Abstracts

Analyzing the functions of genes responsible for mendelian forms of epilepsy may reveal new strategies for epilepsy prevention and treatment. [italic]KCNQ2[/italic] and [italic]KCNQ3[/italic] encode voltage-gated K+ channel subunits. [italic]KCNQ2[/italic] and [italic]KCNQ3[/italic] mutations result in benign familial neonatal convulsions (BFNC). Recently, Devaux [italic]et al.[/italic] showed that KCNQ2 and KCNQ3 proteins are concentrated at unmyelinated portions of axons, the axon initial segments (AIS) and nodes of Ranvier (JNeurosci 2004 24:1236). To begin to assess the importance of axonal KCNQ channels for neuronal physiology and for the BFNC phenotype, we investigated the mechanism underlying channel targeting to axons. Frozen brain sections from wild-type and ankyrin-G knockout mice were stained with antibodies against KCNQ2 and KCNQ3. An [italic]in vivo [/italic]cell surface redistribution assay and fusion proteins were employed to localize KCNQ subunit domains involved in binding ankyrin-G, a scaffold protein previously identified at AIS and nodes. Fluorescence recovery after photobleaching (FRAP) was used to detect interactions between full-length KCNQ subunits and ankyrin-G. Regions implicated in targeting were tested with transfection and immunostaining experiments using cultured hippocampal neurons. Patch clamp recordings were used to examine the effects of ankyrin-G binding on KCNQ2/3 channel gating. Computational modeling was used to analyze the potential effects of axonal KCNQ2/3 channel activity on neuronal excitability. KCNQ2/3 subunits were highly concentrated at AIS throughout the CNS. Targeting of KCNQ2/3 at AIS required ankyrin-G, because it was abolished in neurons from ankyrin-G knockout mice. Surface redistribution experiments implicated a peptide motif near the carboxy termini of KCNQ2 and KCNQ3 in ankyrin-G binding. Remarkably, this motif was strongly homologous with the peptide sequence that targets Na⁺ channels to AIS through interaction with ankyrin-G. FRAP experiments showed that the interaction of heteromeric KCNQ2/3 with ankyrin-G depends on this motif. Nevertheless, ankyrin-G interaction had no effect on channel kinetics. After transfection, fusion proteins bearing the KCNQ2 ankyrin-G binding motif were efficiently targeted to the AIS of cultured hippocampal neurons. A computational model showed that KCNQ2/3 channels at AIS may exert strong effects on action potential firing thresholds and frequency, even though neurons exhibit little KCNQ current at the soma. The axonal targeting of KCNQ2/3 subunits depends on interaction with ankyrin-G. Concentration of KCNQ2/3 subunits at AIS and nodes, which are critical for the initiation and propagation of action potentials, may allow them to play an important role in controlling neuronal excitability. (Supported by NINDS R01 NS49119 and NICHD P30 HD26979.)