Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Structure. Author manuscript; available in PMC 2013 July 3.
Published in final edited form as:
PMCID: PMC3610540

Crystal structure of the ternary complex of a NaV C-terminal domain, a fibroblast growth factor homologous factor, and calmodulin


Voltage-gated Na+ (NaV) channels initiate neuronal action potentials. NaV channels are composed of a transmembrane domain responsible for voltage-dependent Na+ conduction and a cytosolic C-terminal domain (CTD) that regulates channel function through interactions with many auxiliary proteins, including fibroblast growth factor homologous factors (FHFs) and calmodulin (CaM). Most ion channel structural studies have focused on mechanisms of permeation and voltage-dependent gating but less is known about how intracellular domains modulate channel function. Here we report the crystal structure of the ternary complex of a human NaV CTD, an FHF, and Ca2+-free CaM at 2.2 Å. Combined with functional experiments based on structural insights, we present a platform for understanding the roles of these auxiliary proteins in NaV channel regulation and the molecular basis of mutations that lead to neuronal and cardiac diseases. Furthermore, we identify a critical interaction that contributes to the specificity of individual NaV CTD isoforms for distinctive FHFs.


In the axon initial segment of neurons a large and rapid influx of sodium ions (INa) through voltage-gated Na+ (NaV) channels is responsible for the rapid upstroke of the action potential. Dysregulation of NaV channels leads to a variety of neurological disorders. Mutations in SCN1A and SCN2A, which encode NaV1.1 and NaV1.2, respectively, lead to various epilepsy syndromes. Both gain- and loss of-function mutations have been described (Meisler et al., 2010), emphasizing the importance of the precise regulation of channel function. Mutations in the SCN5A-encoded cardiac NaV1.5, with similar biophysical consequences, have been linked to various inherited arrhythmias (Zimmer and Surber, 2008).

NaV channels, which are monomers with tetrameric repeats of 6 transmembrane (TM) segments, share a similar architecture with other voltage-gated cation (Ca2+ and K+) channels. In these 6 TM segments, the first 4 TM segments (S1 – S4) comprise the voltage-sensor domain (VSD) and the last 2 TM segments (S5 and S6) make up a central pore when assembled as a tetrameric configuration. The pore mediates selective cation permeation through its selectivity filter, and this conduction can be allowed (open) or obstructed (closed) by a “gate” near the intracellular membrane surface of the pore lined by S6. In voltage-gated cation channels, VSDs control pore gating upon sensing a change in membrane voltage. Four VSDs are located on the periphery of the central pore (Figure S1).

While the pore-forming α-subunit is responsible for the fundamental properties of voltage-dependent Na+ conduction, auxiliary proteins regulate channel function. Most voltage-gated ion channel structural studies have focused on the general mechanisms of permeation and voltage-dependent gating by studying prototypical Kv channels, such as an archaeal Kv channel (KvAP), the Shaker-type Kv channel (Kv1.2), and the NaV channel from Arcobacter butzleri (NavAb) (Jiang et al., 2003; Lee et al., 2005; Long et al., 2005; Payandeh et al., 2011). Much less is known about how their intracellular domains, such as the NaV cytoplasmic C-terminal domain (CTD), modulate channel function, however. Most NaV channel auxiliary proteins interact with the cytoplasmic C-terminal domain (CTD) of the channel's α-subunit (Figure S1). Among these are the NaV β subunit, through its short cytoplasmic domain (Spampanato et al., 2004); calmodulin (Mori et al., 2003); and members of the fibroblast growth factor homologous factor (FHF) family (Liu et al., 2001). Not only is the NaV CTD important for the regulation of channel activity, but a significant number of disease-causing mutations are localized to NaV CTDs (Lossin, 2009; Ohmori et al., 2006; Spampanato et al., 2004) and their associated proteins (van Swieten et al., 2003).

FHFs belong to the fibroblast growth factor (FGF) superfamily (Smallwood et al., 1996), but are functionally different from canonical FGFs (Olsen et al., 2003). The four FHF genes (FGF11-FGF14) encode 10 isoforms generated by alternative splicing of exons that encode an extended N-terminus preceding a core domain that is homologous with FGFs. Unlike FGFs, FHF N-termini lack signal sequences and so are not secreted from cells (Smallwood et al., 1996). Moreover, they cannot activate FGF receptors because of subtle structural differences in their FGF-like core (Olsen et al., 2003). Rather than serving as extracellular growth factors, FHFs remain intracellular, bind to NaV CTDs, and modulate NaV channel function (Liu et al., 2001; Lou et al., 2005). This NaV-modulating role was highlighted by the identification of FGF14 as the locus for spinocerebellar ataxia 27 (van Swieten et al., 2003), and demonstration that the mutant FGF14 reduces NaV channel currents, alters channel gating, and decreases neuronal excitability in a dominant negative manner (Laezza et al., 2007). Although several FHFs are expressed in the brain, they are not functionally redundant. Gene deletion of Fgf14 in mice causes ataxia, dyskinesia, and dystonia (Wang et al., 2002). Knockout of Fgf12 exacerbates all of those phenotypes and exaggerates the effects on neuronal NaV channels (Goldfarb et al., 2007). In addition to regulating Na+ conduction, FHFs also affect NaV channel trafficking to the plasma membrane and targeting to the axon initial segment (Laezza et al., 2007; Wang et al., 2011a).

The ubiquitous Ca2+ sensor calmodulin (CaM) also binds NaV CTDs, near the putative FHF interaction site. A familial autism-associated mutation localizes to the CaM binding region in NaV1.2 and is postulated to alter regulation of Ca2+-dependent NaV1.2 channel function (Weiss et al., 2003). How CaM controls NaV channel function is not clear, and whether Ca2+ also affects NaV channel function through direct binding to a putative EF hand in the CTD has been controversial (Miloushev et al., 2009; Shah et al., 2006).

The structural basis by which FHFs or CaM regulate NaV channels is not known as current structural information about NaV CTDs and these binding partners is limited to studies of any one component in isolation (Chagot et al., 2009; Goetz et al., 2009; Miloushev et al., 2009; Olsen et al., 2003; Schumacher et al., 2004), or one binding partner and its minimal CTD binding target in the case of CaM and the CaM-binding IQ motif peptide in NaV CTDs (Chagot and Chazin, 2011; Feldkamp et al., 2011). These structures provide little context for how these partners participate in NaV channel regulation and how mutations in NaV CTDs affect channel function. Moreover, individual FHFs impose distinctive modulation on specific NaV channel isoforms (Lou et al., 2005; Wang et al., 2011a), yet how this specificity is encoded is not known. Sequence analysis of the putative FHF and NaV CTD interaction sites (Goetz et al., 2009; Liu et al., 2003) shows that these sites are highly conserved among the NaV CTDs and FHFs, respectively, suggesting that specificity is encoded by unidentified features within these conserved domains. We therefore aimed to solve the structure of the ternary complex of a NaV CTD, an FHF, and CaM to provide a foundation for understanding these interactions and the consequent regulation of NaV function in neurons and the heart.


The structure of the ternary complex of Nav1.5 CTD, FGF13, and Ca2+-free CaM

We co-expressed a NaV1.5 CTD (the predicted structured region of NaV1.5 from the end of IVS6 at amino acid 1773 to amino acid 1940), an FHF (FGF13U, which has a short N-terminal extension), and CaM in E. coli and purified the ternary complex by Co2+ affinity chromatography followed by size exclusion chromatography. We chose this NaV CTD and FHF because the resulting ternary complex was biochemically the most stable and yielded well-diffracting crystals among the several different combinations of NaV CTDs and FHFs tested. The sequence of the NaV CTD is highly conserved among the subtypes (over 75% identical among NaV1.5, NaV1.1, and NaV1.2), and the solution structures of the EF-hand fragments of the NaV1.2 CTD and NaV1.5 CTD are nearly identical (Chagot et al., 2009; Miloushev et al., 2009). We postulated that failed previous attempts to crystallize a NaV CTD in complex with an FHF (Goetz et al., 2009) might be due to the lack of inclusion of CaM and consequent non-specific interactions among complexes containing a NaV CTD and FHF because we observed by size-exclusion chromatography an apparent higher order (stoichiometry > 1:1, NaV1.5 CTD:FGF13) binary complex that was reduced by addition of CaM (data not shown). The ternary complex was stable. Its elution volume on the size exclusion column was consistent with a molecular weight about equal to the sum of its individual components (~60 kDa), indicating the stoichiometry of the ternary complex is 1:1:1 (Figure S2). The resultant ternary complex was crystallized in the space group P213 (a=b=c=126.04 Å) with one copy of the ternary complex in each asymmetric unit. The crystals were grown in the presence of 5 mM EGTA and diffracted to 2.2 Å Bragg spacings. The experimental phases were derived by single anomalous dispersion from selenomethionine-substituted crystals. The final model contains the NaV CTD amino acids 1776-1928; FGF13U amino acids 11-158 (since the N-terminal “U” extension that defines this isoform is not present in the model, it is hereafter designated “FGF13”); and the entire CaM except for the N-terminal 4 amino acids and the loops from residues 22-27 and 57-61 in the N-lobe. The model refined to a free R factor of 22.7 % (Table 1).

Table 1
Data collection, phasing and refinement statistics

Figure 1 shows the overall architecture of the ternary complex. The NaV CTD is comprised of one globular domain (α1-α5) followed by an extended 29-amino acid helix (α6) that contains the CaM-binding IQ motif in its proximal portion. FGF13 binds to the globular domain of the CTD. The five α-helices in the globular domain of the NaV CTD are separated by four short anti-parallel β-strands. The first four helices (α1-α4) with two β-strands (β1 and β3) adopt a paired EF-hand fold, as was observed in NMR structures of NaV CTD fragments containing only α1-α4 (r.m.s.d. of 0.62 Å, Figure S3) (Chagot et al., 2009; Miloushev et al., 2009). Addition of α5 plus a loop between α5 and α6 and a short β strand (β4) to the paired EF-hand fold creates the binding site for FGF13 (Figure 1C). The presence of β4 turns the loop between α3 and β3 into a short β–strand (β2) that forms a short β sheet with the rest of the β-strands (β1-β4) (Figure 1C).

Figure 1
Overall architecture of the ternary complex of the NaV CTD, FGF13, and CaM. (A) Structure of the NaV1.5 CTD (green) in complex with FGF13 (yellow) and CaM (magenta). (B) Same as (A) but rotated 180°. (C) Zoomed-in view of the NaV CTD with the ...

Although our Nav CTD construct starts right after S6 on the basis of the structure of the prokaryotic Nav channel NavAb, we are not able to orient our ternary complex relative to the NavAb structure unambiguously given the lack of experimental constraints. Two hypothetical orientations of the ternary complex relative to the NavAb structure embedded in the membrane are shown in Figure S4.

Specific Interactions between the Nav CTD and FHF

Initial analysis of the interaction surface between the NaV CTD and FGF13 provides a means to understand functional consequences of disease-causing mutations in NaV CTDs, but yields little initial insight into the specificity of CTD-FHF pair-wise interactions and functional consequences. The shape of the NaV CTD binding surface on FGF13 is complementary to that of the FGF13-binding surface on the CTD as indicated by a relatively high shape complementary index of ~0.67 (Lawrence and Colman, 1993). The buried surface area is ~2350 Å. The FGF13-binding surface on the NaV CTD is dominated by a depression formed by α4 and the sequential segments starting with the carboxy-terminal segment of α5 the subsequent extended loop, β4, and the beginning of α6 (amino acids Phe1879 to Arg1898) (Figure 1C and and2B).2B). Consistent with our recent biochemical and functional studies (Wang et al., 2011a), residues distal to Ser1885 contribute significantly to the FHF interaction site on NaV1.5. The NaV CTD binding surface on FGF13, highly conserved among the other FHF family members, contains several knob-like structures that fit into the NaV CTD depression (Figure 2). The main anchor appears to be a knob formed by the side chains of Leu56 and Arg57 on FGF13 that protrudes into the NaV CTD to interact with His1849 and Asp1852 of α4 (Figure 2C). A second protrusion on FGF13 (Asn97) fits into a crevice on the NaV CTD formed between Asp1839 and Pro1841. The involvement of these interacting residues is underscored by disease-associated mutations. For example, epilepsy-associated mutations in NaV1.1, D1866Y (Spampanato et al., 2004) and T1909I (Ohmori et al., 2006), reside in residues (Asp1852 and Thr1895 in the NaV1.5 CTD, respectively) that form contacts with FGF13 (Figure 2 and Figure S5). Additionally, D1839G in NaV1.5, which accepts the knob formed by Asn97 on FGF13, has been associated with the fatal cardiac arrhythmia long QT syndrome (Benhorin et al., 1998). Since the key residues that form the FGF13 binding surface are conserved among the other NaV CTDs and the critical residues forming the complementary NaV CTD binding surface on FGF13 are conserved among the other FHFs, this interaction motif is likely conserved among other NaV CTD – FHF pairs.

Figure 2
Interactions between NaV CTD and FGF13. (A) Surface representation of the ternary complex of the NaV CTD (green), FGF13 (yellow), and CaM (magenta). The interaction surfaces between the NaV1.5 CTD and FGF13 (within 4 Å each other) are colored ...

How, then, is specificity encoded between specific NaVs and FHFs? The variable FHF N-termini may contribute, but cannot explain why FGF12B (which essentially lacks the extended N-terminus found in all other FHF isoforms) can bind to and modulate NaV1.5, but not NaV1.1, as we showed (Wang et al., 2011a). Comparison of the determinants for FGF13 binding in NaV1.5 with the homologous amino acids in the NaV1.1 CTD identified one difference: Glu1890, which interacts with Lys14 of FGF13 by a salt bridge (highlighted in red in Figure 2C; and Figure 3), is replaced by a glutamine (Gln1904) at the homologous position in NaV1.1. We speculated that loss of the salt bridge with Lys9 (homologous to Lys14 in FGF13) contributes to the reduced affinity observed between FGF12B and NaV1.1 (Wang et al., 2011a); as FGF12B has the shortest N-terminal extension among the FHF family members, this loss of binding energy might be particularly costly. To test this hypothesis we performed a charge reversal swap between Lys9 in FGF12B and Glu1890 in NaV1.5 and analyzed the effects on binding affinity between the NaV1.5 CTD and FGF12B using isothermal titration calorimetry. Mutating either Lys9 in FGF12B to Glu (FGF12BK9E) or Glu1890 in NaV1.5 to Lys (Nav1.5 CTDE1890K) reduced binding affinity by at least 20-fold (Figure 3C and Table 2). However, the affinity between both mutants was restored to the level seen with the wild-type pair. Since we have equilibrium binding data for the wild-type and charge-reversed mutants of Nav1.5 and FGF13, we performed a thermodynamic double mutant cycle analysis to measure the degree of energetic coupling between these sites. The resulting energetic coupling constant for the salt bridge interaction is ~1100 (~4 kcal/mol). This high value (>5 is considered significant) strongly suggests that the salt bridge interaction is energetically significant and formed by charged-hydrogen bonding (Hidalgo and MacKinnon, 1995). Consistent with these results, FGF12BK9E co-purified with Nav1.5 CTDE1890K, like the wild-type pair, when co-expressed in E. coli. Neither FGF12BK9E nor Nav1.5 CTDE1890K co-purified with the wild-type binding partner (Figure 3D). The consequences upon channel function were consistent with the binding data. As previously shown, FGF12B induced a depolarizing shift in the V1/2 of steady-state inactivation of NaV1.5 currents (Wang et al., 2011a). In contrast, the FGF12BK9E mutant did not modulate steady-state inactivation, nor was steady-state inactivation of Nav1.5E1890K modulated by wild-type FGF12B (Figure 3F and Table 2). This lack of modulation was not due to decreased expression of either the mutant FHF or NaV (Figure S6), so it likely reflects the absence of interactions with the co-expressed partner. In contrast, FGF12BK9E induced a depolarizing shift in steady-state inactivation for Nav1.5E1890K that was similar to the shift induced by wild-type FGF12B for wild-type NaV1.5.

Figure 3
Specificity between the NaV1.5 CTD and FGF13. (A) Electrostatic interaction between NaV1.5 CTD (E1890) and FGF13 (K14). (B) Sequence alignment among NaV isoforms shows that Glu 1890 in NaV1.5 in conserved among other isoforms except NaV1.1. (C) Isothermal ...
Table 2
Contribution of salt bridge between FGF12B and NaV1.5 to binding affinity and steady state inactivation

To test whether the absence of the salt bridge between Nav1.1 and FGF12B accounts for the fact that FGF12B cannot modulate Nav1.1, we then measured the affinities by ITC between FGF12B and a wild-type NaV1.1 CTD or a NaV1.1 CTD in which Gln1904 was mutated to Glu (NaV1.1 CTDQ1904E). As we previously observed with surface plasmon resonance (Wang et al., 2011b), the affinity between FGF12B and NaV1.1 CTD is very low. The affinity between FGF12B and NaV1.1 CTDQ1904E, was markedly higher by at least 10-fold, albeit not as high as between FGF12B wild-type and NaV1.5 CTD (Table 2). Thus, these data support the hypothesis that binding and function is influenced by the salt bridge between the FHF Lys and the NaV CTD Glu, while also pointing to the existence of other determinants not yet identified.

Interaction between CaM and the Nav1.5 CTD

The interactions between CaM and the NaV CTD are mostly localized on the C-lobe of CaM and the IQ-motif in α6 (Figure 4A). The CaM C-lobe adopts a semi-open conformation as observed in NMR structures of apoCaM bound to a ~20 amino acid NaV IQ motif-containing peptide (Chagot and Chazin, 2011; Feldkamp et al., 2011) (r.m.s.d. = 1.0 Å for the CaM C-lobe with IQ motif-containing peptide). The CaM C-lobe buries the side chains of Ile1908 and Gln1909 of the NaV CTD's IQ motif, and also contacts the side chains of Ser1904, Ala1905, and Phe1912 in α6 (Figure 4A). The importance of these observed interactions is underscored by the epilepsy mutation I1922T in NaV1.1 (homologous to Ile1908) (Harkin et al., 2007); and NaV1.5 mutations S1904L (associated with the life-threatening cardiac arrhythmias Brugada Syndrome (Kapplinger et al., 2010) and Long QT Syndrome (Bankston et al., 2007)) and Q1909R (associated with Long QT Syndrome (Tester et al., 2005)) (Figure S5).

Figure 4
Interactions of NaV CTD with CaM and FGF13. (A) Zoomed-in view of the interactions of the NaV1.5 CTD (green) with the C-lobe of CaM (magenta) shown on the right. The ternary complex structure on the left is shown to help orient the figure on the right. ...

The linker between the N-lobe and C-lobe of CaM adopts a helical structure (Figure 1A-B). In the NMR structure of CaM bound to the short IQ motif peptide, the linker is disordered (Chagot and Chazin, 2011), a discrepancy that is likely because our crystal structure includes a much bigger portion of the NaV CTD or because the crystal structure captured one of many conformational states of the CaM linker in the ternary complex. The structure of the CaM N-lobe in the ternary complex can be superimposed on the N-lobe of the crystal structure of apoCaM with an r.m.s.d. of ~0.8 Å (Schumacher et al., 2004) or the N-lobe in the NMR structure of apoCaM associated with the NaV CTD IQ motif peptide with an r.m.s.d of ~1.1 Å (Chagot and Chazin, 2011).

Even though we grew our crystals in the presence of 5 mM EGTA, we observed two electron density peaks in both Ca2+-binding pockets of the CaM C-lobe. We suspected that the observed peaks were from Mg2+, not Ca2+, since the crystallization solution contained 100 mM Mg2+. To identify the ions in the Ca2+- binding pockets, we performed anomalous diffraction studies on native crystals collected at a long wavelength (1.54 Å). At this wavelength, the anomalous scattering power of Ca2+ (~1.1 e) is higher than that of sulfur (~0.6 e) and much higher than that of Mg2+ (~0.1 e). If there is an anomalous difference peak corresponding to an ion in the Ca2+-binding pocket and it is stronger than those of sulfur in the anomalous difference Fourier map, the ions must be Ca2+. If there is no significant anomalous difference peak corresponding to an ion in the Ca2+-binding pocket or the peak is weaker than those of sulfur, the ions must be Mg2+. Since we observed many anomalous difference Fourier peaks from sulfurs in cysteine and methionine residues, but no notable peaks in the Ca2+-binding loops of CaM, the observed ions in the Ca2+-binding loops of CaM are Mg2+, not Ca2+. To validate this finding, we isolated a crystal grown in the absence of Ca2+ and EGTA and soaked it in crystallization solution containing 5 mM Ca2+ for 45 min before flash-freezing. We then collected data at a long wavelength (1.6 Å). The Ca2+-soaked crystal showed one strong (~7.5 σ) and one weak (~4 σ) anomalous difference peak. The strong peak was much stronger than those from cysteines and methionines (~4 σ) (Figure 5B), suggesting that Ca2+ was successfully incorporated into the Ca2+-binding loop of CaM. There were no notable structural changes with Ca2+ binding. Structures of Ca2+-CaM alone show that the acidic residues (Glu105 and Glu141) on EF hands are involved in coordination of Ca2+ (Figure 5). In our crystal structure, the acidic residues are away from the Ca2+-binding loops, probably because the structure was captured in the apo-CaM conformation (Figure 5B). However, the Ca2+ binding loop in the crystal still has a preference for Ca2+ over Mg2+ as a lower concentration of Ca2+ (5 mM) is enough to replace a high concentration of Mg2+ (100 mM).

Figure 5
Anomalous diffraction studies of the CaM C-lobe and NaV CTD. Anomalous difference Fourier electron density maps of the CaM-C-lobe region from the crystals grown in the presence of EGTA (A) and grown in the absence of EGTA and soaked with 5 mM Ca2+ (B) ...

With our crystallographic approach we also queried whether the NaV CTD's EF-hand could bind Ca2+. Ca2+ regulates NaV channel function but whether it is through Ca2+-CaM or direct binding of Ca2+ to a putative EF hand in the Nav CTD has been controversial (Chagot et al., 2009; Kim et al., 2004; Miloushev et al., 2009; Shah et al., 2006). From the Ca2+-soaked crystal, we observed no anomalous difference peak in the putative Ca2+ binding loops of the NaV CTD's EF-hand (Figure 5C), and there was no peak in the putative Ca2+ binding loops of the Nav CTD's EF-hand in a 2Fo-Fc map (Figure S7). This observation is consistent with the idea that any Ca2+-dependent regulation of NaV channels occurs through the bound CaM.

Ternary interactions among the Nav1.5 CTD, FGF13, and CaM

Superimposition of the FGF13 structure from the ternary complex onto the FGF13 structure obtained in the absence of binding partners (Goetz et al., 2009) shows localized structural changes around the loop between β8 and β9 (Figure 4B). Upon binding the NaV CTD and CaM, that loop is reoriented so that Tyr98 of FGF13 interacts with Arg1898 and Glu1901 of α6 in NaV1.5 via a cation-π interaction (Gallivan and Dougherty, 1999) and a hydrogen bond interaction, respectively (Figure 4B). Interestingly, Glu1901 in the NaV1.5 CTD as well as Tyr98 of FGF13 also are within hydrogen bonding distances with Lys95 of the third Ca2+-binding EF-hand in the CaM C-lobe (Figure 4B). Mutation of Arg1902 in NaV1.2, homologous to Arg1898 in NaV1.5, is associated with familial autism (Weiss et al., 2003) and affects the Ca2+-dependence of the interaction between the NaV1.2 CTD and CaM (Kim et al., 2004; Weiss et al., 2003). It is the only site within the ternary complex containing components of all three molecules.


Our crystal structure, which includes the most structured region of the CTD as well as two key auxiliary proteins, provides a cornerstone for understanding the FHF-mediated and/or CaM-dependent regulation of NaV channels at the molecular level. In conjunction with the recent structure of a prokaryotic NaV channel (Payandeh et al., 2011), which does not contain the CTD, it adds new information that will help unveil the mechanisms of eukaryotic NaV channel regulation. Most immediately, it provides a basis for exploring the consequences of certain pathogenic mutations in NaV channels. For example, the epilepsy-associated mutations D1866Y and T1909 in NaV1.1 reside in residues that form contacts with the FHF (see Figure 2 and Figure S5). These mutations might be particularly disrupting for FHF contacts with NaV1.1, since NaV1.1 appears to have a reduced affinity for FHFs compared to NaV1.5 (see Figure 2 and (Wang et al., 2011b)). Mutations in FGF12 or FGF14 prevent targeting of these FHFs and NaV channels to the axon initial segment (Laezza et al., 2007; Wang et al., 2011b). Thus, a NaV1.1 epilepsy mutation that disrupts interactions with FHF might lead to a loss-of-function phenotype, consistent with the proposed model by which NaV1.1 mutations cause epilepsy (Catterall et al., 2010). Our structure also points to possible explanations for the epilepsy mutation I1922T (Harkin et al., 2007), which alters a key residue for interaction with CaM (Figure 4). Interaction between CaM and NaV channels has been shown to be critical for channel function; loss of CaM interactions by mutation in NaV1.4 or NaV1.6 markedly reduced Na+ channel current density in a heterologous expression system (Herzog et al., 2003). Growing evidence points to CaM, constitutively associated with the CTD of the CaV1.2 Ca2+ channel (Liu et al., 2010; Wahl-Schott et al., 2006), as a regulator of channel biosynthesis and trafficking in several cell types (Poomvanicha et al., 2011; Wang et al., 2007). Indeed, over-expression of CaM was shown to rescue a trafficking-defective epilepsy mutant NaV1.1 channel (Rusconi et al., 2007). By analogy, therefore, the I1922T mutation may also reduce the number of NaV1.1 channels at the plasma membrane, thereby reducing NaV1.1 current. Likewise, the Brugada Syndrome S1904L mutation in NaV1.5 may also lead to a reduced number of NaV1.5 channels and/or Na+ current (Figure S5). This would be consistent with the loss-of-function consequences of functionally characterized NaV1.5 Brugada Syndrome mutations (Antzelevitch et al., 2005).

In addition to the insight into effects of mutations, the interface between the NaV CTD and the FHF offers an unusual model for protein-protein interactions because critical binding components lie at the periphery of the interaction surface. “Hot-spot” residues, those that provide the most binding energy, are usually found toward the center of the interface (Clackson and Wells, 1995) and are sealed from bulk solvent to provide highly energetic interactions (Bogan and Thorn, 1998). The most notable interaction in our structure is the salt bridge between a Lys in the FHF and a Glu in the NaV CTD that contributes greatly to the interaction energy and offers a means to understand the reduced interaction between FHFs and NaV1.1 compared to NaV1.5 (Figure 2 and (Wang et al., 2011a)). In a similar manner, the cation-π interaction between Tyr98 of FGF13 and Arg1898 (Figure 4B) provides another contact point at the periphery of the main interface that may have important functional significance for Ca2+-regulation through a hydrogen bond between Tyr98 of FGF13 and Lys95 of the third Ca2+-binding EF-hand in the CaM C-lobe (Figure 4B). Mutation of the homologous residue in the neuronal NaV1.2 (Arg1902) to cysteine, predicted to disrupt the cation-π interaction, is associated with familial autism (Weiss et al., 2003) and affects the Ca2+-dependence of the interaction between the NaV1.2 CTD and CaM (Kim et al., 2004; Weiss et al., 2003). Together, these observations suggest an intriguing possibility that the ternary interactions, including the cation-π interaction between the NaV CTD and FGF13, may serve as a critical mediator between Ca2+/CaM- and FHF-mediated regulations of NaV currents.

Although a more thorough understanding of how Ca2+ and CaM participate in NaV function will require structural information with a Ca2+-bound CaM, this Ca2+-free structure provides an important starting parameter. Most significantly, the anomalous diffraction studies (Figure 5) support a role for CaM as the Ca2+-sensor rather than the NaV CTD EF-hand. Further structural and functional characterization will help test this hypothesis and unveil the mechanisms of eukaryotic NaV channel regulation.

Experimental Procedures

Molecular biology

The following plasmids, for protein expression and purification, have been previously described: For crystallization the human NaV1.5 CTD (amino acids 1773-1940) was cloned into pET28 (Novagen) (Kim et al., 2004); the human FGF13U (accession # NM_033642) was cloned into the second multiple cloning site of pETDuet-1 (Novagen) (Wang et al., 2011a); and CaM was cloned into pSGC02 (ref. (Kim et al., 2004)). For isothermal titration calorimetry, human FGF12B (accession # NM_004113) and human NaV1.1 CTD (amino acids 1789-1948) were cloned into pET28 (Wang et al., 2011a). For electrophysiology, FGF12B was cloned into pIRES2-acGFP1 (Clontech) as described (Wang et al., 2011a); and NaV1.5 was cloned into pcDNA3.1 as described (Wang et al., 2011a). Site-directed mutagenesis of NaV1.1, NaV1.5 or FGF12B was performed with QuikChange (Stratagene).

Recombinant protein expression and co-purification

The three plasmids containing the His6-Nav1.5 CTD, FGF13U, and CaM were electroporated into BL-21 (DE3) cells. Proteins were expressed after induction with 1 mM isopropyl-1-thio-β-D-galactopyranoside (IPTG) for 64h at 16°C. For expression of selenomethionine incorporated proteins, the cells were grown in M9 medium with the following additions (mg/L): 100 lysine, phenylalanine, and threonine; 50 isoleucine, leucine and valine; and 60 L-selenium-methionine. Cells were grown until OD600 = 0.7, and then induced with IPTG as above. Cells were harvested and resuspended in 300 mM NaCl, 20 mM Tris-HCl, 5 mM imidazole, pH 7.5, supplemented with EDTA-free protease inhibitor mixture (Roche). Cell extracts were prepared by passage twice through an Avestin homogenizer (Emulsiflex-C5, Canada) then centrifuged at 146,300 × g for 25 min. The initial purification protocol has been previously described (Wang et al., 2008). Additional purification was performed by gel filtration on a Superdex 200 10/300L column on an AKTA FPLC (GE Healthcare) in 300mM NaCl, 20mM Tris-HCl, 5mM imidazole, with 5mM EGTA pH 7.5. Protein concentrations were determined by UV absorbance with Thermo NANODROP and were concentrated to A280 = 8 in above buffer for crystallization.

Crystallization, data collection and structure determination

Crystals were grown by vapor diffusion with the sitting-drop method. Crystals were obtained with 20% PEG400, 100mM magnesium acetate and 50mM sodium cacodylate, pH 5.3. Before flash-freezing in liquid nitrogen, the crystals were cryoprotected by gradually increasing the concentration of PEG400 in the well solution to 38% over a period of 10 min. Single anomalous dispersion phases were obtained to 2.9 Å from SeMet-substituted crystals and then the phases were extended for model building. The final model was refined to Rwork/Rfree = 0.210/0.227 and contains NaV1.5 CTD, FGF13, CaM, 3 Mg2+ (two in the Ca2+ binding sites in CaM and one in the crystallographic contacts), and 220 waters. The final model is of good Ramachandran statistics (98% favored and 2% allowed).

The data were collected on beam lines 22ID, 22BM, and 24ID-C at the Advanced Photon Source. The data were processed using HKL2000 (Otwinowski and Minor, 1997). The data for the SeMet-substituted crystal were collected at the Se peak wavelength (0.97936 Å) and the crystal diffracted to 2.9 Å. Phases were obtained by the single anomalous dispersion technique using PHASER (McCoy et al., 2007) from the PHENIX interface (Adams et al., 2010) with a figure of merit of 0.42. Phases were extended to 2.4 Å by solvent flattening using RESOLVE (Terwilliger, 2000) against high-resolution native data (2.2 Å). About 60% of the entire model was built using AutoBuild from the PHENIX interface. Further model building was done manually using COOT (Emsley and Cowtan, 2004). Structure refinement was done using the high-resolution (2.2 Å) native data.

For anomalous diffraction studies, the data for the native crystal grown in the presence of EGTA were collected to 2.6 Å Bragg spacings using the rotating anode generator (Rigaku MicroMax-007 HF micro focus) at the Duke University X-ray Crystallography facility and the data for the Ca2+-soaked crystals to 2.45 Å Bragg spacings were collected at a wavelength of 1.6 Å using the beam line 24ID-C at the Advanced Photon Source. At least 180 images were collected for all the data sets. Anomalous difference Fourier maps were calculated using the phases from the structure of the ternary complex using PHENIX.

Isothermal Titration Calorimetry

Experiments were performed with an ITC-200 (MicroCal) at 20 °C. Solution containing NaV1.5, NaV1.1, NaV1.5E1890K, or NaV1.1Q1904K CTD (20–51 μM) were titrated with 20-30 10-μl injections of solution containing FGF12B or FGF12BK9E (240-510 μM). ITC experiments were repeated with different preparations and different concentrations at least three times to confirm thermodynamic parameters and stoichiometry values. The binding isotherms were analyzed with a single site binding model using the Microcal Origin version 7.0 software package (Originlab Corporation), yielding binding enthalpy (ΔH), stoichiometry (n), entropy (ΔS), and association constant (Ka). Due to the low affinity between the single mutation and wild type interaction, higher concentrations of FGF12B were used. Results are presented as mean ± standard error; statistical significance was assessed using a two-tailed Student's t-test and was set at P< 0.05.


HEK293T cells were transfected at 60% confluence using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. The total amount of DNA for all transfections was kept constant. Transfected cells were identified by GFP fluorescence. Na+ currents were recorded using the whole-cell patch-clamp technique at room temperature (20-22 °C) 48-72 h after transfection. Electrode resistance ranged from 1-2 MΩ. Currents were filtered at 5 kHz and digitized using an analog-to-digital interface (Digidata 1322A, Axon Instruments). Capacitance and series resistance were adjusted (70% to 85%) to obtain minimal contribution of the capacitive transients. The bath solution contained (in mM): NaCl 130, KCl 4, CaCl2 1.8, MgCl2 1, HEPES 10, glucose 10, pH 7.35 (adjusted with NaOH). The intracellular solution contained (in mM): CsF 110, EGTA 10, NaF 10, CsCl 20, HEPES 10, pH 7.35 (adjusted with CsOH). Osmolarity was adjusted to 310 mOsm with sucrose for all solutions. Standard two-pulse protocols were used to generate the steady-state inactivation curves: from a holding potential of -120 mV, cells were stepped to 500-ms preconditioning potentials varying between -130 mV and -10 mV (prepulse), followed by a 20 ms test pulse to -20 mV. Currents (I) were normalized to Imax and fit to a Boltzmann function of the form I/Imax =1/{1+exp[(Vm-V1/2)/k]} in which V1/2 is the voltage at which half of NaV1.5 channels are inactivated, k is the slope factor and Vm is the membrane potential. Data analysis was performed using Clampfit 10.2 software (Axon Instruments) and Origin 8. Results are presented as means ± standard error; the statistical significance of differences between groups was assessed using a two-tailed Student's t-test and was set at P< 0.05.


  1. The first crystal structure of the ternary complex of a human NaV CTD, FHF, and CaM
  2. Identification of determinants of specificity of distinctive NaV channels for FHFs
  3. Insights into the consequences of disease-causing mutations

Supplementary Material



Data for this study were collected at beamlines SER-CAT BM22/ID22 and NE-CAT ID 24-C at the Advanced Photon Source and the Duke X-ray Crystallography Facility. We thank R. Brennan, M. Schumacher, and P. Zhou for providing access to their ITC machines; C. Pemble for help with remote data collection; Z. Johnson for help with ITC experiments and critical reading of the manuscript. This work was supported by start-up funds from the Duke University Medical Center (S.-Y.L.), the Basil O'Connor Starter Scholar Research Award 5-FY10-473 from the March of Dimes foundation (S.-Y.L.), the N.I.H. Director's New Innovator Award 1 DP2 OD008380-01 (S.-Y.L), NHLBI R01 HL71165 (G.S.P.) and R01 HL088089 (G.S.P), and American Heart Association Established Investigator Award (G.S.P.). S.-Y.L. is a McKnight Scholar, Klingenstein fellow, Alfred P. Sloan Research fellow, Mallinckrodt Scholar, and Whitehead Scholar.


Accession Numbers: Atomic coordinates and structure factors for the reported crystal structure have been deposited in the Protein Data Bank under accession codes 4DCK.

Supplemental Information: Supplemental Information includes seven figures and Supplemental Experimental Procedure.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Adams PD, Afonine PV, Bunkoczi G, Chen VB, Davis IW, Echols N, Headd JJ, Hung LW, Kapral GJ, Grosse-Kunstleve RW, et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr D Biol Crystallogr. 2010;66:213–221. [PMC free article] [PubMed]
  • Antzelevitch C, Brugada P, Borggrefe M, Brugada J, Brugada R, Corrado D, Gussak I, LeMarec H, Nademanee K, Perez Riera AR, et al. Brugada Syndrome: Report of the Second Consensus Conference: Endorsed by the Heart Rhythm Society and the European Heart Rhythm Association. Circulation. 2005;111:659–670. [PubMed]
  • Baker NA, Sept D, Joseph S, Holst MJ, McCammon JA. Electrostatics of nanosystems: application to microtubules and the ribosome. Proceedings of the National Academy of Sciences of the United States of America. 2001;98:10037–10041. [PubMed]
  • Bankston JR, Sampson KJ, Kateriya S, Glaaser IW, Malito DL, Chung WK, Kass RS. A novel LQT-3 mutation disrupts an inactivation gate complex with distinct rate-dependent phenotypic consequences. Channels (Austin) 2007;1:273–280. [PubMed]
  • Benhorin J, Goldmit M, MacCluer JW, Blangero J, Goffen R, Leibovitch A, Rahat A, Wang Q, Medina A, Towbin J, et al. Identification of a new SCN5A mutation, D1840G, associated with the long QT syndrome. Mutations in brief no. 153. Online. Hum Mutat. 1998;12:72. [PubMed]
  • Bogan AA, Thorn KS. Anatomy of hot spots in protein interfaces. Journal of Molecular Biology. 1998;280:1–9. [PubMed]
  • Catterall WA, Kalume F, Oakley JC. NaV1.1 channels and epilepsy. The Journal of Physiology. 2010;588:1849–1859. [PubMed]
  • Chagot B, Chazin WJ. Solution NMR structure of Apo-calmodulin in complex with the IQ motif of human cardiac sodium channel NaV1.5. Journal of molecular biology. 2011;406:106–119. [PMC free article] [PubMed]
  • Chagot B, Potet F, Balser JR, Chazin WJ. Solution NMR structure of the C-terminal EF-hand domain of human cardiac sodium channel NaV1.5. J Biol Chem. 2009;284:6436–6445. [PubMed]
  • Clackson T, Wells J. A hot spot of binding energy in a hormone-receptor interface. Science. 1995;267:383–386. [PubMed]
  • Emsley P, Cowtan K. Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr. 2004;60:2126–2132. [PubMed]
  • Feldkamp MD, Yu L, Shea MA. Structural and Energetic Determinants of Apo Calmodulin Binding to the IQ Motif of the Na(V)1.2 Voltage-Dependent Sodium Channel. Structure 2011 [PMC free article] [PubMed]
  • Gallivan JP, Dougherty DA. Cation-π interactions in structural biology. Proceedings of the National Academy of Sciences. 1999;96:9459–9464. [PubMed]
  • Goetz R, Dover K, Laezza F, Shtraizent N, Huang X, Tchetchik D, Eliseenkova AV, Xu CF, Neubert TA, Ornitz DM, et al. Crystal structure of a fibroblast growth factor homologous factor (FHF) defines a conserved surface on FHFs for binding and modulation of voltage-gated sodium channels. J Biol Chem. 2009;284:17883–17896. [PubMed]
  • Goldfarb M, Schoorlemmer J, Williams A, Diwakar S, Wang Q, Huang X, Giza J, Tchetchik D, Kelley K, Vega A, et al. Fibroblast growth factor homologous factors control neuronal excitability through modulation of voltage-gated sodium channels. Neuron. 2007;55:449–463. [PMC free article] [PubMed]
  • Harkin LA, McMahon JM, Iona X, Dibbens L, Pelekanos JT, Zuberi SM, Sadleir LG, Andermann E, Gill D, Farrell K, et al. The spectrum of SCN1A-related infantile epileptic encephalopathies. Brain. 2007;130:843–852. [PubMed]
  • Herzog RI, Liu C, Waxman SG, Cummins TR. Calmodulin binds to the C terminus of sodium channels Nav1.4 and Nav1.6 and differentially modulates their functional properties. J Neurosci. 2003;23:8261–8270. [PubMed]
  • Hidalgo P, MacKinnon R. Revealing the architecture of a K+ channel pore through mutant cycles with a peptide inhibitor. Science. 1995;268:307–310. [PubMed]
  • Jiang Y, Lee A, Chen J, Ruta V, Cadene M, Chait B, MacKinnon R. X-ray structure of a voltage-dependent K + channel. Nature. 2003;423:33–41. [PubMed]
  • Kapplinger JD, Tester DJ, Alders M, Benito B, Berthet M, Brugada J, Brugada P, Fressart V, Guerchicoff A, Harris-Kerr C, et al. An international compendium of mutations in the SCN5A-encoded cardiac sodium channel in patients referred for Brugada syndrome genetic testing. Heart Rhythm. 2010;7:33–46. [PMC free article] [PubMed]
  • Kim J, Ghosh S, Liu H, Tateyama M, Kass RS, Pitt GS. Calmodulin mediates Ca2+ sensitivity of sodium channels. J Biol Chem. 2004;279:45004–45012. [PubMed]
  • Laezza F, Gerber BR, Lou JY, Kozel MA, Hartman H, Marie Craig A, Ornitz DM, Nerbonne JM. The FGF14F145S Mutation Disrupts the Interaction of FGF14 with Voltage-Gated Na+ Channels and Impairs Neuronal Excitability. J Neurosci. 2007;27:12033–12044. [PubMed]
  • Lawrence MC, Colman PM. Shape complementarity at protein/protein interfaces. J Mol Biol. 1993;234:946–950. [PubMed]
  • Lee SY, Lee A, Chen J, MacKinnon R. Structure of the KvAP voltage-dependent K+ channel and its dependence on the lipid membrane. ProcNatlAcadSciUSA. 2005;102:15441–15446. [PubMed]
  • Liu C, Dib-Hajj SD, Waxman SG. Fibroblast growth factor homologous factor 1B binds to the C terminus of the tetrodotoxin-resistant sodium channel rNav1.9a (NaN) J Biol Chem. 2001;276:18925–18933. [PubMed]
  • Liu CJ, Dib-Hajj SD, Renganathan M, Cummins TR, Waxman SG. Modulation of the cardiac sodium channel Nav1.5 by fibroblast growth factor homologous factor 1B. J Biol Chem. 2003;278:1029–1036. [PubMed]
  • Liu X, Yang PS, Yang W, Yue DT. Enzyme-inhibitor-like tuning of Ca2+ channel connectivity with calmodulin. Nature. 2010;463:968–972. [PMC free article] [PubMed]
  • Long SB, Campbell EB, Mackinnon R. Voltage sensor of Kv1.2: structural basis of electromechanical coupling. Science. 2005;309:903–908. [PubMed]
  • Lossin C. A catalog of SCN1A variants. Brain and Development. 2009;31:114–130. [PubMed]
  • Lou JY, Laezza F, Gerber BR, Xiao M, Yamada KA, Hartmann H, Craig AM, Nerbonne JM, Ornitz DM. Fibroblast growth factor 14 is an intracellular modulator of voltage-gated sodium channels. J Physiol. 2005;569:179–193. [PubMed]
  • McCoy AJ, Grosse-Kunstleve RW, Adams PD, Winn MD, Storoni LC, Read RJ. Phaser crystallographic software. J Appl Crystallogr. 2007;40:658–674. [PubMed]
  • Meisler MH, O'Brien JE, Sharkey LM. Sodium channel gene family: epilepsy mutations, gene interactions and modifier effects. The Journal of Physiology. 2010;588:1841–1848. [PubMed]
  • Miloushev VZ, Levine JA, Arbing MA, Hunt JF, Pitt GS, Palmer AG., 3rd Solution structure of the NaV1.2 C-terminal EF-hand domain. J Biol Chem. 2009;284:6446–6454. [PubMed]
  • Mori M, Konno T, Morii T, Nagayama K, Imoto K. Regulatory interaction of sodium channel IQ-motif with calmodulin C-terminal lobe. Biochem Biophys Res Commun. 2003;307:290–296. [PubMed]
  • Ohmori I, Kahlig KM, Rhodes TH, Wang DW, George AL. Nonfunctional SCN1A Is Common in Severe Myoclonic Epilepsy of Infancy. Epilepsia. 2006;47:1636–1642. [PubMed]
  • Olsen SK, Garbi M, Zampieri N, Eliseenkova AV, Ornitz DM, Goldfarb M, Mohammadi M. Fibroblast Growth Factor (FGF) Homologous Factors Share Structural but Not Functional Homology with FGFs. J Biol Chem. 2003;278:34226–34236. [PubMed]
  • Otwinowski Z, Minor W. Processing of X-ray diffraction data collected in oscillation mode. In: Carter CW Jr, Sweet RM, editors. Methods in Enzymology. New York: Academic Press; 1997. pp. 307–326.
  • Payandeh J, Scheuer T, Zheng N, Catterall WA. The crystal structure of a voltage-gated sodium channel. Nature. 2011;475:353–358. [PMC free article] [PubMed]
  • Poomvanicha M, Wegener JW, Blaich A, Fischer S, Domes K, Moosmang S, Hofmann F. Facilitation and Ca2+-dependent inactivation are modified by mutation of the Ca(v)1.2 channel IQ motif. The Journal of biological chemistry. 2011;286:26702–26707. [PubMed]
  • Rusconi R, Scalmani P, Cassulini RR, Giunti G, Gambardella A, Franceschetti S, Annesi G, Wanke E, Mantegazza M. Modulatory Proteins Can Rescue a Trafficking Defective Epileptogenic Nav1.1 Na+ Channel Mutant. J Neurosci. 2007;27:11037–11046. [PubMed]
  • Schumacher MA, Crum M, Miller MC. Crystal structures of apocalmodulin and an apocalmodulin/SK potassium channel gating domain complex. Structure. 2004;12:849–860. [PubMed]
  • Shah VN, Wingo TL, Weiss KL, Williams CK, Balser JR, Chazin WJ. Calcium-dependent regulation of the voltage-gated sodium channel hH1: Intrinsic and extrinsic sensors use a common molecular switch. PNAS. 2006;103:3592–3597. [PubMed]
  • Smallwood PM, Munoz-Sanjuan I, Tong P, Macke JP, Hendry SH, Gilbert DJ, Copeland NG, Jenkins NA, Nathans J. Fibroblast growth factor (FGF) homologous factors: new members of the FGF family implicated in nervous system development. Proc Natl Acad Sci U S A. 1996;93:9850–9857. [PubMed]
  • Spampanato J, Kearney JA, de Haan G, McEwen DP, Escayg A, Aradi I, MacDonald BT, Levin SI, Soltesz I, Benna P, et al. A Novel Epilepsy Mutation in the Sodium Channel SCN1A Identifies a Cytoplasmic Domain for β Subunit Interaction. The Journal of Neuroscience. 2004;24:10022–10034. [PubMed]
  • Terwilliger TC. Maximum-likelihood density modification. Acta Crystallogr D Biol Crystallogr. 2000;56:965–972. [PMC free article] [PubMed]
  • Tester DJ, Will ML, Haglund CM, Ackerman MJ. Compendium of cardiac channel mutations in 541 consecutive unrelated patients referred for long QT syndrome genetic testing. Heart Rhythm. 2005;2:507–517. [PubMed]
  • van Swieten JC, Brusse E, de Graaf BM, Krieger E, van de Graaf R, de Koning I, Maat-Kievit A, Leegwater P, Dooijes D, Oostra BA, et al. A mutation in the fibroblast growth factor 14 gene is associated with autosomal dominant cerebellar ataxia [corrected] Am J Hum Genet. 2003;72:191–199. [PubMed]
  • Wahl-Schott C, Baumann L, Cuny H, Eckert C, Griessmeier K, Biel M. Switching off calcium-dependent inactivation in l-type calcium channels by an autoinhibitory domain. Proceedings of the National Academy of Sciences. 2006;103:15657–15662. [PubMed]
  • Wang C, Hennessey JA, Kirkton RD, Wang C, Graham V, Puranam RS, Rosenberg PB, Bursac N, Pitt GS. Fibroblast Growth Factor Homologous Factor 13 Regulates Na+ Channels and Conduction Velocity in Murine Hearts/Novelty and Significance. CIRCULATION RESEARCH. 2011a;109:775–782. [PMC free article] [PubMed]
  • Wang C, Wang C, Hoch EG, Pitt GS. Identification of Novel Interaction Sites that Determine Specificity between Fibroblast Growth Factor Homologous Factors and Voltage-gated Sodium Channels. JOURNAL OF BIOLOGICAL CHEMISTRY. 2011b;286:24253–24263. [PubMed]
  • Wang C, Wang HG, Xie H, Pitt GS. Ca2+/CaM Controls Ca2+-Dependent Inactivation of NMDA Receptors by Dimerizing the NR1 C Termini. J Neurosci. 2008;28:1865–1870. [PubMed]
  • Wang HG, George MS, Kim J, Wang C, Pitt GS. Ca2+/calmodulin regulates trafficking of Ca(V)1.2 Ca2+ channels in cultured hippocampal neurons. J Neurosci. 2007;27:9086–9093. [PubMed]
  • Wang Q, Bardgett ME, Wong M, Wozniak DF, Lou J, McNeil BD, Chen C, Nardi A, Reid DC, Yamada K, et al. Ataxia and paroxysmal dyskinesia in mice lacking axonally transported FGF14. Neuron. 2002;35:25–38. [PubMed]
  • Weiss LA, Escayg A, Kearney JA, Trudeau M, MacDonald BT, Mori M, Reichert J, Buxbaum JD, Meisler MH. Sodium channels SCN1A, SCN2A and SCN3A in familial autism. Mol Psychiatry. 2003;8:186–194. [PubMed]
  • Zimmer T, Surber R. SCN5A channelopathies--an update on mutations and mechanisms. Prog Biophys Mol Biol. 2008;98:120–136. [PubMed]