Abstracts

Rationale: KCNQ2/3 voltage-gated potassium channels underlie the M-current (I_M) regulating neuronal excitability. About 150 mutations in KCNQ2 have been published, about half leading to mild neonatal-onset epilepsy(Benign Familial Neonatal Seizures) and half leading to severe epileptic encephalopathy. High concentrations of functional KCNQ2/3 channels are found at the axon initial segment (AIS) where they exert control over action potential generation. Experimental determination of the pathophysiological mechanisms in severe KCNQ2 encephalopathy is important for guiding therapy in this disorder. Candidate mechanisms include alterations in voltage-dependent gating, trafficking, or stability. Methods: We used immunofluorescence and confocal imaging mice to determine localization of KCNQ2/3 in both the cortex and hippocampus of transgenic hKCNQ2-G279S, and in transfected cultured rat hippocampal neurons. We used surface biotinylation assays in CHO cells transfected with mutant KCNQ2 in order to determine whether or not localization of the KCNQ2/3 channels are altered or abnormally degraded. Results: Immunofluorescence microscopy performed on tissue sections from transgenic mice over-expressing the dominant negative mutant G279S revealed a strikingly aberrant pattern: KCNQ2 was completely absent from the AIS and was retained at numerous intracellular puncta in the soma and dendrites. KCNQ3 weakly labeled the AIS, and was partially redistributed to these puncta. Voltage-gated sodium channel concentration at the AIS was normal. In addition, cultured hippocampal neurons electroporated with mutant KCNQ2 G279S show absence or abnormal pattern of KCNQ2 labeling at the AIS at DIV7. However, surface biotinylation assays in CHO cells show no changes in mutant KCNQ2's ability to be detected at the surface. Conclusions: Some KCNQ2 mutations may act by preventing trafficking to the AIS, leading to an imbalance in sodium and potassium channel activity, excessive action potential initiation, and cellular hyperexcitability. The KCNQ2 G279S mutation appears to fall into this category. Since such effects that may not be easily revealed through heterologous expression, further development of neuronal expression and in vivo models is warranted.