SEARCH

SEARCH BY CITATION

Keywords:

  • Epilepsy;
  • Excitability;
  • Severe myoclonic epilepsy of infancy;
  • GEFS+;
  • Encephalopathy;
  • Calcium channel

Summary

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

Purpose:  Dravet syndrome (DS), a devastating epileptic encephalopathy, is mostly caused by mutations of the SCN1A gene, coding for the voltage-gated Na+ channel NaV1.1 α subunit. About 50% of SCN1A DS mutations truncate NaV1.1, possibly causing complete loss of its function. However, it has not been investigated yet if NaV1.1 truncated mutants are dominant negative, if they impair expression or function of wild-type channels, as it has been shown for truncated mutants of other proteins (e.g., CaV channels). We studied the effect of two DS truncated NaV1.1 mutants, R222* and R1234*, on coexpressed wild-type Na+ channels.

Methods:  We engineered R222* or R1234* in the human cDNA of NaV1.1 (hNaV1.1) and studied their effect on coexpressed wild-type hNaV1.1, hNaV1.2 or hNaV1.3 cotransfecting tsA-201 cells, and on hNaV1.6 transfecting an human embryonic kidney (HEK) cell line stably expressing this channel. We also studied hippocampal neurons dissociated from NaV1.1 knockout (KO) mice, an animal model of DS expressing a truncated NaV1.1 channel.

Key Findings:  We found no modifications of current amplitude coexpressing the truncated mutants with hNaV1.1, hNaV1.2, or hNaV1.3, but a 30% reduction coexpressing them with hNaV1.6. However, we showed that also coexpression of functional full-length hNaV1.1 caused a similar reduction. Therefore, this effect should not be involved in the pathomechanism of DS. Some gating properties of hNaV1.1, hNaV1.3, and hNaV1.6 were modified, but recordings of hippocampal neurons dissociated from NaV1.1 KO mice did not show any significant modifications of these properties. Therefore, NaV1.1 truncated mutants are not dominant negative, consistent with haploinsufficiency as the cause of DS.

Significance:  We have better clarified the pathomechanism of DS, pointed out an important difference between pathogenic truncated CaV2.1 mutants and hNaV1.1 ones, and shown that hNaV1.6 expression can be reduced in physiologic conditions by coexpression of hNaV1.1. Moreover, our data may provide useful information for the development of therapeutic approaches.

Dravet syndrome (DS), also known as severe myoclonic epilepsy of infancy (SMEI), is a devastating autosomal-dominant genetic epileptic encephalopathy characterized by frequent and prolonged seizures, cognitive impairment, high mortality, and ataxia (Dravet et al., 2005). The most common target of DS mutations is the SCN1A gene (Claes et al., 2001), coding for the voltage-gated Na+ channel NaV1.1 α subunit (Mantegazza et al., 2010a) (Fig. 1A). Other genes have been implicated in DS, but they account for only few cases, whereas hundreds of SCN1A mutations have been identified in about 80% of DS patients (Mantegazza et al., 2010b; Marini & Mantegazza, 2010). DS SCN1A mutations can be missense (causing an amino acid change), nonsense or frame shifts (giving rise to truncated channels), or deletions (see SCN1A variant databases: http://www.molgen.ua.ac.be/SCN1AMutations and http://www.scn1a.info/; Claes et al., 2009; Lossin, 2009).

image

Figure 1.   Truncated hNaV1.1 mutants used in our study. Voltage-gated Na+ channels are essential for neuronal excitability and are formed by a principal pore-forming α subunit composed by four homologous domains of six transmembrane segments and N-terminal and C-terminal cytoplasmic domains (nine isoforms cloned: NaV1.1–NaV1.9) associated with auxiliary β subunits with a single transmembrane segment (four isoforms cloned: β1–β4) (Mantegazza et al., 2010a). (A) Schematic representation of wild-type NaV1.1 principal α subunit (cylinders represent probable α-helical transmembrane segments; IN and OUT are the intracellular and extracellular sides of the membrane, I–IV are the homologous domains). (B) Truncated DS mutant hNaV1.1-R1234*, in which a premature termination codon removes two of the four homologous domains. (C) Left, truncated DS mutant hNaV1.1-R222*, in which a premature termination codon generates a truncated N-terminal region conserving the first three transmembrane segments. Right, western blot from total cell lysate confirming that hNaV1.1-R222* is expressed in tsA-201 cells. Cells were transfected with YFP (yellow fluorescent protein: positive control; lane 1) or hNaV1.1-R222* tagged on the N-terminus with YFP (lane 2; note that the apparent molecular mass is larger than the predicted one because of the YFP tag); untransfected tsA-201 cells were used as negative control (UT; lane 3). The blot was probed with an anti-GFP antibody, which detects also YFP.

Download figure to PowerPoint

The functional effect of DS mutations and of SCN1A mutations in general has been a matter of controversy, with both loss- and gain-of-function effects reported for missense mutations in heterologous expression systems, consistent with either decreased or increased neuronal excitability, respectively (Mantegazza et al., 2010b; Mantegazza, 2011). However, truncations are predicted to cause loss of function, and several pieces of experimental evidence point to a loss of function as the common effect of epileptogenic SCN1A mutations (Mantegazza et al., 2010b). It may seem counterintuitive that a loss of function of a voltage-gated Na+ channel (NaV) can lead to epilepsy. Animal models have helped in clarifying this issue. NaV1.1 knockout (KO) mice express a truncated NaV1.1 channel, similar to patients who carry truncating mutations, and recapitulate several of the clinical features of DS (Yu et al., 2006; Kalume et al., 2007; Oakley et al., 2009). Recordings in neurons dissociated from these mice have shown that NaV1.1 loss of function reduces Na+ current and excitability selectively in γ-aminobutyric acid (GABA)ergic interneurons (Yu et al., 2006); similar results have been obtained also with a knock-in mouse model (Ogiwara et al., 2007). These results are consistent with compromised network GABAergic inhibition as the major pathomechanism in DS. However, the detailed mechanisms of epileptic hyperexcitability, as well as those of the comorbidities, remain elusive. It is important to note that the clarification of these details is essential for developing effective and targeted therapies.

In particular, one of the still incompletely understood issues in the pathomechanism of DS is if NaV1.1 truncated mutants show negative dominance. In fact, it is often hypothesized that DS NaV1.1 truncating mutations cause haploinsufficiency: a 50% reduction of functional NaV1.1 protein in heterozygotes with no effects on the wild-type protein. However, NaV1.1 mutants could act as dominant negative by inhibiting wild-type NaV1.1 or other coexpressed NaV, which would lead to an additional reduction of Na+ current. This issue is important not only for better understanding the pathogenic mechanism, but also for better evaluating possible therapeutic approaches.

Notably, some experimental results are consistent with a dominant negative role for truncated NaV1.1 mutants. In fact, it has been shown that approximately half of the Na+ current is lost in both hippocampal GABAergic neurons and cerebellar Purkinje neurons of heterozygous NaV1.1 KO mice, but a smaller additional decrease is observed in homozygous neurons (Yu et al., 2006; Kalume et al., 2007). This nonlinear loss of Na+ current has been hypothesized to depend on the compensatory upregulation of other NaV. In particular, expression of NaV1.3 has been observed in hippocampal GABAergic neurons of NaV1.1 KO mice, and it has been supposed to be larger in homozygous than in heterozygous neurons (Yu et al., 2006). However, a dominant negative effect of truncated NaV1.1 mutants could produce a similar nonlinear loss of current, because NaV1.1 current in heterozygotes would be reduced by coexpression of the mutants.

Among the several truncated proteins that show negative dominance and are expressed in physiologic conditions or disease, those of voltage-gated Ca2+ channels (CaV), which have high sequence homology with NaV, have been particularly well characterized (Raghib et al., 2001; Page et al., 2004, 2010; Jeng et al., 2008; Mezghrani et al., 2008; Pietrobon, 2010). Truncated CaV2 or CaV3 channels interact with coexpressed wild-type channels and reduce Ca2+ current with a generalized cross-suppression between the different isoforms within the CaV2 or Cav3 family, but without cross-suppression between the two families. Notably, several CaV2.1 mutations causing episodic ataxia type 2 (EA2) generate truncated forms with negative dominance (see Discussion).

Of interest, there is evidence of interactions between Na+ channel α subunits (Poelzing et al., 2006). Moreover, an epileptogenic NaV1.2 truncated mutant can modify the functional properties of wild-type NaV1.2 in overexpression experiments (Kamiya et al., 2004). However, in heterologous expression systems, truncated NaV1.1 mutants have been studied expressed alone (Sugawara et al., 2003; Mantegazza, 2011). Therefore, we have carried out cotransfection experiments to find out if truncated NaV1.1 DS mutants R222* and R1234* show negative dominance.

Methods

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

Plasmids and mutagenesis

The cDNAs of the human NaV1.1 Na+ channel α subunit (hNaV1.1; GenBank sequence NM_006920.4) and of the human NaV1.2 Na+ channel α subunit (hNaV1.2; NM_021007) were provided by Dr. Jeff Clare (GlaxoSmithKline, Stevenage, Herts, United Kingdom). The NaV1.1 clone is a shorter splice variant, with a deletion of 11 amino acids in the intracellular loop between DI and DII, which has already been used in several studies (Mantegazza et al., 2005; Rusconi et al., 2007; Cestele et al., 2008; Rusconi et al., 2009) and could be the predominant NaV1.1 variant expressed in brain (Schaller et al., 1992). We subcloned them in the plasmid pCDM8 because, as already described for NaV1.1 (Mantegazza et al., 2005; Rusconi et al., 2007; Cestele et al., 2008), it was able to reduce rearrangements of NaV cDNAs. hNaV1.2 cDNA was originally in the pCDNA3 vector; we subcloned it by digesting subsequently with EcoRI-EcoRV, EcoRV-NotI, and NotI-XbaI, after inserting the restriction sites by silent mutagenesis with the Quick Change XL Site Directed Mutagenesis Kit (Stratagene-Agilent Technologies, Santa Clara, CA, U.S.A.), and ligating the three obtained fragments into pCDM8. hNaV1.3 has been obtained by DNA synthesis (GeneScript Inc., Piscataway, NJ, U.S.A.) according to GenBank sequence NM_006922 and subcloned into pCDM8 by GeneScript Inc.

The mutation R1234* (R1245* in the longer isoform) was introduced into pCDM8-hNaV1.1 with Quick Change XL using the following primers: 5′-TATATTGATCAGTGAAAGACGATTAAG-3′ forward and 5′-CTTAATCGTCTTTCACTGATCAATA-3′ reverse. The mutation R222* was introduced in an N-terminal (NT) 1,000-bp fragment of hNaV1.1, which was subcloned into the bi-cistronic mammalian expression vector pIRES-YFP (Clontech, Mountain View, CA, U.S.A.): pCDM8-hNaV1.1 was digested using the restriction enzymes EcoRV and NotI, and the resulting fragment was ligated into pIRES-YFP using the same restriction sites. The mutation R222* was introduced in pIRES-YFP-hNaV1.1NT by means of Quick Change XL using the following primers: 5′-CAGAGTTCTCTGAGCATTGAAG-3′ forward and 5′-CTTCAATGCTCAGAGAACTCTGAA-3′ reverse, obtaining a bi-cistronic vector that expresses both hNaV1.1NT-R222* and YFP as reporter. We tagged hNaV1.1NT-R222* with YFP introducing the restriction sites for BglI and SalI by polymerase chain reaction (PCR) amplification of pIRES-YFP-hNaV1.1NT-R222* with the following primers: 5′-TTAAACTTAGATCTCCCGCCGCCAC-3′ forward and 5′-CACAATGGTTTTCAGGCCTGG-3′ reverse. The amplified fragment was then subcloned into the tagging vector pEYFP-C1 using BglI and SalI, obtaining a vector that expresses a chimeric protein with YFP fused to the N-terminus of hNaV1.1-R222*.

To make hNaV1.1 resistant to the specific blocker tetrodotoxin (TTX) we replaced the phenylalanine at position 383 with a serine (F383S) (Sivilotti et al., 1997), using pCDM8-hNaV1.1 and Quick Change XL with the following primes: forward 5′-CTCAGGACAGCTGGGAAAATCTTTATC-3′, reverse 5′-TTTCCCAGCTGTCCTGAGTCATTAG-3′.

pCDM8 constructs were propagated in TOP10/P3 or MC1061/P3 E. coli (Invitrogen-Life Technology, Grand Island, NY, U.S.A.) grown at 28°C for >48 h to minimize rearrangements; other plasmids were propagated in TOP10 E. coli. The entire coding sequences of the channels were sequenced after each amplification in bacteria to rule out the presence of unwanted spurious mutations or rearrangements.

The human clone of KV11.2 (HERG2) K+ channel (GenBank sequence NM_030779.2) subcloned into the pCMV6-XL4 mammalian expression vector was bought from Origene Inc. (Rockville, MD, U.S.A.).

Cell culture and transfection

TsA-201 cells were cultured in modified Dulbecco’s medium and Hams-F12 mix supplemented with 10% fetal bovine serum and transiently transfected by using the CaPO4 method as described previously (Mantegazza & Cestele, 2005). For electrophysiology, cells, plated in 35 mm petri dishes, were transfected with 2 μg of pCDM8-hNaV1.1, pCDM8-hNaV1.2, or pCDM8-hNaV1.3 alone or with 2 μg of truncated constructs (pIRES-YFP-hNaV1.1R222* or pCDM8-hNaV1.1R1234*). Recordings were performed 18–36 h after the transfection. For western blots, cells were cultured in 10 cm petri dishes and transfected with 10 μg of pEYFP- hNaV1.1R222* or with 2.5 μg of pEYFP.

HEK293 cells stably expressing hNaV1.6 (GenBank sequence NM_014191.2) or hNaV1.1 were provided by Dr. Enzo Wanke (Department of Biotechnologies and Biosciences, University of Milano-Bicocca, Milan, Italy) (Oliveira et al., 2004), and cultured in modified Dulbecco’s medium supplemented with 10% fetal bovine serum. Stable cell lines were transfected with 2 μg of pIRES-YFP-hNaV1.1R222*, pCDM8-hNaV1.1R1234*, pCDM8-hNaV1.1F383S, or pCMV6-XL4 HERG2 using Lipofectamine 2000 (Invitrogen).

Western blot

Two days after the transfection with pIRES-YFP-hNaV1.1NT-R222* (R222* N tagged with YFP), tsA-201 cells were washed with PBS, detached using Tris-EGTA containing Complete Mini Protease Inhibitors tablets (Roche, Basel, Switzerland), and centrifuged at 5000 rpm at 4°C for 10 min. The cell pellet was resuspended in RIPA buffer (50 mm Tris, 10 mm EDTA, 150 mm NaCl, 0.1% SDS, 1.25% NP40, 0.5% sodium-deoxycholate, pH 8). After 20 min of lysis in ice, a centrifugation for 10 min at 4,500 g at 4°C was performed to remove the insoluble material. The total amount of proteins contained in the supernatant was quantified through a Bradford assay. Forty micrograms of proteins were separated on a 4–15% SDS-polyacrylamide gradient gel (BioRad, Hercules, CA, U.S.A.) and transferred to a nitrocellulose membrane (BioRad). The protein ladder was a PageRuler plus (Fermentas, Glen Burnie, MD, U.S.A.). Blotted membranes were incubated with a polyclonal anti-GFP antibody (AbCam, Cambridge, United Kingdom) diluted 1:2,000 (which detects also YFP), and then with an anti-rabbit HRP conjugated antibody diluted 1:4,000 (Invitrogen-Life Technology). The signal was visualized by chemiluminescent detection with the ECL Detection System (GE-Healthcare, Little Chalfont, United Kingdom). Experiments were done in triplicate with similar results.

Electrophysiology

Recordings of transfected tsA-201 or stable HEK293 cells were performed 18–36 h after transfection. Cells were selected visually by their fluorescence using a Zeiss Axiovert 100 (Zeiss, Oberkochen, Germany) or a Nikon Eclipse FN1 microscope (Nikon, Tokyo, Japan) equipped with selective epifluorescence filters for YFP (Chroma 41028; Chroma, Bellows Falls, VT, U.S.A.). If used alone, plasmids that do not express a fluorescent protein were cotransfected with the plasmid pEYFP-C1 using a protein-of-interest/YFP molar ratio of 10/1. The extracellular recording solution contained the following (in mm): 140 NaCl, 2 CaCl2, 2 MgCl2, 10 HEPES, pH adjusted to 7.4, with NaOH. The pipette recording solution contained the following (in mm): 195 NMDG (N-methyl-d-glucamine), 10 NaCl, 4 MgCl2, 5 EGTA, 10 HEPES, pH adjusted to 7.2 with H3PO4. For recordings of both HERG2 and hNaV1.6 currents in HEK-hNaV1.6 cells, we used an extracellular solution containing (in mm): 105 NaCl, 40 KCl, 1 MgCl2, 1.8 CaCl2, 10 HEPES, 10 Glucose, pH adjusted to 7.4 with NaOH. The pipette recording solution contained the following (in mm): 100 KAsp, 20 KCl, 2 MgCl2, 1 CaCl2, 10 EGTA, 10 HEPES, 50 glucose, pH adjusted to 7.2 with KOH.

Hippocampal neurons were acutely dissociated from P15 to P20 wild-type or heterozygote (−/+) NaV1.1 KO mice using standard procedures (Yu et al., 2006). Briefly, animals were anesthetized with isoflurane, the whole brain was rapidly isolated, and coronal slices (300 μm) were prepared using a Microm HM 650V vibroslicer (Microm-Thermo Scientific, Waltham, MA, U.S.A.) in ice-cold, low-sodium slicing solution containing (in mm): 249 sucrose, 2.5 KCl, 1.25 NaH2PO4; 10 d-glucose; 26 NaHCO3, 0.1 CaCl2, 2.9 MgSO4, 0.5 ascorbic acid, bubbled with 95%O2/5%CO2 mixture, final pH 7.3. Hippocampal slices were dissected and placed in a holding chamber at room temperature (22–25°C) filled with ACSF, containing (in mm): 126 NaCl, 2.5 KCl, 1.25 NaH2PO4, 10 d-glucose, 26 NaHCO3, 2 CaCl2, 2 MgCl2, 1 pyruvic acid, bubbled with 95%O2/5%CO2 mixture; slices were used within 6 h of slicing. After allowing to recover for at least 1 h, one slice at a time was digested with 1 mg/ml proteases type XIV (Sigma-Aldrich, St. Louis, MO, U.S.A.) at 37°C in Hank’s balanced salt solution (HBSS; Sigma-Aldrich) to which 0.35 g/L of NaHCO3 was added: final pH = 7.3 bubbling with 95% O2/5% CO2. After 30 min of digestion, mechanical dissociation was performed in Na-isethionate solution containing (in mm): 140 Na-isethionate, 23 glucose, 15 HEPES, 2 KCl 4 MgCl2, 0.1 CaCl2; pH 7.2 with NaOH. The cell suspension was placed in a 35 mm tissue culture dish mounted on the recording rig. After allowing the cells to settle for 5 min, the solution was changed to a low Na+ extracellular recording solution containing (in mm): 20 NaCl, 116 glucose, 10 Hepes, 1 BaCl2, 2 MgCl2, 55 CsCl, 1 CdCl2, 20 tetraethylammonium chloride; pH 7.35 with NaOH. The intracellular recording solution contained (in mm): 120 NDMG, 40 Hepes, 4 MgCl2, 10 EGTA; the day of the experiment 25 P-creatine, 2 Na2-ATP and 0.1 Na2-GTP were added and pH was adjusted to 7.2 with acetic acid.

Whole cell patch-clamp recordings were performed at room temperature (22–25°C) using an Axon Multiclamp 700B amplifier (Molecular Devices, Union City, CA, U.S.A.) or a VE-2 amplifier (Alembic Instruments, Montreal, QC, Canada); pClamp 10.2 software (Molecular Devices) was used for stimulation and acquisition. Pipette resistance was between 1.5 and 2.0 MΩ (series resistance between 2.5 and 4.5 MΩ) for cultured cells, and between 3 and 4 MΩ (series resistance between 4.5 and 8 MΩ) for dissociated neurons. Cell capacitance and series resistance were carefully compensated (approximately 85%) before each run of the voltage clamp protocol; maximum accepted voltage clamp error was 2.5 mV. Remaining linear capacity and leakage currents were eliminated by the P/N leak subtraction procedure (mixed hNaV1.6/KV11.2 currents were not leak subtracted). Current signals were filtered at 10 kHz and sampled at 50 or 100 kHz.

The current–voltage (I–V) relationships were obtained by applying depolarizing pulses from a −90 mV holding potential. Activation kinetics was quantified by measuring the time of half activation at −10 mV; kinetics of fast inactivation was quantified by fitting the sum of two exponential functions to the current decay at −10 mV. The conductance–voltage (g–V) relationships (activation curves) were calculated from the I–V relationships according to g = INa/(V − ENa), where INa is the peak Na+ current measured at potential, V, and ENa is the calculated Na+ equilibrium potential (69 mV). The normalized activation and inactivation curves were fit to Boltzmann relationships in the form y = 1/{1 + exp[(V−V½)/k]}, where y is normalized GNa or INa, V, the membrane potential, V, the voltage of half-maximal activation (Va) or inactivation (Vh), and k is a slope factor. The inactivation protocol was formed by a test pulse to 0 mV, preceded by 100-ms–long prepulses at depolarized potentials; holding potential was −90 mV. Recovery from inactivation was studied with a test pulse to 0 mV, preceded by a 100-ms long prepulse to 0 mV and by increasingly longer repolarizations to the holding potential (−90 mV), and was fit with a single exponential relationship. The percentage of unrecovered current observed with this protocol (IUNREC in Table 2, which was presumably caused by the development of slow inactivation induced by the 100 ms-long prepulse to 0 mV) was calculated considering the fraction of current that did not recover after 20 ms at −90 mV (plateau of the curve of recovery from fast inactivation: see Fig. S2).

For HERG2 channels, a steady-state activation curve was obtained by plotting the peak tail currents elicited at −110 mV, after preconditioning for 4 or 15 s at potentials ranging from −60 to +40 mV.

Data analysis

Data were analyzed with pCLAMP 10.2 and Origin 7.5 (OriginLab, Northampton, MA, U.S.A.). Fits were achieved using the Levenberg-Marquardt algorithm with Origin. The results are given as mean ± standard error of the mean (SEM); statistical significance was assessed with Origin using the Student’s t-test, and threshold p-value for statistical significance was 0.05.

Results

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

We engineered in the shorter splice variant of hNaV1.1 (−11aa) the mutations p.R222* and p.R1234* (R1245* in the longer hNaV1.1 splice variant) that introduce premature termination codons and were identified in DS patients (originally reported as R222X and R1245X; Nabbout et al., 2003). The same hNaV1.1 clone has been used in several other functional studies (Mantegazza et al., 2005; Rusconi et al., 2007; Cestele et al., 2008; Rusconi et al., 2009). We selected these mutants in order to test a mutation that deletes most of the protein (R222* generates a truncated N-terminal region with just three transmembrane segments; Fig. 1C) and one that causes a comparatively smaller deletion (R1234* removes two of the four domains of the channel; Fig. 1B). Transcripts containing premature termination codons are often degraded by nonsense-mediated mRNA decay and do not produce proteins (Amrani et al., 2006). However, expression of hNaV1.1 mutants bearing larger truncations than R222* has already been confirmed by western blot (Sugawara et al., 2003); we confirmed the expression of R222* in tsA-201 cells (Fig. 1C). Therefore, premature termination codons do not block expression of NaV1.1 truncated mutants, similarly to what was found for truncated mutants of other proteins (Dreumont et al., 2005). We studied the effects of the two mutants on expression and functional properties of the main α subunit isoforms expressed in the CNS: hNaV1.1, hNaV1.2, hNaV1.3 and hNaV1.6 (Vacher et al., 2008; Mantegazza et al., 2010a).

Effect of hNaV1.1-R222* and hNaV1.1-R1234* on coexpressed hNaV1.1

Effects of truncated hNaV1.1 mutants on wild-type hNaV1.1 are potentially relevant in DS, because they are certainly coexpressed in the same cell types with equivalent (allelic) promoters. We expressed wild-type and mutant channels transfecting tsA-201 cells, a standard ion channel expression system (Mantegazza et al., 2010b). We initially did control experiments comparing currents observed expressing wild-type hNaV1.1 alone with those obtained expressing truncated mutants alone (Fig. 2A). With truncated mutants we recorded just small endogenous currents, which are present also in untransfected cells, whereas wild-type currents were much larger (>50 pA/pF with wild-type hNaV1.1, <15 pA/pF with truncated mutants). Therefore, we confirmed that truncated mutants do not generate Na+ current. Next, we compared the properties of the Na+ currents recorded in cells transfected with hNaV1.1 alone with those observed with cells cotransfected with hNaV1.1 and hNaV1.1-R222* or hNaV1.1-R1234* (Table 1, Fig. 2). We observed no significant differences in the kinetics of activation or inactivation of the transient Na+ current recorded in cotransfected cells in comparison with those expressing hNaV1.1 alone (Table 1), as shown by the comparison of the mean current traces elicited by a depolarizing test pulse to −10 mV, displayed in Fig. 2B. Current density-voltage plots (which compare current amplitude normalizing for the plasma-membrane surface of the cells) are shown in Fig. 2C: The inward Na+ current was maximal at approximately −5 mV and its amplitude was comparable when hNaV1.1 was expressed alone or with the truncated mutants. Moreover, voltage-dependence of activation (Fig. 2D) and inactivation (Fig. 2E) was similar in all the conditions. However, we found a small but significant slowing of the kinetics of recovery from inactivation induced by a 100-ms inactivating prepulse at 0 mV when we cotransfected hNaV1.1 with the mutants (Fig. 2F) (p = 0.003 for hNaV1.1-R222*; p = 0.004 for hNaV1.1-R1234*). Figure 2G shows mean I–V plots of persistent Na+ current (INaP) obtained applying test pulses to membrane potentials between −60 and +20 mV and quantifying INaP as the mean current remaining between 45 and 55 ms after the beginning of the voltage step. INaP amplitude was maximal at approximately −10 mV in all the conditions, and no significant differences in amplitude were found over the range of the tested potentials. These results show that truncated mutants can slow down recovery from fast inactivation, consistently with a small loss of function of coexpressed hNaV1.1 and induction of hypoexcitability.

image

Figure 2.   Effects of the DS truncated mutants hNaV1.1-R222* and hNaV1.1-R1234* on hNaV1.1 properties. (A) Representative Na+ current traces recorded with depolarizing voltage steps between −60 and +15 mV (5 mV increments from a holding potential of −100 mV) in tsA-201 cells transfected with hNaV1.1 (upper panel) or hNaV1.1-R222* (lower panel). (B) Mean normalized currents (solid black line, hNaV1.1; dashed orange line, hNaV1.1 + hNaV1.1-R222*; dotted blue line, hNaV1.1 + hNaV1.1-R1234*) elicited with a depolarizing step to −10 mV. (C) Current density–voltage plots for hNaV1.1 alone (black spheres), hNaV1.1 coexpressed with hNaV1.1-R222* (open orange squares), or with hNaV1.1-R1234* (open blue triangles). (D) Mean voltage dependence of activation; the lines are Boltzmann relationships, the parameters of which were obtained averaging the parameters of the fits of the single cells (see Table 1). (E) Mean voltage dependence of inactivation; the lines are mean Boltzmann relationships obtained from the fits of the single cells. (F) Mean kinetics of recovery at −90 mV from a 100 ms inactivating pulse to 0 mV; the lines are mean single exponential relationships obtained from the fits of the single cells. (G) Mean current–voltage plots of the persistent Na+ current (INaP) plotted as percentage of the maximal transient current. Data are presented as mean ± SEM. Recovery from fast inactivation was slower with both hNaV1.1-R222* and hNaV1.1-R1234* (Table 1).

Download figure to PowerPoint

Table 1.   Functional properties of Na+ currents recorded in cells lines
 CD (pA/pF)Va (mV)ka (mV)Vh (mV)kh (mV)0.5 Act (ms)τIna fast (ms)τRec (ms)INaP (% INaT)
  1. CD, current density; Va, voltage of half-maximal activation; ka, slope factor of the activation curve; Vh, voltage of half-maximal inactivation; kh, slope factor of the inactivation curve; 0.5 Act, time of half activation (50% of the rising phase of the current at −10 mV); τIna fast, fast time constant of inactivation derived from two-exponential fit to the decay of the current (representing >85% of the total decay); τRec, time constant of the single exponential fit to the recovery from inactivation; INaP is shown as % of maximal INaT.

  2. Functional properties of hNaV1.1, hNaV1.2, hNaV1.3, and hNaV1.6 expressed alone or coexpressed with hNaV1.1-R222* or hNaV1.1-R1234* are shown; data are presented as mean ± SEM.

  3. Statistically significant results are *p < 0.05; **p < 0.01.

hNaV1.171 ± 11 (n = 27)−20.5 ± 0.5 (n = 26)6.4 ± 0.2−55.7 ± 1.0 (n = 13)6.7 ± 0.30.25 ± 0.05 (n = 20)0.35 ± 0.06 (n = 20)2.8 ± 0.3 (n = 12)1.2 ± 0.2 (n = 12)
+R222*76 ± 9 (n = 12)−18.9 ± 0.7 (n = 11)6.9 ± 0.2−57.8 ± 1.1 (n = 11)7.1 ± 0.60.25 ± 0.07 (n = 11)0.43 ± 0.09 (n = 11)4.5 ± 0.4* (n = 8)0.9 ± 0.2 (n = 9)
+R1234*75 ± 25 (n = 19)−19.2 ± 0.6 (n = 16)6.6 ± 0.1−54.8 ± 0.7 (n = 10)7.0 ± 0.50.27 ± 0.06 (n = 16)0.45 ± 0.08 (n = 16)3.98 ± 0.06** (n = 11)1.1 ± 0.2 (n = 9)
hNaV1.2209 ± 19 (n = 40)−23.6 ± 0.5 (n = 38)5.3 ± 0.2−63.0 ± 0.3 (n = 13)5.5 ± 0.10.27 ± 0.05 (n = 20)0.41 ± 0.06 (n = 20)4.9 ± 0.2 (n = 25)0.9 ± 0.1 (n = 25)
+R222*194 ± 37 (n = 15)−23.8 ± 0.9 (n = 11)5.9 ± 0.3−63.7 ± 0.7 (n = 11)5.7 ± 0.20.25 ± 0.08 (n = 11)0.37 ± 0.09 (n = 11)4.8 ± 0.4 (n = 10)1.1 ± 0.4 (n = 5)
+R1234*221 ± 29 (n = 17)−24.4 ± 0.7 (n = 16)5.0 ± 0.2−62.4 ± 0.6 (n = 10)5.6 ± 0.20.29 ± 0.06 (n = 16)0.45 ± 0.08 (n = 16)4.5 ± 0.2 (n = 12)1.0 ± 0.1 (n = 7)
hNaV1.3145 ± 33 (n = 16)−20.0 ± 1.8 (n = 11)5.7 ± 0.3−62.7 ± 2.6 (n = 10)6.8 ± 0.40.35 ± 0.06 (n = 7)0.8 ± 0.2 (n = 7)3.7 ± 0.8 (n = 13)3.8.±1.0 (n = 5)
+R222*144 ± 30 (n = 12)−18.5 ± 1.9 (n = 8)5.4 ± 0.5−62.0 ± 3.5 (n = 6)7.0 ± 0.40.37 ± 0.08 (n = 7)1.0 ± 0.2 (n = 7)5.2 ± 0.3** (n = 11)2.2 ± 0.6 (n = 5)
+R1234*174 ± 27 (n = 16)−20.0 ± 1.2 (n = 12)5.3 ± 0.3−64.0 ± 2.2 (n = 10)6.0 ± 0.40.32 ± 0.06 (n = 9)0.6 ± 0.1 (n = 9)3.7 ± 0.7 (n = 13)2.8 ± 0.5 (n = 7)
hNaV1.6198 ± 13 (n = 14)−21.2 ± 3.3 (n = 7)4.2 ± 0.7−49.0 ± 4.2 (n = 7)7.6 ± 0.80.27 ± 0.08 (n = 7)0.4 ± 0.2 (n = 7)3.3 ± 0.7 (n = 7)1.5 ± 0.2 (n = 7)
+R222*140 ± 13** (n = 10)−21.0 ± 2.5 (n = 7)5.0 ± 0.7−52.3 ± 3.3 (n = 7)7.8 ± 0.80.29 ± 0.07 (n = 7)0.4 ± 0.1 (n = 7)4.2 ± 0.5** (n = 7)1.9 ± 0.2 (n = 7)
+R1234*148 ± 19* (n = 11)−21.3 ± 0.9 (n = 7)4.8 ± 0.5−56.3 ± 4.1** (n = 7)8.3 ± 1.60.23 ± 0.08 (n = 7)0.4 ± 0.2 (n = 7)4.9 ± 0.8** (n = 7)1.6 ± 0.1 (n = 7)

Effects of hNaV1.1 DS truncated mutants on coexpressed hNaV1.2

As already shown for CaV channels, dominant negative truncated mutants can suppress expression of different isoforms within the same gene family (Page et al., 2004; Mezghrani et al., 2008). NaV channels share high sequence homology; therefore, NaV1.1 truncated mutants may suppress expression of other NaV isoforms. NaV1.2 is expressed in both unmyelinated and myelinated glutamatergic neurons. In myelinated neurons expression of NaV1.2 deceases during postnatal development, whereas expression of NaV1.6 increases. Therefore, there is a switch between NaV1.2 and naV1.6 in these neurons during postnatal development (adult neurons express basically only NaV1.6) (Boiko et al., 2003; Vacher et al., 2008; Liao et al., 2010). However, some expression of NaV1.2 is maintained in these neurons also in adulthood, where it has a role in generating action potentials that back-propagate into the somatodendritic compartment (Hu et al., 2009). Moreover, it is the Na+ channel most often found concentrated near presynaptic terminals in cortical areas, where it can play an important role for generating presynaptic excitability (Vacher et al., 2008). Therefore, NaV1.1 and NaV1.2 can be coexpressed in several types of neurons. However, our data show that coexpression of hNaV1.1 truncated mutants has no effect on current amplitude or gating properties of hNaV1.2 (see Table 1 and Fig. S1).

Effects of hNaV1.1 DS truncated mutants on coexpressed hNaV1.3

In rodents, NaV1.3 is expressed abundantly in the somatodendritic compartment of central nervous system (CNS) neurons during fetal and early postnatal development, but its expression declines as that of NaV1.1 increases in the second postnatal week (Beckh et al., 1989; Vacher et al., 2008). These data are consistent with a complementary expression of NaV1.1 and NaV1.3 during development. However, as mentioned in the introduction, it has been observed that NaV1.3 is specifically upregulated in hippocampal GABAergic interneurons of NaV1.1 KO mice, as a compensatory effect for the loss-of-function of NaV1.1 channels (Yu et al., 2006). Notably, NaV1.3 is upregulated in several pathologies of neuronal excitability (Mantegazza et al., 2010a). Moreover, expression of NaV1.3 in humans extends to adulthood, with cellular and subcellular distribution similar to those found in rodents (Chen et al., 2000; Whitaker et al., 2001; Vacher et al., 2008). Therefore, NaV1.1 truncated mutants can be coexpressed also with NaV1.3 in some neurons. We studied whether hNaV1.1-R222* or hNaV1.1-R1234* can modify the properties of coexpressed hNaV1.3 in tsA-201 cells (Table 1, Fig. 3). Kinetics of activation and current decay were not significantly modified by coexpression of truncated mutants (Table 1), although with coexpression of hNaV1.1-R1234* there was a trend toward a faster current decay, as it is shown by the comparison of the average mean currents elicited with a step pulse to −10 mV (Fig. 3B). Current density-voltage plots (Fig. 3C) showed no differences in Na+ current amplitude, which was maximal at approximately −10 mV in all the conditions. Voltage dependence of activation (Fig. 3D) and inactivation (Fig. 3E) was not modified. However, recovery from fast inactivation (Fig. 3F) was slightly but significantly slowed down with coexpression of hNaV1.1-R222* (p < 0.001); notably, coexpression of hNaV1.1-R1234* did not induce this effect. Analysis of the plot of the percentage of INaP over a range of potentials (Fig. 3G) did not show any significant differences. Therefore, similarly to hNaV1.1, the only hNaV1.3 functional property modified was recovery from fast inactivation, but only with coexpression of hNaV1.1-R222*.

image

Figure 3.   Effects of hNaV1.1-R222* and hNaV1.1-R1234* on hNaV1.3 properties. (A) Representative Na+ current traces recorded from tsA-201 cells transfected with hNaV1.3 applying depolarizing voltage steps from −60 to +20 mV, 10 mV increment from a holding potential of −100 mV. (B) Mean normalized Na+ currents elicited by a depolarizing voltage step to −10 mV from a holding potential of −100 mV for hNaV1.3 (solid line), hNaV1.3 + hNaV1.1-R222* (dashed orange line), hNaV1.3 + hNaV1.1-R1234* (dashed-dotted blue line). (C) Current density-voltage relationships for hNaV1.3 alone (black stars) or coexpressed with hNaV1.1-R222* (open orange squares) or hNaV1.1-R1234* (open blue triangles). (D) Mean voltage dependence of activation; the lines are mean Boltzmann relationships, the parameters of which were calculated averaging the parameters of the fits of the single cells. (E) Mean voltage dependence of steady-state inactivation; the lines are mean Boltzmann relationships. (F) Mean kinetics of recovery at −90 mV from a 100 ms inactivating pulse to 0 mV; the lines are mean-fits of single exponential relationships obtained from the fits of the single cells. (G) Mean current–voltage plots of the persistent Na+ current (INaP) plotted as percentage of maximal transient current. Data are presented as mean ± SEM. Recovery from inactivation was slower with hNaV1.1-R222* (Table 1).

Download figure to PowerPoint

Effects of hNaV1.1 DS truncated mutants on coexpressed hNaV1.6

NaV1.6 is widely expressed in the brain and is the principal Na+ channel isoform expressed in adult myelinated neurons (Vacher et al., 2008). It is coexpressed with NaV1.1 in several types of neurons, for example in cerebellar Purkinje neurons (Kalume et al., 2007) and probably also in hippocampal pyramidal ones, although the expression level of NaV1.1 in these neurons is likely low (Yu et al., 2006; Vacher et al., 2008). Because hNaV1.6 cDNA was not available, we studied the effects of coexpression with truncated NaV1.1 DS mutants using a HEK293 cell line stably expressing hNaV1.6 (HEK-hNaV1.6; Oliveira et al., 2004) (Fig. 4 and Table 1); green fluorescent protein was used as a reporter of transfection.

image

Figure 4.   Effects of hNaV1.1-R222* and hNaV1.1-R1234* on hNaV1.6. (A) Representative Na+ current traces recorded with depolarizing voltage steps between −60 and +15 mV (5 mV increments from a holding potential of −100 mV) in HEK293 cells stably expressing hNaV1.6 (HEK-hNaV1.6 stable cell line) transfected with the control plasmid pCDM8. (B) Mean normalized current traces (solid line, hNaV1.6 + pCDM8; dashed orange line, hNaV1.6 + hNaV1.1 R222*; dash-dotted blue line, hNaV1.6 + hNaV1.1-R1234*) elicited with a depolarizing voltage step to −10 mV. (C) Current density–voltage plots for HEK-hNaV1.6 cells transfected with pCDM8 (black pentagons), hNaV1.1-R222* (open orange squares) or with hNaV1.1-R1234* (open blue triangles); both the mutants significantly reduced hNaV1.6 current density (Table 1). (D) Mean voltage dependence of activation; the lines are mean fits with Boltzmann relationships, the parameters of which were calculated averaging the parameters of the fits to the experimental data of the single cells. (E) Mean voltage dependence of inactivation; the lines are mean fits with Boltzmann relationships; hNaV1.1-R1234* leftward shifted hNaV1.6 inactivation curve. (F) Mean kinetics of recovery from a 100 ms inactivating pulse to 0 mV; the lines are mean fits of single exponential relationships to the data. (G) Bar graph displaying mean maximum current density in HEK-hNaV1.6 cells transfected with pCDM8 for control (1.6 stable, black bar; 251 ± 23 pA/pF, n = 8; different series of experiments in comparison with those shown above) or with the TTX resistant mutant hNaV1.1-F383S (+1.1-R; 185 ± 22 pA/pF, n = 7; p = 0.02); in these cells, hNaV1.1-F383S selective current was recorded upon perfusion with TTX 500 nm (+1.1-R +TTX; 70 ± 11 pA/pF; p = 0.009 in comparison with hNaV1.6 in control) and hNaV1.6 selective current was obtained subtracting the current recorded after perfusion with TTX (hNaV1.1-F383S current) from that recorded before perfusion with TTX (Δ+/− TTX; 115 ± 14 pA/pF; p < 0.005). (H) Bar graph of the mean maximum Na+ current density in HEK-hNaV1.6 cells transfected with hKV11.2 (HERG2) K+ channel in order to study the effect of the expression of a K+ channel on hNaV1.6 current (287 ± 44 pA/pF, n = 6, in control in a different series of experiments in comparison with those shown above; 294 ± 36 pA/pF, n = 8, with coexpression of hKV11.2). (I) Bar graph of the mean maximum Na+ current density in HEK293 cells stably expressing hNaV1.1 (HEK-hNaV1.1 stable cell line) transfected with pCDM8 for control (1.1 stable, black bar; 155 ± 22 pA/pF, n = 12) or with the truncated mutant hNaV1.1-R222*, which did not induce a significant modification of Na+ current density (+R222*; 120 ± 25 pA/pF, n = 10). Data are presented as mean ± SEM.

Download figure to PowerPoint

Figure 4A shows representative Na+ current traces recorded applying depolarizing voltage steps between −60 and +15 mV in control conditions (transfection with empty pCDM8). Figure 4B shows the comparison of the mean current traces elicited by a depolarizing test pulse to −10 mV in the different conditions. We found no significant differences in the kinetics of activation or in the current decay of the transient Na+ current (Table 1). Moreover, Na+ current amplitude was maximal at approximately −5 mV in all the conditions, as it is shown by the current density-voltage plots in Fig. 4C. However, we found an about 30% reduction of current density in cells transfected with the truncated mutants in comparison with those transfected with the control empty plasmid (hNaV1.1-R222*: p = 0.005; hNaV1.1-R1234*: p = 0.03). The voltage dependence of activation (Fig. 4D) was similar in all the conditions. Conversely, the voltage dependence of inactivation (Fig. 4E) showed a significant 7 mV negative shift with coexpression of hNaV1.1-R1234* (p = 0.006), but with hNaV1.1-R222* the shift was not statistically significant. Therefore, hNaV1.1-R1234* could induce a stabilization of the inactivated state of hNaV1.6. Furthermore, similarly to the effects observed with hNaV1.1 and hNaV1.3 described earlier, we found a significant deceleration of the recovery from fast inactivation (Fig. 4F) in cells transfected with the truncated mutants (hNaV1.1-R222*: p = 0.001; hNaV1.1-R1234*: p = 0.002). INaP was maximal at around −5 mV in all the conditions and there was no significant difference in its amplitude (not shown in Fig. 4, see Table 1). The effects that we have observed are consistent with a loss of function of hNaV1.6 when hNaV1.1 truncated mutants are coexpressed, and thus with cellular hypoexcitability. However, at least some of these effects, in particular reduction of current amplitude, could depend on the experimental system. In fact, proteins stably expressed in cell lines (NaV1.6 in this case) have often lower expression levels than those expressed with transient transfections (truncated mutants in this case). Therefore, down-regulation of hNaV1.6 might be an unspecific effect caused by the difference in the expression levels. To test this hypothesis and find out if the effect is selective for hNaV1.6, we carried out further experiments. In a first series of experiments, we investigated if the observed effects are induced specifically by truncated hNaV1.1 mutants. In order to study the effect of coexpression of the functional full length hNaV1.1, we transfected HEK-hNaV1.6 cells with an engineered hNaV1.1 channel (hNaV1.1-F383S) that is resistant to the specific blocker TTX (Sivilotti et al., 1997) (Fig. 4G). In these conditions, the current selectively generated by hNaV1.1-F383S (indicated as +1.1-R +TTX in Fig. 4G) can be quantified in each cell upon block of hNaV1.6 current by perfusion with TTX (500 nm), whereas hNaV1.6 current (indicated as Δ+/− TTX in Fig. 4G) can be obtained by subtracting the hNaV1.1-F383S current obtained with TTX from the total current obtained before perfusion with TTX (hNaV1.6 + hNaV1.1-F383S mixed current, indicated as +1.1-R in Fig. 4G). Notably, the hNaV1.6 mean current density obtained from HEK-hNaV1.6 cells transfected with hNaV1.1-F383S (Fig. 4G) showed a 45% reduction in comparison with that recorded from HEK-hNaV1.6 cells in control (p < 0.005). Therefore, coexpression of a functional full-length hNaV1.1 impairs the expression of hNaV1.6 in this cell line, similarly to what we observed with the truncated mutants. We then performed a further series of experiments to find out whether the reduction depends on the expression in stable cell lines and to rule out other unspecific effects. We studied the effect of coexpression of another plasma-membrane protein on hNaV1.6 current by transfecting HEK-hNaV1.6 cells with hKV11.2 (HERG2) K+ channels. We initially used solutions that allow for recording of both Na+ and K+ currents to confirm that hKV11.2 and hNaV1.6 were actually coexpressed, and we observed functional coexpression in all of the cells tested (Fig. S2; hKV11.2 maximal tail current density was 215 ± 75 pA/pF, n = 5; perfusion with 500 nm TTX blocked completely the Na+ current but did not modify the K+ current, n = 3, not shown). Subsequently, we quantified hNaV1.6 current density using recording solutions that are specific for Na+ currents (those used in the previous experiments). Notably, we observed no significant differences in hNaV1.6 current density between HEK-hNaV1.6 cells in control and those transfected with hKV11.2 (Fig. 4H). Moreover, we tested the effect of coexpression of hNaV1.1-R222* and hNaV1.1 in a HEK293 cell line stably expressing hNaV1.1 (HEK-hNaV1.1; Oliveira et al., 2004) (Fig. 4I), in which we did not observe a decrease of hNaV1.1 current density, similarly to the results that we obtained with transient transfections of tsA-201 cells. Therefore, both full length and truncated hNaV1.1 can specifically inhibit coexpressed hNaV1.6.

Is kinetics of recovery from fast inactivation also altered in mice expressing a truncated NaV1.1 mutant?

The most consistent effect of the truncated mutants that we observed on the gating properties in transfected cells was a small deceleration of recovery from inactivation of hNaV1.1, hNaV1.3, and hNaV1.6. Therefore, we asked if recovery from inactivation is modified in neurons from NaV1.1 KO mice, which express a truncated NaV1.1 mutant (Yu et al., 2006). In fact, the properties of recovery from inactivation have not been studied thus far in these mice. We dissociated hippocampal neurons from heterozygous NaV1.1 KO mice and their wild-type littermates (dissociated neurons limit space-clamp errors, allowing faithful recordings of Na+ currents), and recorded from both glutamatergic pyramidal and bipolar fusiform GABAergic neurons (Yu et al., 2006).

We observed in heterozygous NaV1.1 KO mice a significant decrease of Na+ current density selectively in GABAergic interneurons, and no modifications in the voltage dependence of activation or inactivation (Table 2). These data confirm the results obtained previously (Yu et al., 2006). Moreover, we found no differences in the kinetics of the decay of the current, as well as in the kinetics of the recovery from inactivation (Fig. S3). In all of the recorded neurons the recovery from inactivation reached an apparent steady state after 20 ms of repolarization at −90 mV, but the amplitude of the current did not fully recover to the original level (see insets in Fig. S3 and Methods for more details). This unrecovered fraction was probably caused by the development of slow inactivation induced by the inactivating pulse. However, the percentage of unrecovered current was not modified in heterozygous NaV1.1 KO mice. Overall, these results show that the expression of a truncated NaV1.1 mutant in hippocampal neurons of NaV1.1 KO mice does not modify the properties of Na+ current inactivation.

Table 2.   Functional properties of Na+ currents recorded in pyramidal (Pyr.) and bipolar (Bip.) neurons dissociated from wild type (wt) and heterozygous (Het) Nav1.1 KO mice
 CD (pA/pF)Va (mV)ka (mV)Vh (mV)kh (mV)τDECAY (ms)τREC (ms)IUNREC. (%)
  1. Va, voltage of half-maximal activation; ka, slope factor of the activation curve; Vh, voltage of half-maximal inactivation; kh, slope factor of the inactivation curve; τDECAY, time constant of the exponential fit of the current decay at 0 mV; τREC, time constant of the exponential fit of the recovery from inactivation at −90 mV; IUNREC, % of current that did not fully recover during the protocol of recovery of fast inactivation (see insets in Fig. S3).

  2. The only statistically significant difference is about maximal current density (CD) in bipolar neurons.

  3. The results are given as mean ± SEM; **p < 0.01.

wt Pyr.405 ± 28 (n = 18)−37.8 ± 0.2 (n = 16)6.7 ± 0.2−60.7 ± 0.2 (n = 16)9.4 ± 040.75 ± 0.03 (n = 16)5.6 ± 0.4 (n = 16)25.0 ± 2.0 (n = 16)
het Pyr.401 ± 37 (n = 13)−39.0 ± 0.5 (n = 10)7.0 ± 0.4−62.5 ± 0.2 (n = 10)8.9 ± 0.20.71 ± 0.04 (n = 12)5.7 ± 0.4 (n = 10)27.5.±2.4 (n = 10)
wt Bip.516 ± 52 (n = 15)−40.1 ± 0.5 (n = 16)7.5 ± 0.4−58.2 ± 0.2 (n = 16)9.5 ± 0.20.71 ± 0.02 (n = 18)5.5 ± 0.4 (n = 16)24.2 ± 2.2 (n = 16)
het Bip.330 ± 43** (n = 13)−40.3 ± 0.5 (n = 11)7.2 ± 0.4−59.0 ± 0.2 (n = 11)9.3 ± 0.10.77 ± 0.05 (n = 14)5.5 ± 0.3 (n = 11)25.0 ± 2.4 (n = 10)

Discussion

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

Our results show that there are no modifications of Na+ current amplitude in experiments of coexpression of hNaV1.1-R222* or hNaV1.1-R1234* with wild-type hNaV1.1, hNaV1.2, or hNaV1.3 done by means of transient transfections of tsA-201 cells. We observed a significant reduction in current amplitude expressing the truncated mutants in a HEK cell line in which hNaV1.6 is stably expressed. This effect was selective for hNaV1.6, because coexpression of a truncated mutant did not reduce hNaV1.1 current in a similar stable cell line. However, reduction of hNaV1.6 current was also induced by coexpression of functional full-length hNaV1.1, mimicking physiologic conditions, whereas control experiments done coexpressing KV11.2 did not show a reduction of NaV1.6 current. Therefore, this effect should not play a role in the pathogenic mechanism of DS. Although we have not investigated the molecular mechanism, it could be similar to that of dominant negative truncated CaV, which can be generated by premature stop codons introduced by physiologic alternative splicing or EA2 pathogenic mutations (Pietrobon, 2010). These truncated CaV decrease expression of cotransfected wild type CaV because of interactions between the N-terminus of truncated and wild type CaV (Page et al., 2010), which induce misfolding of the interacting proteins, and thus their targeting to the proteasome by the quality control system of the endoplasmic reticulum and subsequent degradation (Jeng et al., 2008; Mezghrani et al., 2008). Interestingly, folding defects can also be caused by EA2 missense mutations of CaV2.1 (Pietrobon, 2010), and by some epileptogenic missense mutations of NaV1.1 (Rusconi et al., 2007, 2009). Truncated CaV mutants can also induce reduction of protein synthesis by triggering the unfolded protein response (Page et al., 2004). CaV truncated mutants show a generalized cross-suppression between the different isoforms within the CaV2 or CaV3 family. Differently than for CaV our results show that there is no generalized cross-suppression between Nav isoforms, but a selective suppression of NaV1.6 by NaV1.1 and its truncated mutants. Perhaps, the suppression depends on the strength of the interaction between coexpressed proteins; hNaV1.1 truncated mutants possibly bind with sufficient strength only to hNaV1.6 and cannot activate this mechanism with other NaV.

Furthermore, we observed that gating properties of cotransfected wild-type NaV channels were in some cases modified by coexpressed hNaV1.1 truncated mutants. In particular, recovery from fast inactivation of hNaV1.1, hNaV1.3, and hNaV1.6 was slightly slower, and voltage dependence of hNaV1.6 inactivation was shifted toward negative potentials. Although small, these effects can reduce Na+ current and are consistent with a partial loss of function of coexpressed wild-type channels. Moreover, some truncated mutants may in theory compete with wild-type channels for binding to proteins that localize channels to specific subcellular domains (e.g., binding to ankyrin-G for localization in the axon initial segment). Accordingly, truncated mutants might have some negative dominant effects even if they do not inhibit expression of wild-type channels.

However, in order to induce these effects, truncated mutants should be inserted into the plasma-membrane and interact there with wild-type channels and other proteins. Two gene targeted mice have been generated that express truncated NaV1.1 mutants and show similar DS-like phenotypes: NaV1.1 KO mice, which we have used in our study (Yu et al., 2006), in which transmembrane segments 4–6 of domain IV and the C-terminus are deleted, and NaV1.1-R1407X knock-in mice (Ogiwara et al., 2007), bearing a larger deletion comprising transmembrane segment 6 of domain III, domain IV and the C-terminus. Notably, immunoblots of brain membrane fractions and immunostaining of brain sections from these two mouse lines have shown that truncated mutants are not found in the plasma-membrane. Consistent with these data, our recordings in neurons dissociated from NaV1.1 KO mice did not show any significant modifications in gating properties, although Na+ current amplitude was reduced in bipolar GABAergic neurons, confirming and extending previous results (Yu et al., 2006). Moreover, NaV1.1 is coexpressed with NaV1.6 in cerebellar Purkinje neurons, but, differently than in transfected cells, the voltage dependence of inactivation was not shifted in Purkinje neurons of NaV1.1 KO mice (Kalume et al., 2007). Therefore, the effects observed on gating properties of wild-type channels in transfected cells may be caused by the overexpression of the truncated mutants, which in these conditions can leak to the plasma membrane, whereas in vivo they are probably degraded before to reach it (Hirsch et al., 2009).

Hence, our conclusion is that NaV1.1 truncated mutants are not dominant negative and, when expressed at real pathophysiologic levels, are probably not inserted into the plasma membrane and thus do not alter gating properties of coexpressed wild-type channels: NaV1.1 DS truncating mutations induce pure haploinsufficiency. Notably, the effect observed on hNaV1.6, although physiologic, could be an obstacle for possible therapeutic approaches aimed at increasing expression levels of hNaV1.1, because upregulation of hNaV1.1 would reduce hNaV1.6 current in cells that coexpress the two channels (e.g., in Purkinje neurons of the cerebellum, which may aggravate ataxia).

It is interesting to note that wild-type channels may have interactions that are similar to those observed for truncated mutants and, different from truncated mutants, they are definitely inserted into the plasma membrane. Therefore, according to the results that we have obtained with truncated mutants, wild-type NaV1.1 may interact with coexpressed NaV1.1, NaV1.3, and NaV1.6, but not with NaV1.2 (because we have not seen any effect on this isoform in transfected cells). These effects appear isoform-specific and might lead to cross-modulation of the properties of coexpressed wild-type channels in vivo, although this should not be relevant for the pathomechanisms of DS, as highlighted above.

As for CaV channels, expression of truncated NaV channels has been observed in physiologic conditions. In fact, premature termination codons can sometimes be introduced by alternative splicing: a physiologic RNA processing that generates proteins with different sequences from a single gene transcript by differentially combining exons (Lipscombe, 2005). It has been proposed that introduction of premature termination codons can play a functional role in regulating the expression of several proteins, inhibiting the synthesis of full-length active ones (Lewis et al., 2003). For example, alternative splicing at exon 18 of NaV1.6 introduces a termination codon that generates a two-domain truncated channel, which is the predominant NaV1.6 variant in mouse fetal brain and is also expressed in human brain; this may be a mechanism for preventing expression of functional NaV1.6 (Plummer et al., 1997). Alternative splicing of NaV1.2 and NaV1.3 at exon 17 and of NaV1.7 at exon 16 can insert termination codons and generate two-domain truncated channels. Moreover, a termination codon can be inserted into NaV1.3 by alternative splicing at exon 11, generating a shorter one-domain truncated channel (Kerr et al., 2008). NaV1.1 undergoes alternative splicing at exon 5, which introduces three amino acid changes, and at exon 11, which generates −33 bp (−11aa, the variant used in our study) and −84 bp (−28aa) shorter variants, with deletions within the intracellular loop between DI and DII (Lossin, 2009). NaV1.1 variants truncated at different points in the third domain have been identified with reverse transcription-PCR experiments (Oh & Waxman, 1998; Alessandri-Haber et al., 2002), but it has not been investigated if these physiologic truncated variants can regulate the properties of wild-type NaV. Of interest, negative dominance of truncated variants may be in some cases species- or splice variant-specific, as it has been shown for CaV2.1 mutants (Pietrobon, 2010). Moreover, binding between α subunits and, consequently, negative dominance may be regulated by polymorphisms, as it has been shown for NaV1.5 (Poelzing et al., 2006).

Acknowledgments

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

We thank Dr. Jeff Clare (GlaxoSmithKline, Stevenage, Herts, United Kingdom) for providing the cDNAs of human NaV1.1 and NaV1.2, Dr. Enzo Wanke (Department of Biotechnologies and Biosciences, University of Milano-Bicocca, Milan, Italy) for providing HEK293 cells stably expressing hNaV1.6, Dr. Giuliano Avanzini (Besta Foundation Neurological Institute, Milan, Italy) for support, and Camille Liautard (IPMC, CNRS UMR6097 and University of Nice-Sophia Antipolis) for help with the genotyping of NaV1.1 KO mice. This study was supported by the European Integrated Project EPICURE EFP6-037315 (MM and SF), the Italian League Against Epilepsy (LICE) – Genetic committee (MM and SF), the Fédération pour la Recherche sur le Cerveau (MM), the Fondation pour la Recherche Medicale (MM), the ATIP/Avenir start up grant (MM), and the University of Nice-Sophia Antipolis CSI grant (MM).

Disclosure

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

We have no conflicts of interest and we confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

References

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information
  • Alessandri-Haber N, Alcaraz G, Deleuze C, Jullien F, Manrique C, Couraud F, Crest M, Giraud P. (2002) Molecular determinants of emerging excitability in rat embryonic motoneurons. J Physiol 541:2539.
  • Amrani N, Sachs MS, Jacobson A. (2006) Early nonsense: mRNA decay solves a translational problem. Nat Rev Mol Cell Biol 7:415425.
  • Beckh S, Noda M, Lubbert H, Numa S. (1989) Differential regulation of three sodium channel messenger RNAs in the rat central nervous system during development. EMBO J 8:36113616.
  • Boiko T, Van Wart A, Caldwell JH, Levinson SR, Trimmer JS, Matthews G. (2003) Functional specialization of the axon initial segment by isoform-specific sodium channel targeting. J Neurosci 23:23062313.
  • Cestele S, Scalmani P, Rusconi R, Terragni B, Franceschetti S, Mantegazza M. (2008) Self-limited hyperexcitability: functional effect of a familial hemiplegic migraine mutation of the Nav1.1 (SCN1A) Na+ channel. J Neurosci 28:72737283.
  • Chen YH, Dale TJ, Romanos MA, Whitaker WR, Xie XM, Clare JJ. (2000) Cloning, distribution and functional analysis of the type III sodium channel from human brain. Eur J Neurosci 12:42814289.
  • Claes L, Del Favero J, Ceulemans B, Lagae L, Van Broeckhoven C, De Jonghe P. (2001) De novo mutations in the sodium-channel gene SCN1A cause severe myoclonic epilepsy of infancy. Am J Hum Genet 68:13271332.
  • Claes LR, Deprez L, Suls A, Baets J, Smets K, Van Dyck T, Deconinck T, Jordanova A, De Jonghe P. (2009) The SCN1A variant database: a novel research and diagnostic tool. Hum Mutat 30:E904E920.
  • Dravet C, Bureau M, Oguni H, Fukuyama Y, Cokar O. (2005) Severe myoclonic epilepsy in infancy: Dravet syndrome. Adv Neurol 95:71102.
  • Dreumont N, Maresca A, Boisclair-Lachance JF, Bergeron A, Tanguay RM. (2005) A minor alternative transcript of the fumarylacetoacetate hydrolase gene produces a protein despite being likely subjected to nonsense-mediated mRNA decay. BMC Mol Biol 6:1.
  • Hirsch C, Gauss R, Horn SC, Neuber O, Sommer T. (2009) The ubiquitylation machinery of the endoplasmic reticulum. Nature 458:453460.
  • Hu W, Tian C, Li T, Yang M, Hou H, Shu Y. (2009) Distinct contributions of Na(v)1.6 and Na(v)1.2 in action potential initiation and backpropagation. Nat Neurosci 12:9961002.
  • Jeng CJ, Sun MC, Chen YW, Tang CY. (2008) Dominant-negative effects of episodic ataxia type 2 mutations involve disruption of membrane trafficking of human P/Q-type Ca2+ channels. J Cell Physiol 214:422433.
  • Kalume F, Yu FH, Westenbroek RE, Scheuer T, Catterall WA. (2007) Reduced sodium current in Purkinje neurons from Nav1.1 mutant mice: implications for ataxia in severe myoclonic epilepsy in infancy. J Neurosci 27:1106511074.
  • Kamiya K, Kaneda M, Sugawara T, Mazaki E, Okamura N, Montal M, Makita N, Tanaka M, Fukushima K, Fujiwara T, Inoue Y, Yamakawa K. (2004) A nonsense mutation of the sodium channel gene SCN2A in a patient with intractable epilepsy and mental decline. J Neurosci 24:26902698.
  • Kerr NC, Holmes FE, Wynick D. (2008) Novel mRNA isoforms of the sodium channels Na(v)1.2, Na(v)1.3 and Na(v)1.7 encode predicted two-domain, truncated proteins. Neuroscience 155:797808.
  • Lewis BP, Green RE, Brenner SE. (2003) Evidence for the widespread coupling of alternative splicing and nonsense-mediated mRNA decay in humans. Proc Natl Acad Sci U S A 100:189192.
  • Liao Y, Deprez L, Maljevic S, Pitsch J, Claes L, Hristova D, Jordanova A, Ala-Mello S, Bellan-Koch A, Blazevic D, Schubert S, Thomas EA, Petrou S, Becker AJ, De Jonghe P, Lerche H. (2010) Molecular correlates of age-dependent seizures in an inherited neonatal-infantile epilepsy. Brain 133:14031414.
  • Lipscombe D. (2005) Neuronal proteins custom designed by alternative splicing. Curr Opin Neurobiol 15:358363.
  • Lossin C. (2009) A catalog of SCN1A variants. Brain Dev 31:114130.
  • Mantegazza M. (2011) Dravet syndrome: insights from in vitro experimental models. Epilepsia 52(Suppl. 2):6269.
  • Mantegazza M, Cestele S. (2005) Beta-scorpion toxin effects suggest electrostatic interactions in domain II of voltage-dependent sodium channels. J Physiol 568:1330.
  • Mantegazza M, Gambardella A, Rusconi R, Schiavon E, Annesi F, Cassulini RR, Labate A, Carrideo S, Chifari R, Canevini MP, Canger R, Franceschetti S, Annesi G, Wanke E, Quattrone A. (2005) Identification of an Nav1.1 sodium channel (SCN1A) loss-of-function mutation associated with familial simple febrile seizures. Proc Natl Acad Sci USA 102:1817718182.
  • Mantegazza M, Curia G, Biagini G, Ragsdale DS, Avoli M. (2010a) Voltage-gated sodium channels as therapeutic targets in epilepsy and other neurological disorders. Lancet Neurol 9:413424.
  • Mantegazza M, Rusconi R, Scalmani P, Avanzini G, Franceschetti S. (2010b) Epileptogenic ion channel mutations: from bedside to bench and, hopefully, back again. Epilepsy Res 92:129.
  • Marini C, Mantegazza M. (2010) Sodium channelopathies and epilepsy: recent advances and new perspectives. Expert Rev Clin Pharmacol 3:371384.
  • Mezghrani A, Monteil A, Watschinger K, Sinnegger-Brauns MJ, Barrere C, Bourinet E, Nargeot J, Striessnig J, Lory P. (2008) A destructive interaction mechanism accounts for dominant-negative effects of misfolded mutants of voltage-gated calcium channels. J Neurosci 28:45014511.
  • Nabbout R, Gennaro E, Dalla BB, Dulac O, Madia F, Bertini E, Capovilla G, Chiron C, Cristofori G, Elia M, Fontana E, Gaggero R, Granata T, Guerrini R, Loi M, La Selva L, Lispi ML, Matricardi A, Romeo A, Tzolas V, Valseriati D, Veggiotti P, Vigevano F, Vallee L, Dagna BF, Bianchi A, Zara F. (2003) Spectrum of SCN1A mutations in severe myoclonic epilepsy of infancy. Neurology 60:19611967.
  • Oakley JC, Kalume F, Yu FH, Scheuer T, Catterall WA. (2009) Temperature- and age-dependent seizures in a mouse model of severe myoclonic epilepsy in infancy. Proc Natl Acad Sci USA 106:39943999.
  • Ogiwara I, Miyamoto H, Morita N, Atapour N, Mazaki E, Inoue I, Takeuchi T, Itohara S, Yanagawa Y, Obata K, Furuichi T, Hensch TK, Yamakawa K. (2007) Na(v)1.1 localizes to axons of parvalbumin-positive inhibitory interneurons: a circuit basis for epileptic seizures in mice carrying an Scn1a gene mutation. J Neurosci 27:59035914.
  • Oh Y, Waxman SG. (1998) Novel splice variants of the voltage-sensitive sodium channel alpha subunit. Neuroreport 9:12671272.
  • Oliveira JS, Redaelli E, Zaharenko AJ, Cassulini RR, Konno K, Pimenta DC, Freitas JC, Clare JJ, Wanke E. (2004) Binding specificity of sea anemone toxins to Nav 1.1-1.6 sodium channels: unexpected contributions from differences in the IV/S3-S4 outer loop. J Biol Chem 279:3332333335.
  • Page KM, Heblich F, Davies A, Butcher AJ, Leroy J, Bertaso F, Pratt WS, Dolphin AC. (2004) Dominant-negative calcium channel suppression by truncated constructs involves a kinase implicated in the unfolded protein response. J Neurosci 24:54005409.
  • Page KM, Heblich F, Margas W, Pratt WS, Nieto-Rostro M, Chaggar K, Sandhu K, Davies A, Dolphin AC. (2010) N terminus is key to the dominant negative suppression of Ca(V)2 calcium channels: implications for episodic ataxia type 2. J Biol Chem 285:835844.
  • Pietrobon D. (2010) CaV2.1 channelopathies. Pflugers Arch 460:375393.
  • Plummer NW, McBurney MW, Meisler MH. (1997) Alternative splicing of the sodium channel SCN8A predicts a truncated two-domain protein in fetal brain and non-neuronal cells. J Biol Chem 272:2400824015.
  • Poelzing S, Forleo C, Samodell M, Dudash L, Sorrentino S, Anaclerio M, Troccoli R, Iacoviello M, Romito R, Guida P, Chahine M, Pitzalis M, Deschenes I. (2006) SCN5A polymorphism restores trafficking of a Brugada syndrome mutation on a separate gene. Circulation 114:368376.
  • Raghib A, Bertaso F, Davies A, Page KM, Meir A, Bogdanov Y, Dolphin AC. (2001) Dominant-negative synthesis suppression of voltage-gated calcium channel Cav2.2 induced by truncated constructs. J Neurosci 21:84958504.
  • Rusconi R, Scalmani P, Cassulini RR, Giunti G, Gambardella A, Franceschetti S, Annesi G, Wanke E, Mantegazza M. (2007) Modulatory proteins can rescue a trafficking defective epileptogenic Nav1.1 Na+ channel mutant. J Neurosci 27:1103711046.
  • Rusconi R, Combi R, Cestele S, Grioni D, Franceschetti S, Dalpra L, Mantegazza M. (2009) A rescuable folding defective Nav1.1 (SCN1A) sodium channel mutant causes GEFS+: common mechanism in Nav1.1 related epilepsies? Hum Mutat 30:E747E760.
  • Schaller KL, Krzemien DM, McKenna NM, Caldwell JH. (1992) Alternatively spliced sodium channel transcripts in brain and muscle. J Neurosci 12:13701381.
  • Sivilotti L, Okuse K, Akopian AN, Moss S, Wood JN. (1997) A single serine residue confers tetrodotoxin insensitivity on the rat sensory-neuron-specific sodium channel SNS. FEBS Lett 409:4952.
  • Sugawara T, Tsurubuchi Y, Fujiwara T, Mazaki-Miyazaki E, Nagata K, Montal M, Inoue Y, Yamakawa K. (2003) Nav1.1 channels with mutations of severe myoclonic epilepsy in infancy display attenuated currents. Epilepsy Res 54:201207.
  • Vacher H, Mohapatra DP, Trimmer JS. (2008) Localization and targeting of voltage-dependent ion channels in mammalian central neurons. Physiol Rev 88:14071447.
  • Whitaker WR, Faull RL, Waldvogel HJ, Plumpton CJ, Emson PC, Clare JJ. (2001) Comparative distribution of voltage-gated sodium channel proteins in human brain. Brain Res Mol Brain Res 88:3753.
  • Yu FH, Mantegazza M, Westenbroek RE, Robbins CA, Kalume F, Burton KA, Spain WJ, McKnight GS, Scheuer T, Catterall WA. (2006) Reduced sodium current in GABAergic interneurons in a mouse model of severe myoclonic epilepsy in infancy. Nat Neurosci 9:11421149.

Supporting Information

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

Figure S1. Effects of hNav1.1R222X and hNav1.1R1234X on hNav1.2 properties.

Figure S2. Representative mixed Na+/K+ current traces recorded in HEK-hNaV1.6 cells co-expressing hKV11.2 (HERG2) K+ channel.

Figure S3. Time course of recovery from fast inactivation in hippocampal neurons dissociated from Nav1.1 heterozygous (het) knock-out mice and wild type (wt) littermates.

FilenameFormatSizeDescription
EPI_3346_sm_FigS1.docx652KSupporting info item
EPI_3346_sm_FigS2.docx248KSupporting info item
EPI_3346_sm_FigS3.docx3663KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.