Chapter 3 Mechanisms of Action of Levetiracetam and Newer SV2A Ligands
Levetiracetam (LEV; ucb L059; (S)-α-ethyl-2-oxo-pyrrolidine acetamide; Figure 3-1) is the (S)-enantiomer of the ethyl analog of the nootropic drug piracetam. Random screening in audiogenic susceptible mice initially showed that both systemic and central administration of LEV protect against seizure activity in these animals, whereas the R-enantiomer (ucb L060) and main metabolite (ucb L057) of LEV were inactive.1 This suggests that the parent compound mediates an antiepileptic effect by a central action. However, the therapeutic utility of this observation was challenged by other findings, which showed LEV to be inactive in the two conventional screening models in rodents for antiepileptic drugs (AEDs), the maximal electroshock and pentylenetetrazol seizure tests.2
Despite these conflicting results, UCB Pharma initiated a number of small, open-label studies in patients with treatment refractory seizures. The outcome suggested LEV to be an effective and well-tolerated add-on therapy devoid of clinically relevant pharmacokinetic interactions. This triggered a clinical development program resulting in approval of LEV as add-on treatment of adult patients with partial onset seizures by the FDA in 1999 and the EMEA in 2000.3 Both agencies later extended this approval to include children, and more recent approvals include adjunctive treatment of both myoclonic seizures in patients with juvenile myoclonic epilepsy and primary generalized tonic clonic seizures in patients with idiopathic generalized epilepsy and monotherapy treatment of patients with new onset epilepsy (by EMEA only).3 Taken together, these studies demonstrate LEV’s efficacy as a broad-spectrum add-on therapy as well as a monotherapy treatment for epilepsy.
More than 1 million patients had been treated with LEV by 2006, and by 2007 the drug had become the most prescribed new AED for the treatment of epilepsy.3 Together with its novel pharmacological profile and unique mechanisms of action, this triggered significant drug discovery activities at UCB Pharma targeting the identification of LEV analogs with improved mechanistic and antiepileptic properties. The first successful outcome of these efforts was the discovery of brivaracetam (BRV) and seletracetam (SEL) (Figure 3-1). These two clinical AED candidates currently undergo phase III and II studies, respectively, as add-on treatment of drug-refractory adult patients with partial onset seizures. This chapter will briefly describe the epilepsy pharmacology and review the mechanisms of action of levetiracetam, brivaracetam, and seletracetam.
The lack of activity of LEV in the maximal electroshock and pentylenetetrazol seizure test appears to reflect a general absence of anticonvulsant activity in rodents in acute seizure tests employing either maximal electroshocks or administration of CD97 doses of chemoconvulsants.2 This contrasts a significant ability of LEV to suppress seizures in animals with an acquired, chronic epilepsy, as revealed by LEV’s potent seizure protection in a number of different kindling models, or in various genetic animal models of epilepsy.4 These results reveal a selective, broad-spectrum action of LEV in animal models of epilepsy that mimics both partial and generalized seizures in man—a profile that it does not share with any other AED.
Several studies have shown LEV’s remarkable ability to counteract kindling acquisition, induced by either PTZ administration to mice1 or electrical stimulation of amygdala in rats.5 Two independent experiments in the latter model showed that LEV permanently abolishes the kindling-induced increase in afterdischarge duration, even after cessation of treatment.5,6 This persistent effect against kindling acquisition after prolonged treatment distinguishes LEV from other AEDs7 and suggests that it does not simply mask the expression of kindled seizures through seizure suppression, but potentially possesses antiepileptogenic properties.
Systemic administration of high doses of LEV to rodents only induces minor sedative and ataxic effects.2 Psychomimetic behaviors are absent2 and cognitive performance unaltered.8 Combined with the potent seizure suppression by LEV in animal models of epilepsy, this results in a very high separation between doses inducing seizure protection and CNS-related adverse effects.2
The findings summarized earlier indicate that the pharmacological properties of LEV in animal models of seizures and epilepsy are unique and distinguish it from all other AEDs.4 LEV is the only AED to reveal an absence of anticonvulsant activity in conventional screening tests. This contrasts broad-spectrum seizure suppression and kindling inhibition with a wide safety margin in animal models of acquired and genetic epilepsy. This novel preclinical profile has nourished the desire to determine LEV’s mechanism of action.
A vast number of both in vitro and in vivo electrophysiological studies in rodents has consistently shown an absence of effect of LEV on normal neuronal responses and neuronal characteristics.9 This contrasts several other studies reporting that LEV exerts a preferential action against hypersynchronization of epileptiform activity.
It has been observed that LEV differs from other AEDs by its ability to reduce increases in the amplitude of evoked population spikes, reflecting hypersynchronization, in rat hippocampal slices expressing epileptiform activity due to perfusion with high K+/low Ca2+.10 Further exploration with simultaneous dual extra- and intracellular recordings in the same model showed that LEV was the only AED to decrease the number of population spikes per extracellular response, without altering the number of action potentials per intracellular burst.11 These results suggest that LEV may counteract the transition from interictal to ictal activity. This could explain both its absence of activity against acute seizures induced by immediate, ictal activity in normal animals as well as its selective seizure protection, and low induction of adverse effects in animals with acquired and genetic epilepsy.2
The antiepileptic action of AEDs traditionally relates to a primary action on one or more of three main mechanisms. These consist of facilitation of GABAA/BZ receptors; inhibition of voltage-gated Na+ channels, and inhibition of low voltage-gated (T-type) Ca2+ channels. Numerous studies have failed to show a direct interaction of LEV with any of these three mechanisms.9 This confirms that LEV’s unique profile in epilepsy pharmacology must reflect a novel mechanism of action.
It has repeatedly been observed that LEV possesses an ability to inhibit AMPA-gated currents in rat hippocampal and cortical neurons.12,13 This effect only becomes significant at concentrations above therapeutic relevance and can therefore not be expected to contribute to LEV’s antiepileptic action. However, other studies conducted at therapeutically relevant concentrations have discovered two novel cellular effects that probably contribute to LEV’s unique mechanism of action.
LEV was shown to differ from other AEDs by an ability to reverse the inhibition of Zn2+ and β-carbolines of both GABA-gated currents in rat hippocampal and dentate granule neurons and glycine-gated currents in spinal cord neurons (Table 3-1).14 This distinguishes LEV from other AEDs and associates it with a potentially novel mechanism, in particular when viewed in the context of the “sprouted mossy fiber/Zn2+-sensitive GABAA receptor” hypothesis.15 Indeed, it has been proposed that epileptogenesis involves alterations in both the subunit composition of the GABAA receptor as well as mossy fiber sprouting from dentate granule cells with Zn2+ containing terminals. This is believed to induce a vicious circle of disinhibition in hippocampus resulting in epileptic discharges—a condition that LEV may counteract.
LEV was also been shown to reduce high-voltage gated Ca2+ currents.16,17 This effect appears to relate to a selective modulation of N-type Ca2+ channels.18 Furthermore, several studies have reported on LEV’s ability to reduce both ryanodine and IP3-receptor mediated Ca2+ release from the endoplasmic reticulum.19,20 Taken together, these results suggest a novel effect of LEV on Ca2+ transients by a dual action on both N-type Ca2+ channels, incorporated in the plasma membrane, as well as on intraneuronal Ca2+ stores.
LEV’s ability to oppose Zn2+ inhibition of GABA- and glycine-gated currents and on Ca2+ transients are only of a modest magnitude. Thus, although novel, these mechanisms cannot constitute the primary antiepileptic mechanism of action of LEV. This appears, instead, to relate to LEV’s binding to the synaptic vesicle protein 2A.
LEV (10 µM) has not been found to displace radioligands specific for a variety of receptors, ion channel proteins, reuptake sites, and second messenger systems.21 This contrasts to the observation of a specific binding site for 3H-LEV in rat brain membranes, which has become known as the Levetiracetam Binding Site (LBS).21 LEV was found to bind saturably, reversibly, and stereospecifically to LBS. A strong correlation also existed between the affinity of a series of LEV analogs to the LBS and their seizure protection in epilepsy models.21 This suggests an important functional role for LBS in the antiepileptic mechanism of LEV and provided a strong rationale to identify its molecular nature. Approximately 10 years after the discovery of LBS, it was finally documented that synaptic vesicle protein 2A (SV2A) is the molecular correlate to LBS.22
SV2 is a 12-transmembrane protein incorporated in the membrane of synaptic vesicles, present in the presynaptic terminal. It exists in three isoforms, SV2A, SV2B, and SV2C, of which SV2A is the most widely distributed in the brain and also present on many neuroendocrine cells.23