Abstracts

Rationale: The striatum is especially vulnerable to pro-apoptotic effects of certain AEDs given to neonatal rats. We asked if this toxicity alters the maturation of GABA and glutamate transmission in striatum. We examined synaptic currents in medium spiny neurons (MSNs, striatal output neurons), from postnatal day (P)10 to 18 in rats exposed on P7 to one treatment of phenobarbital (PB), phenytoin (PHT), lamotrigine (LTG), or levetiracetam (LEV). To assess long-term outcomes, separate groups were evaluated for locomotor activity and reversal learning as juveniles (P21), and for water maze navigation as adults. Methods: Inhibitory and excitatory post-synaptic currents (I and EPSCs) were recorded from MSNs on P10, P14 and P18 using whole cell voltage clamp in corticostriatal slices. P7 pups received: saline, PB (37.5 or 75mg/kg), PHT (50mg/kg), LTG (20mg/kg), or LEV (200mg/kg). Another group received melatonin (MEL 20 mg/kg) pretreatment before PB (75mg/kg) on P7 to determine if MEL, a compound effective for preventing PB-induced cell death, could attenuate the effects of PB. For behavioral testing in juveniles, P7 pups received PB (75mg/kg) or saline, and were tested on P21 on reversal learning in a T-maze, and locomotor activity in the with and without amphetamine (AMPH 1.5mg/kg). For behavioral testing in adults, pups were exposed daily from P7-14 with PB (75mg/kg) or saline and tested at 12 months in a standard water maze (hidden platform) task. Results: In controls, m and sIPSC frequency increased 2x from P10 to 14 and an additional 50% by P18. IPSC amplitude and decay decreased from P10 to 18. EPSC frequency increased from P14 to 18 with no change in amplitude. P7 exposure to PB or PHT abolished the increase in IPSC and EPSC frequency from P10 to 18. LTG exposure at P7 resulted in similar but transient deficits. LEV was without effect. MEL pretreatment prevented the PB-induced disruption of IPSC frequency on P14. PB exposed pups showed normal t-maze acquisition but significant impairment in reversal learning at P21. They also showed greater locomotor activation than controls in response to AMPH. Adults exposed as pups to PB had longer escape latencies than controls during acquisition in the water maze and displayed excessive thigomotaxis; in a probe test (platform removed) they performed at chance levels. Conclusions: Our data indicate that the normal developmental pattern of maturation of striatal GABAergic and glutamatergic synapses, as measured by age-dependent increases in IPSCs and EPSCs, is disrupted by P7 exposure to PB, PHT, or LTG, but not by exposure to LEV, a drug that does not cause cell death in neonatal rats even at high doses. Together with the ability of MEL to prevent the effect of PB, the results indicate that cell death may be a determinant of this disruption. The behavioral impairment seen in both juveniles and adults exposed to PB postnatally, may reflect long-term consequences of earlier disruption in synaptic maturation in striatum. These results underscore the need to identify and select antiepileptic treatments that avoid acute and long-term neurotoxicity in the developing brain.