Antiepileptic activity of preferential inhibitors of persistent sodium current

Abstract

Objective: Evidence from basic neurophysiology and molecular genetics has implicated persistent sodium current conducted by voltage-gated sodium (NaV ) channels as a contributor to the pathogenesis of epilepsy. Many antiepileptic drugs target NaV channels and modulate neuronal excitability, mainly by a use-dependent block of transient sodium current, although suppression of persistent current may also contribute to the efficacy of these drugs. We hypothesized that a drug or compound capable of preferential inhibition of persistent sodium current would have antiepileptic activity.

Methods: We examined the antiepileptic activity of two selective persistent sodium current blockers ranolazine, a U.S. Food and Drug Administration (FDA)-approved drug for treatment of angina pectoris, and GS967, a novel compound with more potent effects on persistent current, in the epileptic Scn2a(Q54) mouse model. We also examined the effect of GS967 in the maximal electroshock model and evaluated effects of the compound on neuronal excitability, propensity for hilar neuron loss, development of mossy fiber sprouting, and survival of Scn2a(Q54) mice.

Results: We found that ranolazine was capable of reducing seizure frequency by approximately 50% in Scn2a(Q54) mice. The more potent persistent current blocker GS967 reduced seizure frequency by >90% in Scn2a(Q54) mice and protected against induced seizures in the maximal electroshock model. GS967 greatly attenuated abnormal spontaneous action potential firing in pyramidal neurons acutely isolated from Scn2a(Q54) mice. In addition to seizure suppression in vivo, GS967 treatment greatly improved the survival of Scn2a(Q54) mice, prevented hilar neuron loss, and suppressed the development of hippocampal mossy fiber sprouting.

Significance: Our findings indicate that the selective persistent sodium current blocker GS967 has potent antiepileptic activity and that this compound could inform development of new agents.

Keywords: Epilepsy; Mossy fiber sprouting; Neurophysiology; Sodium channel.

Figure 1. Ranolazine reduces seizure frequency in Scn2a^Q54 mice

(A) Number of seizures for individual male mice in 30 minutes before and 30 minutes after treatment with either vehicle (left) or ranolazine (right). (B) A histogram of percent change in seizure frequency following a single i.p. injection of either vehicle or ranolazine (40 mg/kg). Percent change was calculated in response to treatment, with n = 9 for each treatment. (**p < 0.005; repeated measures ANOVA)

Figure 2. GS967 inhibits persistent sodium current

(A) Representative trace of sodium current in the absence (black trace) or presence (red trace) of 1 µM GS967. The inset illustrates persistent sodium current on an expanded scale. (B) Representative trace of voltage dependent sodium current in the absence (black trace) or presence (blue trace) of 10 µM phenytoin. The inset illustrates persistent sodium current on an expanded scale. (C) Concentration response of inhibition of persistent sodium current (closed symbols, solid lines) and transient sodium current (open symbols, dashed lines) by GS967 (red circles) and phenytoin (blue squares). (D) Steady-state use-dependent inhibition of transient sodium current by either 1 µM GS967 (red symbols) or 10 µM phenytoin (blue symbols) at stimulation frequencies of 10, 30, or 100 Hz. Values represent ratios of use-dependent inhibition in the presence of drug to that in the absence of drug. All data are expressed as mean ± SEM, with n = 7–11 for each condition.

Figure 3. GS967 inhibits persistent current and spontaneous firing in Scn2a^Q54 neurons

(A) Representative normalized trace of sodium currents from hippocampal pyramidal neurons from Scn2a^Q54 mice recorded in the absence (black trace) or presence (red trace) of 1 µM GS967. (B) Summary data for persistent sodium current (expressed as % of peak current) recorded from pyramidal neurons in the absence or presence of 1 µM GS967. Peak current densities were not significantly different between control and GS967 conditions (control: 187.2 ± 32.6 pA/pF; GS967: 172.3 ± 46.2 pA/pF). (C) Representative spontaneous action potential firing recorded from a pyramidal neuron from either wild-type or Scn2a^Q54 mice. Membrane potential was clamped at −80 mV and spontaneous action potentials were recorded in the absence and presence of 1 µM GS967. (D) Concentration response of inhibition of spontaneous action potential firing by GS967 (closed circles) and inhibition action potential firing by 1 µM phenytoin (open square). Data are expressed as mean ± SEM, with n = 5–7 for each concentration.

Figure 4. GS967 reduces seizure frequency and improves survival of Scn2a^Q54 mice

(A) Number of seizures in 30 minutes for individual mice at baseline (day 0) and after treatment (day 1 and 2) with either control chow or 1.5 mg/kg/d GS967. (B) A histogram of percent change in seizure frequency following oral administration of either control or chow containing GS967 at two dosage levels. Percent change was calculated in response to treatment, with n = 6 for each treatment (*p < 0.05 and **p < 0.005 compared to control; repeated measures ANOVA) (C) Survival curves of Scn2a^Q54 mice placed on control chow or chow containing GS967 (dose 1.5 mg/kg/d). Treatment began at 3 weeks of age indicated by the dashed line, with n = 18–20 per group. Survival difference between groups was significant at p < 0.005; Cox proportional hazards model.

Figure 5. GS967 prevents hilar neuron loss and suppresses mossy fiber sprouting in Scn2a^Q54 mice

A–C Cresyl violet stained sections of the dentate gyrus from a representative wild-type mouse (WT, panel A), untreated Scn2a^Q54 mouse (panel B) and a Scn2a^Q54 mouse treated with 1.5 mg/kg/d GS967 from P21-P60 (panel C). Images in the middle and lower panels are high-magnification views. Scale bars represent 500 µm, 200 µm and 100 µm for the upper, middle and lower panels, respectively. D–F, Timm stained sections of the dentate gyrus from a wild-type (WT, panel D) untreated Scn2a^Q54 mouse (panel E) and a Scn2a^Q54 mouse treated with GS967 from age P21-P60 (panel F). Images in the lower panels are high-magnification views. Scale bars represent 500 µm and 200 µm for the upper and lower panels, respectively. Quantification of the average density in the inner molecular layer of the dentate gyrus normalized to background is reported below images. Data are expressed as mean ± SEM, with n = 6–8 per group (*p < 0.05; one-way ANOVA followed by Tukey's post-hoc).