix) after amplification with an AC amplifier (EEG amplifier, Grass model 7P511; 10 and 100 Hz low and high pass filter settings; 200 /~V/cm). EEG activity was quantitated by electronic spike counting. EEG spikes were defined as waveforms with a frequency greater than 10 Hz and an amplitude greater than three times that of background (preseizure) activity. These spikes were fed into a data acquisition board and stored as spikes per minute in an electronic buffer until the end of the experiment. Permanent storage was made by downloading the buffers onto floppy discs via an IBM-PCXT computer.
2.3. Treatment protocol 2.5. Data analysis Each rat was employed only once in these studies. The animals were administered pentylenetetrazole (35 m g / k g i.p.) and seizure and convulsive activity allowed to progress over the ensuing 30 rain. Typically, at 30 min post-pentylenetetrazole, a consistent spiking pattern was observed on the EEG. For each drug three log-spaced doses were given in a cumulative dosing regimen. The lowest dose of calcium channel antagonist was administered i.c.v. 30 min after pentylenetetrazole, followed 15 min later by the next dose of calcium channel antagonist and at 30 rain the highest dose of calcium channel antagonist. After the last dose the animals were observed both visually and via the electroencephalogram for the subsequent 60 min. This method of drug administration allowed for the generation of a dose-response curve for each animal. In two separate groups, a constant repeated dose of diltiazem (100 ~tg) or flunarizine (10 /~g) was administered with increasing doses of nimodipine. Repeated vehicle injections over a similar time course were found to have no effect on seizure activity.
Data analysis was facilitated by using a basic program (SPIKE.BAS, available from author, C.W. Stevens) which calculated the following: the average baseline spikes per minute, determined from the 10 min period preceding the first calcium channel antagonist administration; and the average spikes per minute for the 10 min periods following all three doses of calcium channel antagonist, these latter values were compared to the baseline value and reported as percent inhibition of spiking. For statistical analysis the percent inhibition values for each rat were grouped and compared to the baseline values by one-way ANOVA. Cumulative dose-response curves were analyzed to give EDs0 ( + 95 % confidence limits), slopes and regression values via a software package (Tallarida and Murray, 1986). Significance was set at P < 0.05.
3. Results
3.1. Vehicle injection 2.4. EEG monitoring EEG activity was monitored bipolar between the two hemispheres with respect to the third lead
As a control group (n = 14) three vehicle i.c.v. injections (10 /~l/injection) were given at 30, 45 and 60 min after pentylenetetrazole. Multiple
302 Time After PTZ
PTZ BL
3O
35
V
50
V
65
V
V
Vehicle
Time
i,.,,,,.,,i
.................................
iI
NM dose
,,,,,
li ......
' .............
i,J ........
30
; ...........
100
V
..........
I,,,
iIiiIh
300
V
V
r
NM -t- FLZ (10)
i
r
~
Fig. 1. EEG activity (2 min segments). Top set of tracings show the EEG of pentylenetetrazole-treated rats receiving multiple vehicle injections via i.c.v, administration at 35, 50 and 65 min after pentylenetetrazole. Nimodipine alone (2nd set of tracings) reduces EEG spiking at 100 and 300/~g i.c.v. Nimodipine + diltiazem (100 fig) (3rd set of tracings) causes a reduction in EEG spiking for all three doses of nimodipine. The bottom set of tracings show flunarizine's (10 fig) ability to completely suppress nimodipine's antiseizure activity at all doses of nimodipine examined. BL: baseline, pre-pentylenetetrazole; V: vehicle/drug injection).
Spikes/minute
Spikes/minute *
1501
150-
NM Dose (/~g)
30
2001
200.
:
100 300
NM injection
*
*
Time(rnin)
32
45 65
-k
100. 50-
0-
0
2s
so
7s
loo
12s
1so
Time ( m i n ) Fig. 2. Multiple vehicle i.c.v, injections. EEG spikes per minute versus time for a typical single rat. Multiple vehicle injections (36, 49 and 63 rain injection times in this example) had no effect on pentylenetetrazole-induced spiking activity (* vehicle injection).
0
25
50
75 100 Time (minutes)
125
150
Fig. 3. Nimodipine alone, EEG spikes per minute vs. time. Increasing doses of nimodipine (30, 100 and 300 fig i.c.v.) injected at 32, 45 and 65 min. The two highest nimodipine doses significantly reduced (P < 0.05) spiking activity (* nimodipine injection).
303
vehicle injections did not alter pentylenetetrazoleinduced seizure activity time course as noted in the EEG (fig. 1, top tracing) and spikes per minute values (fig. 2). The average baseline spikes per minute value (determined from the 10 rain period preceding the first vehicle injection) was not statistically different from the average spikes per minute values for the 10 rain period following each vehicle administration.
Spikes/minute
NM Dose (~g)
Time (min)
30
200 "Jr: NM injection 150
39
100
55
300
68
"Jr
OO
50.
3.2. Single calcium channel antagonists O'
I.c.v. administered nimodipine at three doses (30, 100 and 300 #g) produced a dose-dependent reduction in seizure activity (fig. 1, 2nd tracing) and a reduction in spikes per minute (fig. 3) from which a dose-response curve was generate (fig. 4). The EDs0 and 95% confidence limits calculated for nimodipine was 135.3 /zg (108.7-168.4). The other two calcium channel antagonists, diltiazem and flunarizine, did not cause a reduction in seizure activity when administered alone (fig. 4). Diltiazero was examined at doses of 30, 100 and 300 t~g i.c.v.; while flunarizine was examined at 1, 30 and 100 #g i.c.v. Higher doses of flunarizine were not examined because of toxicity. Thus, flunarizine at 300 t~g i.c.v, in three rats to death within 30 min preceded by respiratory symptoms that included gasping and frothing.
Inhibition of Spiking 1000--0 NM A--A
0
50
25
75
1O0
125
150
Time (minutes) Fig. 5. l.c.v, nimodipine in combination with a constant dose of diltiazem (100 #g). Spikes per minute versus time. A low dose of nimodipine in combination with diltiazem transiently reduces spiking activity, an effect not seen when nimodipine is administered alone at this dose. Higher doses of nimodipine in combination with diltiazem suppress spiking activity to levels observed with nimodipine alone. In this example nimodipine was administered at 39, 55 and 69 min (* injection).
3.3. Coadministration of diltiazem and flunarizine with increasing nimodipine concentrations A constant dose of diltiazem (100 /zg) was coadministered with increasing doses of nimodipine ( n = 12). This combination resulted in a greater reduction of EEG seizure activity (fig. 1, 3rd tracing) than nimodipine alone. This was especially prominent at the lowest dose of nimodipine TABLE 1
DL I ~~
Summary of the dose-response analysis of the antiseizure effects of nimodipine, dihiazem and flunarizine alone and in combination given i.c.v.
II--II fkZ 50-
Agent
Slope of dose-response
EDs0 (95% CL)
N
curve
-50 10
100
1000
Dose (/zcj i.c.v) Fig. 4. Dose-response curve for nimodipine, diltiazem and flunarizine. Percent inhibition of spiking activity versus dose. Administered alone, only nimodipine reduced spiking in a dose-dependent fashion. Each curve represents mean + S.E. for 8 or 12 animals per drug.
Nimodipine Nimodipine + diltiazem a Nimodipine +flunarizine a Diltiazem Flunarizine
119.2+13.4
135.3 (109-168)
66.8 + 12.5
66.9 (52-87)
_ b _ b
> 300 c > 300 c
_
•
b
300
c
12 12 12 8 8
a Dose-response curves for nimodipine in combination with either 100 ~ g diltiazem or 10 /~g flunarizine, b Slope not different from zero. c EDs0 not calculated due to zero slope.
304
(30 ~g) at which nimodipine alone was ineffective. This EEG seizure reduction was reflected in the spike counting (fig. 5). The ED50 and 95% confident limit values for nimodipine in combination with a constant dose of diltiazem was 67/~g (52-87), which is significantly less (P < 0.05) than nimodipine alone (see table 1). In a separate group of rats (n = 12), a constant flunarizine dose (10 ~g i.c.v.) was coadministered with increasing doses of nimodipine. This combination resulted in a complete inhibition of nimodipine's antiseizure activity as reflected on the EEG (fig. 1, bottom tracing) and the spikes per minute.
3.4. Behavior An advantage of this seizure model (pentylenetetrazole-induced seizures in conscious rats with chronically implanted EEG electrodes) is that behavior can be monitored. The dose or pentylenetetrazole (35 m g / k g i.p.) utilized in this study produced several tonic-clinic fits occurring only within the first 20 min after pentylenetetrazole administration. Subsequently, while EEG activity displayed frequent repetitive bursting, motor seizures were not observed. However, in all rats receiving pentylenetetrazole, characteristic in-
Inhibition of Spiking 100&--& • --"
so
-50
0
+ DIL(IO0) ~ - / NM + F L Z ( I ~ / / T ~ NM
T~ ~ "
1O0 Dose (/~g ICV)
1000
Fig. 6. Dose-response curve for nimodipine alone and in combination with diltiazem and flunarizine. Nimodipine + diltiazem (100 ~g) significantly shifts (P < 0.05) the low dose portion of the curve to the left, reflecting the EEG observations. Flunarizine completely suppresses nimodipine inhibition of spiking. Each curve represents mean _+S.E. of 12 animals per drug.
creases in levels of motor activity and augmented startled responses to auditory and tactile stimuli were observed. Though not systematically quantified, administration of diltiazem and flunarizine alone did not alter these behavior changes induced by pentylenetetrazole. Administration of nimodipine at 100 and 300 /xg i.c.v, reversed pentylenetetrazole-induced excitatory behavior and led to a state of relative behavioral depression, i.e. decreased motor activity and diminished response to auditory and tactile stimuli. When administered alone, the lowest dose of nimodipine (30 ~g) examined did not produce behavioral depression, however when this low dose is coadministered with diltiazem behavioral depression resulted (fig. 6). Coadministration of flunarizine with nimodipine resulted in behavioral depression only at the two highest doses of nimodipine (100 and 300/~g i.c.v.), a finding similar to administration of nimodipine alone at these two doses.
4. Discussion
4.1. Dihydropyridine effects following i.c.v, administration." modulation by diltiazem and flunarizine The dihydropyridine calcium channel antagonists bind to specific and high affinity sites in nervous tissue thought to be associated with calcium channels (Miller and Freedman, 1984). This dihydropyridine binding can be positively and negatively allosterically modulated by the benzothiazepines (diltiazem) and the phenylalkylamines/piperazines (verapamil/flunarizine), respectively. The dihydropyridine class of calcium channel antagonists are thought to mediate their antiseizure effects by binding to the calcium channel associated binding sites. An attempt to associate these binding interactions to a physiological correlate was made in this investigation by coadministering diltiazem and flunarizine with nimodipine to determine their effects on nimodipine's antiseizure capabilities against pentylenetetrazoleinduced seizures in the conscious rat. Nimodipine alone was shown to reduce seizure activity in a dose-dependent fashion (fig. 4) and this activity was significantly enhanced through coadministra-
305 tion with diltiazem (fig. 6). If the dihydropyridines do indeed mediate their antiseizure actions through binding to highly specific sites, the affinity for which is positively enhanced by diltiazem, then this finding of enhanced antiseizure activity may be a reflection of this binding interaction. Conversely, flunarizine coadministration completely suppressed nimodipine's antiseizure activity, and is thus consistent with the negative interaction of flunafizine on nimodipine binding. These related findings suggest that the dihydropyridine calcium channel antagonists mediate their antiseizure activity through interaction with a site similar to that defined by the binding studies. It should be stressed that although these studies provide evidence that the dihydropyridines mediate their antiseizure actions through neuronal calcium channel blockade, they do not rule out the possibility that these agents may be acting through more than one mechanism. The dihydropyridine calcium channel antagonists have been reported to possess actions other than calcium channel blockade, these include inhibition of adenosine uptake (Murphy and Snyder, 1982; Wu et al., 1983; Marangos et al., 1984; Phillis et al., 1984), and inhibition of the calcium dependent regulatory protein calmodulin (Bostrom et al., 1981; Johnson et al., 1982). Inhibition of adenosine reuptake would cause increased extracellular adenosine. Adenosine is a potent anticonvulsant (Dunwiddie and Worth, 1981; Barraco et al., 1984) and is thought by some to be the brain's endogenous anticonvulsant (Dragunow, 1985). 4.2. Considerations of the pentylenetetrazole i.c.v. model The present studies employed the i.c.v, route of administration to systematically assess drug effects. Repeated vehicle injections were without effect on pentylenetetrazole seizures, indicating the lack of a cumulative injection artifact. The use of the i.c.v, route assumes that the site of drug action is sufficiently close to the ventricular lumen for the drug to reach. The present studies clearly display a dose-dependent effect of nimodipine. Similar results have been recently observed with this model for a wide range of dihydropyridine
analogues (Moron et al., 1988). The failure of diltiazem or flunarizine to exert antiseizure activity would not appear to be due to simply drug access. Indeed, an i.c.v, route of administration diminished the role played by peripheral metabolism, plasma protein binding and drug brain penetration. Moreover, as the drugs were coadministered, and the dose-response curves reflect the maximum effect and not seizure duration, it is unlikely that the pronounced effects observed with these agents could be accounted for by differential clearance. 4.3. Diltiazem's and flunarizine's lack of antiseizure activity In our investigations diltiazem alone was found to lack antiseizure activity although when coadministered with nimodipine it enhanced nimodipine's antiseizure activity. Wauquier et al. (1985) reported diltiazem's lack of protection against bicuculline induced convulsions in rats. Additionally, Ascioti et al. (1986) reported only weak antiseizure activity of diltiazem against audiogenic induced seizures in D B A / 2 mice. Both of these previous investigations utilized systemic routes of administration of diltiazem; and both investigators suggested that diltiazem's lack of antiseizure activity may be due to poor CNS penetrance. However, in this investigation we report the lack of antiseizure activity for diltiazem after central (i.c.v.) administration. Though flunarizine has been reported to be active after systemic administration in a variety of seizure models (Desmedt et al., 1975; Ashton and Wauquier, 1979; Wauquier et al., 1979; Lee et al., 1986) and in man (Declerck and Wauquier, 1983; Overweg et al., 1984), we failed to observe activity even at doses which were toxic. We do not know the reason for this apparent lack of effect. It is possible that, when administered systematically, flunarizine exerted its actions at sites different from those acted upon by nimodipine, lying more distant from the ventricles. Another possible explanation for the lack of antiseizure activity by flunarizine in this model may be due to its slow onset of action. Others (Desmedt et al., 1975; Wauquier et al., 1979; Drago et al., 1986) have noted this delayed onset of action
306 r e q u i r i n g 2 h or m o r e a f t e r a d m i n i s t r a t i o n b e f o r e a n t i s e i z u r e a c t i v i t y is n o t e d . I n these studies t h e total e x p e r i m e n t a l p e r i o d l a s t e d less t h a n 2 h, t h u s s u g g e s t i n g a n e x p l a n a t i o n for l a c k of effect. F l u n a r i z i n e ' s d e l a y e d o n s e t of a c t i o n is a s s o c i a t e d w i t h its l o n g d u r a t i o n of action, D e s m e d t et al. (1975) n o t e d seizure p r o t e c t i o n u p to 72 h a f t e r a single f l u n a r i z i n e dose.
4.4. Conclusion I n s u m m a r y , t h e effects of a g e n t s t h a t p o s i tively and n e g a t i v e l y r e g u l a t e n i m o d i p i n e b i n d i n g w e r e e x a m i n e d for effects u p o n n i m o d i p i n e ' s a n t i seizure a c t i v i t y d u r i n g c o a d m i n i s t r a t i o n w i t h nimodipine. Diltiazem, which increases nimodipine binding, enhanced nimodipine's antiseizure activity. I n c o n t r a s t , f l u n a r i z i n e , w h i c h r e d u c e s n i m o d i p i n e b i n d i n g , s u p p r e s s e s n i m o d i p i n e ' s antiseizure activity. T h e s e f i n d i n g s are c o n s i s t e n t w i t h the h y p o t h e s i s t h a t n i m o d i p i n e (a d i h y d r o p y r i d i n e c a l c i u m c h a n n e l a n t a g o n i s t ) exerts its a n t i s e i z u r e action, at least in part, t h r o u g h i n t e r a c t i o n s w i t h sites classified b y b i n d i n g studies to b e a s s o c i a t e d with a calcium channel.
Acknowledgement Supported in part by Grant R01 NS24329.
References Ascioti, C., G.B. De Sarro, B.S. Meldrum and G. Nistico, 1986, Calcium entry blockers as anticonvulsant drugs in DBA/2 mice, Br. J. Pharmacol. 88, 379p. Ashton, D. and A. Wauquier, 1979, Behavioural analysis of the effects of 15 anticonvulsants in the amygdaloid kindled rat, Psychopharmacology 65, 7. Barraco, R.A., T.H. Swanson, J.W. Phillis and R.E. Berman, 1984, Anticonvulsive effects of adenosine analogues on amygdaloid-kindled seizures in rats, Neurosci. Lett. 46, 317. Bolger, G.T., P. Gengo, R. Klockowski, E. Luchowski, H. Siegel, R.A. Janis, A.M. Triggle and D.J. Triggle, 1983, Characterization of binding of the C a 2 + channel antagonists (3H)nitrendipine, to guinea pig ileal smooth muscle, J. Pharmacol. Exp. Ther. 225, 291. Bostrom, S.L., B, Ljung, S. Mardh, S. Forsen and E. Thulin, 1981, Interactions of antihypertensive drug felodipine with calmodulin, Nature 292, 777.
Declerck, A.C. and A. Wauquier, 1983, Double blind study of the effectiveness of flunarizine in therapy resistant epilepsy in mentally retarded children, Boll. Chim. Farm. 122, 9. DePover, A., I.L. Grupp, G. Grupp and A. Schwartz, 1983, Diltiazem potentiates the negative inotropic action of nimodipine in the heart, Biochem. Biophys. Res. Commun. 114, 922. DePover, A., M.A. Matlib, S.W. Lee, G.P. Dube, I.L. Grupp, G. Grupp and A. Schwartz, 1982, Specific binding of (3H)nitrendipine to membranes from coronary arteries and heart in relation to pharmacological effects. Paradoxical stimulation by diltiazem, Biochem. Biophys. Res. Commun. 108, 110. Desmedt, L.K.C., C.J.E. Niemegeers and P.A.J. Janssen, 1975, Anticonvulsant properties of cinnarizine and flunarizine in rats and mice, Arzneim. Forsch. (Drug Res.) 25, 1408. Drago, F., C. Valerio, G. Clementi and U. Scapagini, 1986, Effects of flunarizine on experimentally induced convulsions in animals, Funct. Neurol. 1,529. Dragunow, M., 1985, Is adenosine an endogenous anticonvulsant?, Epilepsia 26, 480. Dunwiddie, T.V. and T. Worth, 1982, Sedative and anticonvulsive effects of adenosine analogues in the mouse and rat, J. Pharmacol. Exp. Ther. 220, 70. Ehlert, F.J., E. Itago, W.R. Roeske and H.I. Yamamura, 1982a, The interaction of (3H)nitrendipine with receptors for calcium antagonists in the cerebral cortex and heart and rats, Biochem. Biophys. Res. Commun. 104, 937. Ehlert, F.J., W.R. Roeske, E. Itagon and H.I. Yamamura, 1982b, The binding of (3H)nitrendipine to receptors for calcium channel antagonists in the heart, cerebral cortex, and ileum of rats, Life Sci. 30, 2191. Ferry, D.R. and H. Glossmann, 1983, Tissue-specific regulation of (3 H)nimodipine binding to putative calcium channels by the biologically active isomer of diltiazem, Br. J. Pharmacol. 78, 81p. Glossmann, H., D.R. Ferry, A. Goll, J. Stiessing and M. Schober, 1985, Calcium channels: basic properties revealed by radioligand binding studies, J. Cardiol. Pharmacol. 7 (Suppl. 16), $20. Goll, A., D.R. Ferry and H. Glossmann, 1983, Target size analysis of skeletal muscle channels: positive allosteric heterotrophic regulation by d-cis-diltiazem is associated with apparent channel oligmer dissociation, FEBS Lett. 157, 63. Janis, R.A., J.G. Sarmiento, S.G. Maurer, G.T. Bolger and D.J. Triggle, 1984, Characteristics of the binding of (3H)nitrendipine to rabbit ventricular membranes: modification by other calcium channel antagonists and by the Ca ++ channel agonist Bay k 8644, J. Pharmacol. Exp. Ther. 231, 8. Johnson, J.D., P.L. Vaghy, T.H. Crouch, J.D. Potter and A. Schwartz, 1983, A hypothesis for the mechanism of action of some of the Ca ++ antagonistic drugs: calmodulin as a receptor, in: Advances in Pharmacology and Therapeutics II, Vol. 3, eds. H. Yoshida, Y. Hagihara and S. Ebashi (Pergamon, Oxford) p. 121. Lee, S., H. Kawawaki, O. Matsuoka and R. Murata, 1986, Effect of Ca-antagonist (flunarizine) on kindling seizures in suckling rats, No To Hattatsu 18, 292.
307 Marangos, P.J., M.S. Finkel, A. Verma, M.F. Maturi, J. Patel and R.E. Patterson, 1984, Adenosine uptake sites in dog heart and brain: interaction with calcium channel antagonists, Life Sci. 35, 1109. Meyer, F.B., R.E. Anderson, T.M. Sundt, Jr. and F.W. Sharbrough, 1986, Selective central nervous system calcium channel blockers - a new class of anticonvulsant agents, Mayo Clin. Proc. 61, 239. Miller, R.J. and S.B. Freedman, 1984, Are dihydropyridine binding sites voltage sensitive calcium channels, Life Sci. 34, 1205. Morocutti, C., F. Pierelli, L. Sanarelli, E. Stefano, A. Peppe and G.L. Mattioli, 1986, Antiepileptic effects of calcium antagonist (nimodipine) on cefazolin-induced epileptogenic foci in rabbits, Epilepsia 27, 498. Mor6n, M.A., T.L. Yaksh and C.W. Stevens, 1988, The antiepileptic activity of eight dihydropyridine calcium channel antagonists: mechanism of action, Pharmacologist 30, A120. Murphy, K.M.M., R.J. Gould and S.H. Snyder, 1984, Regulation of (3H)nitrendipine binding: a single allosteric site for verapamil, diltiazem and prenylamine, in: Nitrendipine, eds. A. Scriabine, S. Vanov and K. Deck (Urban and Schwarzenberg, Baltimore) p. 107. Murphy, K.M.M. and S.H. Snyder, 1982, Adenosine receptor binding and specific receptors for calcium channel drugs, in: Calcium Entry Blockers, Adenosine, and Neurohumors: Advances in Coronary Vascular and Cardiac Control, eds. G.F. Merrill and H.R. Weiss (Urban and Schwarzenberg, Baltimore) p. 295. Ots, M.E., T.L. Yaksh, R.E. Anderson and T.M. Sundt, Jr., 1987, Effect of dihydropyridines and diphenylalkylamines on pentylenetetrazole-induced seizures and cerebral blood flow in cats, J. Neurosurg. 67, 406.
Overweg, J., C.D. Binnie, J.W.A. Meijer, H. Meinardi, S.T.M. Nuijten, S. Schmaltz and A. Wauquier, 1984, Double-blind placebo controlled trial of flunarizine as add on therapy in epilepsy, Epilepsia 25, 217. Phillis, J.W., T.H. Swanson and R.A. Barraco, 1984. Interactions between adenosine and nifedipine in the rat cerebral cortex. Neurochem. Int. 6, 693. Schoemaker, H. and S.Z. Langer, 1985, [3H]Diltiazem binding to calcium channel antagonists recognition sites in the rat cerebral cortex, European J. Pharmacol. 111. 273. Tallarida R.J. and R.B. Murray, 1984, Graded dose-response, in: Manual of Pharmacological Calculations with Computer Programs (Springer-Verlag, New York, NY) p. 82. Walden, J., E.J. Speckmann and O.W. Witte, 1985, Suppression of focal epileptiform discharges by intraventricular perfusion of a calcium antagonist, Electroencephalogr. Clin. Neurophysiol. 61,299. Wauquier, A., D. Ashton and W. Melis, 1979, Behavioural analysis of amygdaloid kindling in beagle dogs and the effects of clonazepam, diazepam, phenobarbital, diphenylhyndantoin and flunarizine on seizure manifestation, Exp. Neurol. 64, 579. Wauquier, A., D. Ashton, G. Clincke and J. Fransen, 1985, Calcium entry blockers as cerebral protecting agents: comparative activity in tests of hypoxia and hyperexcitability, Jap. J. Pharmacol. 38, 1. Wu, J.H., J.W. Phillis and V.L. Coffin, 1983, Calmodulin antagonists inhibit adenosine uptake by rat brain cortical synaptosomes, Neurosci. Lett. 37, 187. Zobrist, R.H., K.M. Giacomini, W.L. Nelson and J.C. Giacomini, 1986, The interaction of phenylalkylamine calcium channel blockers with the 1,4-dihydropyridine binding site, J. Mol. Cell. Cardiol. 18, 963.