Sodium channel mutations in epilepsy
and other neurological disorders (original) (raw)
Twenty SCN1A missense mutations have been evaluated in functional assays (23, 24, 33, 39–46). Functional analysis is complicated by the difficulty of cloning neuronal sodium channel cDNAs, which are uniquely unstable during propagation in bacterial cultures (unlike the muscle sodium channel cDNAs or the calcium and potassium channel cDNAs). Functional studies in the Xenopus oocyte system and in transfected mammalian cells do not always agree, and there is little experimental basis for extrapolation to in vivo effects. Nonetheless, some interesting patterns have emerged relating altered channel function to the neuronal hyperexcitability that is thought to underlie seizure disorders.
One common functional abnormality is impaired channel inactivation leading to increased persistent current. Normally, the voltage-gated sodium channels open rapidly in response to altered membrane potential and inactivate rapidly, declining to 1% of maximal current within a few milliseconds (Figure 3A). In 3 different GEFS+ mutations (14, 19), persistent current was increased to 2–5% of peak current (39) (Figure 3A). In the context of the neuron, this persistent current is thought to reduce the depolarization threshold required for firing, resulting directly in hyperexcitability. Another common mechanism is demonstrated by the GEFS+ mutation D188V, which spends less time in the inactivated state than the wild-type channel (Figure 3B). The result is greater availability of channels for opening in response to depolarization, another route to hyperexcitability. The altered biophysical properties of representative mutant channels are summarized in Table 3.
The effects of GEFS+ mutations on SCN1A channel properties have been studied in the Xenopus oocyte system and in transfected mammalian cells. (A) Whole-cell recordings from HEK tsA201 cells transfected with the indicated mutant SCN1A cDNAs demonstrate increased persistent current from the mutant channels (39). (B) Mean normalized amplitudes of sodium currents elicited by 80-Hz pulse trains in HEK cells expressing SCN2A cDNA containing the GEFS+ mutation D188V demonstrate reduced cumulative inactivation of the mutant channel during high-frequency trains of channel activation (40). (C) Voltage-dependent gating of SCN1A cDNA constructs in Xenopus oocytes expressed in the absence (left) or presence (right) of the β1 cDNA (85). See text for discussion.
A variety of functional abnormalities in mutant alleles of SCN1A encoding the sodium channel Nav1.1
A unique biochemical mechanism was observed for the missense mutation D1866Y, located in the C-terminal domain of SCN1A, in a family with GEFS+ (23). In the Xenopus oocyte system, the mutant channel exhibited a depolarized shift in voltage dependence of fast inactivation; the effect was tenfold greater in the presence of the β1 subunit (Figure 3C). Modeling with the program NEURON (http://www.neuron.yale.edu/neuron/) indicated that this shift is sufficient to produce neuronal hyperexcitability. Because the difference between the wild-type and mutant channels was increased by the presence of the β1 subunit, the effect of the mutation on subunit interaction was tested. Yeast 2-hybrid screen and co-immunoprecipitation demonstrated direct interaction between the C-terminal cytoplasmic domains of the α and β subunits, which was impaired by the D1866Y mutation (23). The D1866Y mutation thus defines an intracellular interaction domain that appears to be required, in combination with the extracellular interaction domain (23), to form the stable α/β complex. Since mutations in either SCN1A or β1 can result in GEFS+, it is not surprising that impaired interaction between the 2 subunits could also cause the disease.
The GEFS+ mutation R1648H has been examined in 3 expression systems with different outcomes. When the mutation was introduced into the rat SCN1A cDNA and examined in the Xenopus oocyte expression system, accelerated recovery from inactivation was observed (45). In the human cDNA in transfected mammalian cells, persistent current was the major effect (Figure 3A) (39). Alekov et al. introduced R1648H into the SCN4A cDNA and expressed the clone in mammalian HEK tsA201 cells. In this context, they observed slowed inactivation and accelerated recovery from inactivation, leading to increased channel availability, but no persistent current (42). A second substitution at the same residue, R1648C, was identified in a patient with SMEI (47). In transfected cells, persistent current was generated by R1648C at a level indistinguishable from that of the mutation R1648H that causes GEFS+ (33). Thus, the heterologous expression systems are not able to distinguish between missense mutations that lead to mild disease in vivo and those that lead to severe disease.
The data indicate that seizures can result from increased SCN1A channel activity, as in the missense mutations described above, and from reduced activity, as in the truncation mutations. Neuronal firing patterns appear to be extremely sensitive to subtle changes in sodium channel function. In the future, the most physiologically relevant data are likely to be obtained from measurements of neuronal currents in knock-in mouse models carrying human mutations.
We investigated the in vivo effect of an SCN2A mutation with impaired inactivation in the Q54 transgenic mouse (48). Analysis of the mutation SCN2AGAL879–881QQQ in the Xenopus oocyte system revealed an increase in persistent current and in the percentage of current that inactivated with a slow time constant (Figure 4, B and C) (49). When the mutant cDNA was expressed in transgenic mice under the control of the neuron-specific enolase promoter, the mice exhibited progressive seizures of hippocampal origin accompanied by loss of neurons called hippocampal sclerosis (Figure 4, A and E). Persistent sodium current was detected in recordings from CA1 neurons of the transgenic mice (Figure 4D), demonstrating agreement between the Xenopus assay and the in vivo effect. The phenotype of the Q54 mice most closely resembles human mesial temporal lobe epilepsy.
An SCN2A mutation with persistent current causes seizures in the Q54 transgenic mouse. (A) Focal motor seizure in a Q54 mouse. (B) The GAL879–881QQQ mutation is located in the D2S4–S5 linker. (C) The mutant channel generates persistent current in Xenopus oocytes. (D) Whole-cell sodium currents recorded from CA1 hippocampal neurons from presymptomatic Q54 mice demonstrate increased persistent current. (E) Nissl-stained sections of hippocampus area CA3 reveal extensive neuronal cell loss in a Q54 mouse compared with a wild-type littermate. Adapted with permission from Neuroscience (48).