Michael J. Mclean MD, PhD
Department of Neurology, Vanderbilt University Medical Center, Nashville, Tennessee
Detailed mechanistic information could help to resolve two issues that affect the use of oxcarbazepine (10,11-dihydro-10-oxo-carbamazepine; Trileptal, Novartis, East Hanover, NJ). First, oxcarbazepine is indicated as first-line therapy for the treatment of partial and secondarily generalized tonic-clonic seizures [see full prescribing information (3,17,60,61)]. The percentage of patients achieving complete seizure control on oxcarbazepine was comparable with that of patients taking carbamazepine (60). At the 10 position of the central 7-membered ring, oxcarbazepine has a keto group that is rapidly converted to an active monohydroxy derivative in humans (63). Carbamazepine differs structurally in having a 10,11 double bond that is converted to an active epoxide (15). Thus, oxcarbazepine achieves significant efficacy with one active compound, whereas the efficacy of carbamazepine results from two active compounds. Pharmacokinetic considerations aside, it would be useful to know whether different profiles of cellular actions account for the comparable efficacy of these two compounds. Pharmacologic differentiation would provide a rational basis for considering oxcarbazepine and carbamazepine to be distinct therapeutic agents.
Second, there is a trend toward treating disorders other than epilepsy with antiepileptic drugs. In this regard, three tricyclic compounds are notable. Amitriptyline, a tertiary amine antidepressant, is effective in migraine prophylaxis (19). It also has been shown to relieve the pain of postherpetic neuralgia (35,77) and painful diabetic neuropathy independent of effects on mood (33,34). In addition to antiepileptic effects, carbamazepine is effective in the treatment of trigeminal neuralgia (8,27), painful diabetic neuropathy (20,56), and bipolar disorder (12,37). Oxcarbazepine and its monohydroxy derivative have both been shown to be effective in the treatment of trigeminal neuralgia (16,81) and of affective and schizoaffective symptoms of patients with mood disorders (14,26,42,71). A distinctive mechanistic profile could help to choose among therapeutic agents for selected patients.
Known cellular actions of oxcarbazepine and its monohydroxy derivative are examined here in an effort to focus on these two questions.
BLOCKADE OF SODIUM-DEPENDENT ACTION POTENTIAL FIRING
Oxcarbazepine and its monohydroxy derivative limited sustained, high-frequency repetitive firing (SRF) of sodium-dependent action potentials of cultured mouse spinal cord neurons equipotently at concentrations between 10-8 and 10-6 mol/L (75). This is lower than maximum plasma (up to 60 µmol/L) and tissue concentrations achieved after chronic use (presumably as high as 60 µmol/g) (22), suggesting that this is a therapeutically relevant mechanism of action. The experimental paradigm for showing this effect on a single neuron is shown at the top of Figure 44.1. In the control state, SRF is elicited by depolarizing current pulses applied intracellularly (PRE in Figure 44.1). Then, in the presence of the drug, depolarizing steps were applied to examine firing (DRUG in Figure 44.1). In the example shown, action potential amplitude diminishes in parallel with declining maximal rate of rise (top trace), an indirect reflection of inward sodium current generating the upstroke of the action potential, until firing ceases for the remainder of the depolarizing step. Passage of hyperpolarizing current into the neuron led to restoration of SRF in the continuing presence of the drug, indicative of the voltage dependence of the block. After washout of the drug (POST in Figure 44.1), SRF was restored at transmembrane potentials near to or less negative than the resting potential, demonstrating reversibility of the voltage-sensitive block. In population studies, oxcarbazepine, the monohydroxy derivative, and carbamazepine reduced the percentage of neurons with SRF in a concentration-dependent manner (Figure 44.1, bottom) (38,75). The concentration dependence of this effect for carbamazepine
P.452
was shifted to the right compared with the other two drugs (Figure 44.1, bottom) (39,75). Carbamazepine has been shown to block sodium current directly in a number of preparations (51,52,64,73), including the slow sodium currents of pain-processing neurons (57,67). Oxcarbazepine and its monohydroxy derivative have not been shown to block sodium current directly, but are assumed to do so based on voltage- and use-dependent limitation of action potential firing, and by analogy with carbamazepine (75). Amitriptyline also has been reported to block sodium current by affecting both activation and inactivation kinetics in bovine chromaffin cells (2,47,48), in human embryonic kidney cells expressing human cardiac sodium channels (43), and in dorsal root ganglion cells (47,67).
|
FIGURE 44.1. Top: An experimental paradigm for showing effects of oxcarbazepine (OXC), its monohydroxy derivative (MHD), and carbamazepine (CBZ) on sodium-dependent action potential firing. PRE: Control trace showing sustained repetitive firing (SRF) elicited by an intracellularly applied depolarizing current pulse. DRUG: During superfusion with an effective concentration of a drug, firing was limited at less negative resting potentials (-55 mV) but was restored by hyperpolarization (-66 mV), indicating voltage dependence of the block. POST: After washout, repetitive firing was sustained throughout depolarizing pulses at the original resting potential. Calibrations at right apply to all traces. Bottom: Concentration dependence of limitation of SRF. Percentage of neurons with SRF (ordinate) versus drug concentration is shown. |
CALCIUM CHANNEL BLOCKADE
The monohydroxy derivative of oxcarbazepine reduced glutamatergic synaptic transmission and high-voltage-activated, N-type calcium currents in cortical and striatal neurons (6,69). Carbamazepine had only minor effects on high-voltage-activated calcium currents in neurons isolated from human hippocampus (62) and neocortex (59) obtained at epilepsy surgery. Amitriptyline relaxed smooth muscle cells isolated from human mesenteric arteries at supratherapeutic concentrations (72) and decreased motility of human vas deferens smooth muscle cells at concentrations slightly higher than the therapeutic range (40). These findings suggest an L-type blocking action of amitriptyline, but effects on neuronal L-calcium channels have not been clearly demonstrated at therapeutically relevant concentrations.
POTASSIUM CURRENTS
Enhancement of hyperpolarizing potassium currents could contribute to anticonvulsant, pain-relieving, and mood-stabilizing effects of antiepileptic drugs. Both enantiomers of the monohydroxy derivative blocked penicillin-induced
P.453
bursting in hippocampal slices. This effect was prevented by the potassium channel blocker, 4-aminopyridine. These findings suggested that the monohydroxy derivative blocked bursting by increasing a hyperpolarizing potassium current (Olpe et al., unpublished results). Carbamazepine had similar effects on penicillin-induced bursting (45) and enhanced a voltage-sensitive potassium current in rat cortical neurons in cell culture (82). Amitriptyline opened several types of potassium channels (voltage-gated, adenosine triphosphate-activated, and Ca2+-activated) that could contribute to analgesic efficacy (18). However, blockade of certain cardiac potassium channels may contribute to the arrhythmogenic potential of amitriptyline (11,25). Potassium channel blockade by amitriptyline in the central nervous system also could be proconvulsant in some instances. Interestingly, amitriptyline has anticonvulsant effects in some experimental models (10,30,79), but is associated with seizures in some patients, especially at high concentrations (50).
NMDA RECEPTORS
The importance of glutamatergic hyperexcitability, particularly that mediated by the N-methyl-D-aspartate (NMDA) type of receptors, to human (65,66) and animal (41) epileptogenesis is well documented. The monohydroxy derivative of oxcarbazepine, the principal active anticonvulsant metabolite of oxcarbazepine, decreased stimulation-evoked field potentials elicited from hippocampal slices in buffers containing low magnesium concentration to enhance the NMDA component of this potential (6,7). This occurred at concentrations of 10 to 100 µmol/L monohydroxy derivative. Assuming that total-brain concentrations of the monohydroxy derivative could reach 60 µmol/L, the peak plasma concentration obtained in epileptic patients receiving repeated doses (22), up to 30% of the NMDA-mediated field potential could be blocked.
Therapeutically relevant concentrations of carbamazepine (including concentrations in brain with chronic administration) blocked NMDA-activated current in cultured spinal cord neurons (28), inhibited NMDA-elicited calcium influx into rat cerebellar granule cells (9,24,53), inhibited NMDA-induced depolarizations in cortical wedges from epileptic mice (29) and blocked glutamato release (44).
Amitriptyline blocked NMDA-induced toxicity but not NMDA-induced elevation of extracellular glutamate in cerebellar granule cells at supratherapeutic concentrations (36). NMDA-induced currents in Xenopus oocytes injected with rat brain RNA were only slightly blocked by high concentrations of amitriptyline (70,76). At concentrations near the top of the therapeutically useful range, amitriptyline reduced NMDA-induced intracellular calcium elevations (5). Pretreatment with amitriptyline intrathecally prevented hyperalgesia produced by NMDA in rats (13). These findings suggest that some benefit of amitriptyline, particularly for the treatment of pain, may result from NMDA antagonism. However, oxcarbazepine, its monohydroxy derivative, and carbamazepine seem to have greater NMDA-blocking capabilities.
Figure 44.2 shows the results of pressure application of 10-5 mol/L glutamate in magnesium-free buffer to cultured spinal cord neurons, to enhance the NMDA component of the response. The neurons depolarized and fired continuously throughout the application (Control, top row). The monohydroxy derivative and carbamazepine blocked the depolarization and firing at concentrations that might be reached in brain with chronic administration. A relatively high concentration of amitriptyline (near top of the antidepressant range) limited action potential firing, but did not diminish the depolarization significantly. This suggests that there may be differences in concentration dependence of blockade of the NMDA component by these three tricyclic drugs that determine the extent to which NMDA blockade contributes to their therapeutic clinical utility.
SEROTONIN
Blockade of the serotonin (5-HT) transporter with a resultant increase in interstitial serotonin concentrations is thought to contribute to the antidepressant effect of amitriptyline (58). Carbamazepine at therapeutically relevant concentrations also may increase interstitial serotonin by enhancing release and by modestly blocking the transporter at higher concentrations (80). The increase in interstitial serotonin concentrations may contribute to antiepileptic efficacy of carbamazepine by stimulating γ-aminobutyric acid (GABA)-ergic interneurons bearing 5-HT3-type receptors (55) or by activation of inhibitory serotonin receptors. Oxcarbazepine has not been reported to alter extracellular serotonin concentrations or serotonin receptor profiles.
GABA RECEPTORS
The importance of GABA-mediated inhibition in controlling seizures has been well documented (46), and GABAA receptors are the target of several antiepileptic drugs (31). Carbamazepine did not alter GABAA- or GABAB-mediated responses of hippocampal CA3 neurons at concentrations up to 500 µmol/L (supratherapeutic) (4). Also, carbamazepine did not change the amplitude of responses to GABA applied by electrophoresis to cultured spinal cord neurons (38). Blockade of peripheral benzodiazepine receptors by Ro 5-4864, but not blockade of central benzodiazepine receptors by Roche compound (Ro) 15-1788, prevented the anticonvulsant effect of carbamazepine in amygdala-kindled rats (78). Carbamazepine and phenytoin potentiated
P.454
the response of GABAA receptors expressed in human embryonic kidney cells and in cultured rat cortical neurons as long as the recombinant receptors included α1, α3, or α5 subunit isoforms (23). Thus, there is mixed evidence for effects of carbamazepine in different systems. It is possible that concentration dependence of effects determines the contribution of GABAA receptor-mediated effects to the anticonvulsant efficacy of carbamazepine.
|
FIGURE 44.2. Effects of tricyclic compounds on glutamate-induced depolarization and action potential firing in cultured spinal cord neurons. Glutamate (10-5 mol/L) was applied in magne-sium-free buffer to emphasize the NMDA component of the response. Top row: Control (CONT) response to a 50-second application of glutamate by pressure ejection from a micropipette positioned near the neuron under study. Middle rows: The monohydroxy derivative of oxcarbazepine (MHD) and carbamazepine (CBZ) blocked depolarization and action potential firing at concentrations that might be reached in the brain during chronic administration. Bottom row: Amitriptyline (AMT) limited firing but did not reduce glutamate-induced depolarization substantially at a high therapeutically relevant concentration encountered in treatment of depression. Concentrations are shown below drugs. Calibrations at lower right apply throughout. (Unpublished data.) |
There is little information about the effects of amitriptyline on GABA receptors. Blockade of GABAA receptors could contribute to proconvulsant effects of amitriptyline under some conditions (32).
There are no reports of oxcarbazepine effects on GABA receptors or GABA concentrations in brain.
ACETYLCHOLINE RECEPTORS
Anticholinergic side effects of amitriptyline can limit therapeutic utility (21). These include reduced activity of bladder and gut smooth muscle leading to urinary retention and constipation, pseudodementia, and cardiac arrhythmias. Amitriptyline blocked cloned muscarinic receptors (68) and reduced acetylcholine binding to muscarinic receptors (54).
Carbamazepine-induced urinary retention also is likely to result from muscarinic cholinergic blockade. Its interaction with cholinoreceptors may be indirect. Carbamazepine uncouples muscarinic receptors and β-adrenergic receptors from G proteins (1). Mutant nicotinic cholinergic receptors that are associated with autosomal dominant nocturnal frontal lobe epilepsy are more sensitive to carbamazepine than are human native nicotinic receptors (49).
There are no reports of effects of oxcarbazepine or its monohydroxy derivative on cholinergic neurotransmission.
DISCUSSION
Based on a series of indirect assays, the mechanistic profile of oxcarbazepine and its monohydroxy derivative at therapeutically
P.455
relevant concentrations is likely to include a variety of actions. Sodium channel blockade could interrupt high-frequency repetitive firing of neurons; N-calcium channel blockade is expected to damp synaptic activity diffusely; opening of potassium channels should reduce burst activity; and blockade of NMDA receptor-mediated activity could limit hyperexcitability. All of these actions could contribute to the broad clinical utility of oxcarbazepine outlined in the introductory paragraphs of this chapter.
At face value, the mechanistic profiles of action of carbamazepine and amitriptyline appear to overlap that of oxcarbazepine substantially (Figure 44.3), but differences appear on closer inspection. Limitation of firing of sodium-dependent action potentials by oxcarbazepine and the monohydroxy derivative occurred at lower concentrations than with carbamazepine. However, the N-calcium effect of the monohydroxy derivative differentiates it further from carbamazepine. Potassium channel and NMDA receptor effects seem comparable with those of carbamazepine. In the absence of comparative data, no difference in effects on GABA and cholinergic receptors or on serotonin levels can be assumed. The sodium channel blocking action of amitriptyline rivals that of oxcarbazepine and carbamazepine in terms of occurrence at concentrations in the therapeutic range. A combination of antimuscarinic cholinergic activity, blockade of certain potassium conductances, and possibly GABAA receptor antagonism confers cardiac arrhythmogenic and proconvulsant effects in some cases. In the balance, this profile disqualifies amitriptyline as an anticonvulsant. Serotonergic actions of amitriptyline and carbamazepine may be important for pain control, and the serotonin-releasing effect of carbamazepine is essential for anticonvulsant effects in the genetically epilepsy-prone rat (80). It would be of considerable interest to know if oxcarbazepine and the monohydroxy derivative affect interstitial serotonin levels in the nervous system. This action, N-calcium channel blockade, and some pharmacokinetic properties (e.g., less induction of hepatic enzymes and fewer drug-drug interactions) could make oxcarbazepine preferable to carbamazepine in some instances.
|
FIGURE 44.3. Tricyclic compounds. |
The ability of the monohydroxy derivative and carbamazepine to block glutamate responses (Figure 44.2) begs for further mechanistic resolution to differentiate the two drugs. The potency of oxcarbazepine and carbamazepine in
P.456
inhibiting glutamate release was comparable with that of lamotrigine (74). Clarification of monohydroxy derivative effects on glutamate receptor-mediated activities also could clarify situations in which oxcarbazepine might be selected over lamotrigine for a variety of conditions.
In conclusion, current knowledge of the mechanistic profile of oxcarbazepine does not provide a simple formula for selection over other antiepileptic drugs. There are some subtle differences from carbamazepine. Multiple known actions of oxcarbazepine and its monohydroxy derivative provide a rationale for utility in the treatment of a broad range of clinical disorders, including epilepsy, neuropathic pain, and mood disorders. However, there still is much to learn about these compounds mechanistically.
REFERENCES
P.457
P.458