Abstract
Voltage-gated sodium channels in the mammalian CNS initiate and propagate action potentials when excitatory inputs achieve threshold membrane depolarization. There are multiple sodium channel isoforms expressed in rat brain (types I, II, III, 6, and NaG). We have constructed a full-length cDNA clone encoding type I and compared the electrophysiological properties of type I (Rat1) and II (Rat2) channels in the absence and presence of the two accessory subunits β1 and β2. Injection intoXenopus oocytes of RNA encoding Rat1 resulted in functional sodium currents that were blocked by tetrodotoxin, with Kapp = 9.6 nm. Rat1 sodium channels had a slower time course of fast inactivation than Rat2. Coexpression of β1 accelerated inactivation of both Rat1 and Rat2, resulting in comparable inactivation kinetics. Rat1 recovered from fast inactivation more rapidly than Rat2, regardless of whether β1 or β2 was present. The voltage dependence of activation was similar for Rat1 and Rat2 without the β subunits, but it was more positive for Rat1 when β1 and β2 were coexpressed. The voltage dependence of inactivation was more positive for Rat1 than for Rat2, and coexpression with β1 and β2 accentuated that difference. Finally, sodium current amplitudes were reduced by 7–9% for both Rat1 and Rat2 channels when protein kinase A phosphorylation was induced. It has been suggested previously that Rat1 and Rat6 channels mediate transient and maintained sodium conductances, respectively, in Purkinje cells, and the electrophysiological properties of Rat1 currents are consistent with a role for this channel in mediating the rapidly inactivating, transient current.
Various sodium channel isoforms have been detected by molecular cloning, biochemical purification, and electrophysiological recording (for review, seeGoldin, 1995). Multiple isoforms have been identified in the rat CNS, including types I (Rat1) (Noda et al., 1986a), II (Rat2) (Noda et al., 1986a) and a splice variant termed Rat2A (Auld et al., 1988), III (Rat3) (Kayano et al., 1988; Joho et al., 1990), 6 (Rat6) (Schaller et al., 1995) and the species variant Scn8a (Burgess et al., 1995), and a partial cDNA sequence for NaG (Gautron et al., 1992). In addition to the pore-forming α subunit of the sodium channel, there are two accessory subunits termed β1 and β2(Hartshorne and Catterall, 1984). The β1 subunit modulates channel function by accelerating the kinetics of inactivation and shifting its voltage dependence in the hyperpolarizing direction (Isom et al., 1992). The β2 subunit is covalently bound to the α subunit, and it accelerates inactivation slightly (Isom et al., 1995).
The sodium channel isoforms in the CNS are present at different times during development and in different locations. Rat1 becomes detectable shortly after birth and increases until adulthood, Rat2 becomes detectable during embryonic development and reaches maximal levels during adulthood, Rat3 peaks at birth and becomes undetectable by adulthood (Beckh et al., 1989), and Rat6 peaks during late embryonic and early postnatal periods (Felts et al., 1997) but is also present at high levels during adulthood (Schaller et al., 1995). Levels of Rat2 are highest in the rostral regions of the CNS, Rat1 is the predominant channel in the caudal regions and the spinal cord (Gordon et al., 1987;Beckh et al., 1989), and there is no rostral–caudal gradient of Rat6 mRNA (Schaller et al., 1995). In the cerebellum, Rat1 is detectable in Purkinje cells but not in granule cells, Rat2 is expressed in both Purkinje (Black et al., 1994) and granule cells (Furuyama et al., 1993), and Rat6 is expressed predominantly in granule cells (Schaller et al., 1995). Rat1 is localized in the soma of neurons in various CNS regions, including the hippocampus, cerebellum, and spinal cord, whereas Rat2 is axonal in distribution (Westenbroek et al., 1989). Because there are multiple α sodium channel isoforms in the mature brain, it is conceivable that each has a distinct role in determining electrical excitability. For example, Rat1 and Rat6 channels have been predicted to mediate transient and maintained sodium conductances, respectively, in Purkinje cells (Vega-Saenz de Miera et al., 1997).
By examining the functional properties of the different sodium channel isoforms, it should be possible to gain a better understanding of how each isoform affects electrical excitability. The properties of the Rat2 (Auld et al., 1988) and Rat3 (Joho et al., 1990) sodium channels have been extensively characterized, but neither the Rat1 nor Rat6 channels have been expressed in an exogenous system. Attempts to characterize the properties of the original Rat1 clone inXenopus oocytes were unsuccessful (Noda et al., 1986a), and that clone has not been available for study. We therefore constructed a full-length clone encoding Rat1 and compared the functional properties of Rat1 and Rat2 channels in the absence and presence of β1 and β2 subunits.
MATERIALS AND METHODS
Isolation of rat brain RNA. Total rat brain RNA was isolated from 15 to 18-d-old rats by a modified lithium chloride/urea procedure (Dierks et al., 1981). RNA was suspended in sterile, RNase-free water at a concentration of 1 mg/ml and stored at −75°C.
Reverse transcription and PCR. The Rat1 coding region was amplified from total rat brain RNA by reverse transcription followed by PCR (RT-PCR) in two fragments, with the middle boundary defined by the unique Sphl site in the coding region. Primers for reverse transcriptase and PCR were designed based on the published sequence (Noda et al., 1986b). The four primers used were (A) 5′ end primer: 5′-GGCCATAT(GCGGCCGC)ATCAGGAATCTCACATGAAG-3′; (B) primer upstream of Sphl (5′): 5′-CTGGTGCTGGCCATCATCG-3′; (C) primer downstream of Sphl (3′): 5′-GCAGTCAGTGGCAATTTTGC-3′; and (D) 3′ end primer: 5′-GGCCATAT(GCGGCCGC)AGTCCTTTGA CTTCACAGG-3′. Each of the outer 5′ and 3′ primers contains a four-base G–C clamp and an extra four nucleotides to promote efficient cutting at the Notl sites (indicated in parentheses).
Total brain RNA (2 μg) was heat-denatured at 65°C for 5 min, followed by rapid cooling on ice. RT was performed with 0.5 mm deoxynucleotide triphosphates, 10 mmdithiothreitol, 100 pmol of oligonucleotide primers (C and D), 40 U RNasin (Promega, Madison, WI), and 500 U M-MLV reverse transcriptase (Life Technologies, Gaithersburg, MD) in a total volume of 50 μl. Reactions were incubated at room temperature for 5 min and then at 37°C for 2 hr. Reaction products were purified by phenol/chloroform extraction and ethanol precipitation and were resuspended in 10 μl of distilled water.
The two RT products were amplified using the primer pairs A–C (amino terminal portion of Rat1) and B–D (carboxy portion of Rat1). The RT product was combined with 2.5 mm MgCl2, 200 μmol deoxynucleotide triphosphates, 0.2 μm of each primer, and 2.5 U LA Taq Polymerase (PanVera, Madison, WI). Thermal cycle parameters for A–C primer amplification were one cycle consisting of denaturation at 95°C for 4 min, annealing at 50°C for 3 min, and polymerase extension at 72°C for 6 min, followed by 30 cycles consisting of denaturation at 95°C for 30 sec, annealing at 50°C for 1 min, and polymerase extension at 72°C for 6 min. Parameters for B–D amplification were the same except that annealing steps were at 55°C. These conditions resulted in two PCR fragments of size 3429 and 2851 bp for primer pairs B–D and A–C, respectively. These fragments were extracted with phenol/chloroform, precipitated with ethanol, and resuspended in distilled water before digestion with the appropriate restriction enzymes.
Construction of full-length Rat1 cDNA. The full-length cDNA for Rat1 was constructed by making a series of Rat1–Rat2 chimeras and then combining Rat1 sequences. Chimeras were constructed using AatII and SphI restriction sites that are common to both channels. The B–D PCR product (carboxy end of Rat1) was cloned into the corresponding region of Rat2 using SphI and NotI restriction sites to generate a chimeric Rat2–1 channel (written Rat2211, with each number corresponding to approximately one domain of the channel). This chimera expressed sodium current in oocytes (data not shown).
A Rat1–2 chimera (Rat1122) was constructed by ligation of the Rat1 amino terminal PCR product A–C (NotI–SphI) with the complementary Rat2 SphI–NotI fragment into the NotI site of pLCT1, a modified version of pBSTA (Goldin, 1991) that contains the gene for tetracycline resistance and an origin of replication from the plasmid pACYC184. An isolate was obtained that contained Rat1 sequence positioned downstream from the T3 promoter; however, RNA transcribed by T3 RNA polymerase did not express sodium current in oocytes. To transfer additional Rat1 sequence into Rat2211, the AatII–SphI region from Rat1 122 (domain 2) was incorporated to yield a Rat2 111 chimera. This chimera did express sodium current in oocytes.
We reasoned that the lack of expression by the Rat1122 construct was most likely attributable to three extra ATG start codons that followed the T3 promoter but were upstream and in a different reading frame than the authentic start codon. Therefore, the 5′ region was modified by a secondary round of PCR to exclude the three extra start codons and to attach an XhoI site at the 5′ end for subsequent ligation into the XhoI site in Rat2. The modification was accomplished using a new 5′ primer [GCGCGC(CTCGAG)TGACAAGATGGAGCAAAC (XhoI site in parentheses)] and a new primer downstream from the AatII restriction site CACACTGAGACAGAACACGG. The second round of PCR was performed using 2 ng of Rat1122 plasmid DNA, 20 μm deoxynucleotide triphosphates, 1 μm each primer, and 2.5 U Pfu DNA polymerase (Stratagene, La Jolla, CA). Thermal cycling parameters were as described for the initial RT-PCR, except that the annealing temperature was 57°C. The secondary PCR product (801 bp) was cut with XhoI and AatII and ligated into the corresponding sites in Rat2111 to construct the full-length Rat1 sequence (Rat 1111). Expression of sodium current was observed in oocytes after injection of RNA transcribed from this clone.
The sequence of the entire coding region was determined using the dideoxy chain termination method and the Taq dye terminator cycle sequencing kit with an Applied Biosystems sequencer Model 373 Stretch (Applied Biosystems). To increase the level of expression in oocytes, the Rat1 coding region was inserted into the BgIII site of pLCT1, which resulted in the 5′ and 3′ noncoding regions from the Xenopusβ-globin gene being positioned on either side of the Rat1 coding region and a poly-A tail at the 3′ end of the insert. The Rat1 coding region was inserted into pLCT1 by cutting it with XhoI and NotI, followed by generation of blunt ends with Klenow DNA polymerase, attachment of BgIII linkers, and ligation into the BgIII site of pLCT1.
Complementary DNA clones encoding the β1 subunit (Isom et al., 1992) and β2 subunit (Isom et al., 1995) were isolated from total rat brain RNA by RT-PCR, using published sequence information to design the following primers: β1(5′)-TCGAGATCTATGGGGACGCTGCT; β1(3′)GCCAGATCTATTCAGCCACCTGG; β2(5′)-GCATCGATGGCCTGAAAATGCACAGGGATGC; and β2(3′)-GCATCGATGGATCCGGACACAGGAAGGGGCTTC. RT and PCR conditions were similar to those used for Rat1, except that annealing temperatures were adjusted to 5°C below the melting temperature of each primer pair.
Expression and electrophysiology. RNA transcripts were synthesized from NotI linearized DNA templates using a T7 RNA polymerase Message Machine transcription kit (Ambion, Austin, TX). The yield of RNA was estimated by glyoxal gel analysis. Stage V oocytes were removed from adult female Xenopus laevis frogs and prepared as described previously (Goldin, 1991), and incubated in ND-96 media, which consists of (in mm) 96 NaCl, 2 KCl, 1.8 CaCl2, 1 MgCl2, and 5 HEPES, pH 7.5, supplemented with 0.1 mg/ml gentamicin, 0.55 mg/ml pyruvate, and 0.5 mm theophylline. Rat2 sodium channel RNA was injected at 100 pg/oocyte, and Rat1 RNA was injected at 50 ng/oocyte. The oocytes were incubated for 40 hr at 20°C in ND-96.
Sodium currents were recorded using the cut-open oocyte technique (Taglialatela et al., 1992) with the CA-1 high performance oocyte voltage clamp (Dagan, Minneapolis, MN) and Digidata 1200A interface (Axon Instruments, Foster City, CA) and pCLAMP 6.0.3 software (Axon Instruments), as described previously (Kontis et al., 1997). Temperature was maintained at 20°C using an HCC-100A temperature controller (Dagan). The intracellular solution consisted of (in mm) 88 K2SO4, 10 EGTA, 10 HEPES, 10 Na2SO4, pH 7.5, and the extracellular solution consisted of (in mm) 120 sodium MES, 10 HEPES, and 1.8 Ca–Cs, pH 7.4. Capacitive transients and leak currents were corrected by P/4 subtraction. Sodium current amplitudes were between 1 and 5 μA.
For analysis of recovery from inactivation and modulation by protein kinase A (PKA), a two-electrode voltage clamp was used at room temperature as described previously (Patton and Goldin, 1991). Although this voltage clamp does not provide the fast time resolution of the cut-open oocyte clamp, the oocytes are more stable over long periods of time, which was essential for analyzing these two channel properties. Capacitive and leak currents were eliminated from the recovery from inactivation records by subtraction of comparable records obtained in the presence of 400 nm tetrodotoxin, and from the PKA modulation records by P/4 subtraction. The bath solution consisted of ND-96. Oocytes were clamped at −100 mV for 5–10 min before recording to allow for recovery from slow inactivation. After steady-state current levels were established for 10 min, PKA was induced by perfusing oocytes with a mixture consisting of 25 μmforskolin, 10 μm chlorophenylthio-cAMP (cpt-cAMP), 10 μm dibutyryl-cAMP (db-cAMP), and 10 μm3-isobutyl-1-methylxanthine (IBMX) for 10 min. We have shown previously that this mixture reduces Rat2 sodium currents by activation of PKA (Smith and Goldin, 1997). The rate of bath perfusion was carefully adjusted to 0.3 ml/min to minimize fluctuations in current amplitude resulting from changes in flow rate. Sodium current amplitudes were measured every 60 seconds during depolarizations to −10 mV from a holding potential of −100 mV.
Data analysis. Analysis was performed using pCLAMP 6.0.3 software (Axon Instruments), Excel 7.0 (Microsoft, Redmond, WA), and Sigmaplot 4.0 (Jandel, San Rafael, CA). Inactivation time constants were determined using the Chebyshev method to fit current traces with a single exponential equation: I = Aslow· exp[−(t − K) /τslow] + C, or a double exponential equation: I = Afast · exp[−(t − K) /τfast] + Aslow· exp[−(t − K) /τslow] + C, where I is the current, Afast and Aslow represent the percentage of channels inactivating with time constants τfast and τslow, K is the time shift, and C is the steady-state asymptote. The time shift was manually selected by fitting the traces at the time when the currents were just starting to decrease exponentially. Recovery data were fit using a double or triple exponential equation of the form I = 1 − [A1 · exp(− t /τ1) + A2 · exp(− t /τ2)] and I = 1 − [ A1 · exp(− t /τ1) + A2 · exp(− t /τ2) + A3 · (− t /τ3)], where A1, A2, and A3 are the relative proportions of current recovering with time constants τ1, τ2, and τ3, and t is the recovery interval.
Conductance values were calculated using the formula G = I/(V − Vr), where G is conductance, I is current amplitude, V is the depolarized membrane potential, and Vr is the reversal potential. Reversal potentials were individually estimated for each data set by fitting the I–V data with the equation I = [1 + exp(−0.03937 · z · (V − V1/2))]−1 · g · (V − Vr), where z is the apparent gating charge, g is a factor related to the number of channels contributing to the macroscopic current, V is equal to the voltage potential of the pulse, and V1/2 is the half-maximal voltage. Conductance values were fit with a two-state Boltzmann equation, G = 1/(1 + exp[−0.03937 · z · (V − V1/2)]), with z equal to the apparent gating charge, V equal to the pulse potential, and V1/2 equal to the voltage required for half-maximal activation. The voltage dependence of fast inactivation data was fit with a two-state Boltzmann equation, I = 1/(1 + exp[( V − V1/2)/a]), with I equal to the current amplitude measured during the test depolarization, V equal to the inactivating depolarization potential, a equal to the slope factor, and V1/2 equal to the voltage depolarization required for half-maximal inactivation. For analysis of current after PKA induction, there was drift in the peak current amplitude in some cases, even after allowing for recovery from slow inactivation. In those cases, the peak current measurements were adjusted by subtracting a linear relationship that was fit to data acquired during the first 10 min before PKA stimulation.
RESULTS
Construction of a full-length cDNA clone encoding the Rat1 sodium channel
The type I rat brain sodium channel sequence (Rat1) was amplified by RT-PCR from total rat brain RNA using primers based on the previously published sequence (Noda et al., 1986a). The continuous, full-length cDNA containing the coding region was sequenced to determine the predicted amino acid sequence. The sequence of this clone is similar to that of the previously published Rat1 sequence (Noda et al., 1986a). The two sequences differ at only 11 nucleotide positions. Most of these differences are silent, so that there are only four amino acid differences, which are represented by the circled residues in Figure 1. The amino acid indicated within each circle is the residue present in the clone that we have constructed, and the amino acid indicated in parentheses is present in the clone isolated by Noda et al. (1986a). Three of the differences are located in putative cytoplasmic regions of the channel (R97K, G427E, and V1823M), and two of these (R97K and V1823M) are conservative changes. The final difference (G979R) is a nonconservative change in the S6 transmembrane region of domain II.
Diagram of the Rat1 sodium channel. The predicted transmembrane topology of the voltage-gated sodium channel consists of four homologous domains (I–IV), each composed of eight transmembrane segments. The linkers connecting the four domains and the amino and C termini are on the inside of the membrane. The four amino acids that differ between the cDNA clone described in this paper (circled) and the cDNA clone originally isolated by Noda et al. (1986a) (parentheses) are indicated. The five consensus PKA sites in the I–II linker are depicted by solid circles.
Functional expression of Rat1 sodium channels and modulation by β1 and β2
RNA encoding the Rat1 sodium channel was transcribed in vitro and injected into Xenopus oocytes, which resulted in significant levels of sodium current. The Rat1 current was sensitive to tetrodotoxin, with a Kapp = 9.6 ± 3.2 nm, similar to the value obtained for Rat2 currents in this study (8.8 ± 4.0 nm). Sodium currents were recorded using a cut-open oocyte voltage clamp, and the properties of Rat1 were compared with those of Rat2 (Fig. 2). Both channel isoforms were tested by injection of RNA encoding the α subunit alone, α + β1 subunits, α + β2subunits, and α + β1 + β2 subunits. Sodium currents were elicited by depolarizations ranging from −65 to +25 mV in 10 mV increments from a holding potential of −100 mV. To obtain comparable levels of current for the two channels, a 500-fold greater amount of Rat1 RNA was injected. The properties of both Rat1 and Rat2 were modulated by β1, most notably resulting in an accelerated time course of fast inactivation (Fig. 2B). The β2 subunit resulted in a slight acceleration of inactivation that was more pronounced for Rat1 (Fig. 2C). The combination of β1 and β2 resulted in kinetics of inactivation that were similar to those observed when just β1 was added to either α subunit (Fig. 2D). The effects of β1 and β2 on Rat2 are consistent with previous studies (Isom et al., 1992, 1995; Patton et al., 1994).
Representative Rat1 and Rat2 sodium channel currents. Rat1 and Rat2 sodium currents are compared for (A) α subunits alone, (B) α + β1 subunits, (C) α + β2 subunits, and (D) α + β1 + β2 subunits. Rat1 and Rat2 sodium channels were expressed in Xenopus oocytes, and currents were recorded using a cut-open oocyte voltage clamp at 20°C as described in Materials and Methods. Currents were elicited by membrane depolarizations ranging from −65 to +25 mV in 10 mV increments from a holding potential of −100 mV. Calibration: 10 msec, 0.5 μA.
Rat1 sodium channels inactivate more slowly than Rat2 channels
In the absence of β1 and β2, the Rat1 currents demonstrated a slower time course of fast inactivation than did the Rat2 currents (Fig. 2A). To quantify the differences, the time constants for inactivation were determined by fitting the current traces with single and double exponential equations, as described in Materials and Methods (Fig. 3A). For depolarizations between −30 and 0 mV, current traces were fit with a single exponential equation, resulting in one time constant of inactivation (τslow, squares). For depolarizations between +10 mV and +50 mV, current traces were best fit with a double exponential equation, resulting in two time constants (τfast, circles, and τslow, squares). The time constants for both components of inactivation were significantly greater for Rat1 (filled symbols) than for Rat2 (open symbols). The fraction of current represented by τfast is indicated in the bottom panel of Figure 3. The fraction represented by τfast is 0 between −30 mV and 0 mV, because those traces were fit with a single exponential equation representing the slow time constant. Between +10 mV and +50 mV, the fast component of inactivation became increasingly prominent, as indicated by the fraction of current that was fit with τfast. At depolarizations equal to or greater than +20 mV, both channel isoforms demonstrated approximately equal fractions of the fast and slow components.
Time constants for fast inactivation of Rat1 and Rat2 sodium channels. Currents were recorded from oocytes expressing Rat1 or Rat2 sodium channels as described in the legend to Figure 2. The kinetics of inactivation were fit with single or double exponential equations as described in Materials and Methods, and the time constants representing the fast and slow components are shown on a logarithmic scale in the top panels for (A) α subunits alone, (B) α + β1 subunits, (C) α + β2 subunits, and (D) α + β1 + β2 subunits. The fast component (τfast) is represented by circles, and the slow component (τslow) is represented bysquares. Solid symbols indicate Rat1 (•, ▪), and open symbols indicate Rat2 (○, □). The fraction of current inactivating with τfast is shown in thebottom panels. Solid bars indicate Rat1, andopen bars indicate Rat2. In all cases, the fraction of τfast plus the fraction of τslow equals 1. Values represent averages, and error bars indicate SDs. Sample sizes were Rat1 α (7), Rat2 α (9), Rat1 α + β1 (5), Rat2 α + β1 (7), Rat1 α + β2 (6), Rat2 α + β2 (5), Rat1 α + β1 + β2(7), and Rat2 α + β1 + β2 (6).
When Rat1 and Rat2 were coexpressed with the β1subunit, sodium current traces were best fit with two exponential equations over the entire range of depolarizations (Fig. 3B). The principal effect of β1 for both channels was to accelerate the fast component of inactivation, as has been shown previously to be the case for Rat2 (Isom et al., 1992;Patton et al., 1994). The β1 subunit had a somewhat greater effect on Rat1 currents, so that τfast values for both channel isoforms were comparable in the presence of β1. In addition, β1 caused the fast component to predominate for both Rat1 and Rat2 throughout the entire range of depolarizations tested (Fig. 3B, bottom panel). The effect of β1 on τslow was to cause a considerable reduction for Rat1 and a slight reduction for Rat2, so that the time constants in the presence of β1 were comparable. Therefore, β1 caused both channel isoforms to be functionally equivalent with respect to τfast and τslow.
When Rat1 and Rat2 were coexpressed with the β2 subunit, the kinetics were best fit with a single exponential (τslow) between −30 and 0 mV, and two exponentials between +10 and +50 mV (Fig. 3C). These results were similar to those observed when the α subunits were expressed alone, with one notable difference. The kinetics of inactivation for the Rat1 channel were accelerated significantly, and there was only a minimal effect on the Rat2 channel. Therefore, the kinetics of inactivation were comparable for Rat1 and Rat2 in the presence of the β2 subunit.
Coexpression of β1 and β2 produced a modest increase in τslow compared with β1 alone over a range of depolarizations for Rat1 (Fig. 3D). In addition, a smaller fraction of Rat1 channels inactivated with the fast component at more negative depolarizations. Most notably, the fraction of τfast at −30 mV was only 25%, but that fraction increased to 80% by −10 mV (Fig. 3D, bottom panel). The combination of β1 and β2 resulted in inactivation kinetics of Rat2 that were comparable to those observed in the presence of β1alone.
Recovery from inactivation is faster for Rat1 compared with Rat2 sodium channels
Because Rat1 differed from Rat2 with respect to the kinetics of entry into the inactivated state, it was likely that the two channels would also differ with respect to the kinetics of recovery from inactivation. We therefore examined the kinetics of recovery from inactivation of Rat1 and Rat2 sodium channels. A two-pulse protocol was used to measure recovery over a time interval of 1–3000 msec, as described in the legend to Figure 4. The kinetics of recovery for Rat1 and Rat2 without β subunits are shown on a log scale in Figure 4A. Recovery is a multicomponent process for both channels, as can be seen by the multiple slopes of the curves. The kinetics were fit with a triple exponential equation as described in Materials and Methods, and the values for the time constants are shown in Table 1. The time constants for all three components of recovery are smaller for Rat1 than for Rat2, although this is somewhat offset by the fact that a greater percentage of Rat1 current recovered with the slowest time constant (τ3). In general, however, Rat1 recovered from inactivation more rapidly than Rat2.
Recovery from fast inactivation of Rat1 and Rat2 sodium currents. Recovery from inactivation was measured using a two-pulse protocol consisting of an initial conditioning pulse to −10 mV for 50 msec (which inactivated >95% of the channels), a variable recovery interval, and a test pulse to −10 mV to measure the amount of current that had recovered. Fractional recovery was calculated by dividing the current amplitude during the test pulse by the amplitude measured during the corresponding conditioning pulse. Fractional recovery is plotted on a log scale as a function of recovery time for (A) α subunits alone, (B) α + β1 subunits, (C) α + β2subunits, and (D) α + β1 + β2subunits. Data for Rat1 are indicated by solid circles, and data for Rat2 are indicated by open circles. Values represent averages, and error bars indicate SDs. The data were fit with a double or triple exponential equation as described in Materials and Methods, and the parameters of the fits are shown in Table 1. Sample sizes were Rat1 α (4), Rat2 α (5), Rat1 α + β1 (3), Rat2 α + β1 (4), Rat1 α + β2 (5), Rat2 α + β2 (3), Rat1 α + β1 + β2 (4), and Rat2 α + β1 + β2 (5).
Parameters of the voltage dependence of activation and inactivation and recovery from inactivation
When the β1 subunit was coexpressed with the α subunits, recovery from inactivation was faster for both Rat1 and Rat2, with Rat1 recovering more quickly than Rat2 at short recovery times (Fig. 4B). The faster recovery resulted from three effects of β1. First, >80% of the current recovered with the fast time constant. Second, both τ1 and τ2were decreased for both channels. Third, the very long time constant (τ3) was minimal or nonexistent. Coexpression of just β2 with the α subunits resulted in recovery kinetics that were similar to those observed for the α subunits alone, except that the fast time constant (τ1) was decreased for both Rat1 and Rat2 (Fig. 4C, Table 1). When both β1 and β2 were coexpressed with the α subunits, recovery was similar to that observed with just α and β1, in that most of the current recovered from inactivation with the fast time constant (Fig. 4D, Table1).
The voltage dependence of activation and inactivation is more positive for Rat1 than for Rat2 sodium channels
The effects of sodium channels on electrical excitability of a neuron depend on both the kinetics and voltage dependence of sodium channel activation and inactivation. We therefore compared the voltage dependence of activation and inactivation for the Rat1 and Rat2 channels to determine whether there were any differences in these properties (Fig. 5). The curves were fit with two-state Boltzmann equations as described in Materials and Methods, and the parameters of the fits are shown in Table 1. When the α subunits were expressed alone, there were no significant differences in the voltage dependence of conductance between Rat1 (Fig. 5A, filled circles) and Rat2 (Fig. 5A,open circles). When β1 was coexpressed with the α subunits, the conductance curves for the two channel isoforms were still similar, although the voltage for half-maximal activation (V1/2) for Rat2 was shifted slightly in the negative direction, although not to a statistically significant extent (Fig. 5B, Table 1). Coexpression of the β2 subunit with the α subunits did not significantly affect the voltage dependence of conductance compared with the α subunits alone (Fig. 5C, Table 1). When both β1 and β2 were coexpressed with the α subunits, the V1/2 for Rat1 was shifted in the positive direction, whereas the V1/2 for Rat2 was shifted in the negative direction, so that these two values (−15 and −22 mV) were significantly different from each other (Fig. 5D, Table 1). The slope factors (z) were not significantly different between Rat1 and Rat2, and only the Rat1 slope factor was significantly decreased by coexpression of β1(Table 1). Coexpression of β2 did not significantly change the slope factor of either Rat1 or Rat2. In summary, the voltage dependence of conductance was similar for Rat1 and Rat2 α subunits alone, but it was significantly more positive for Rat1 + β1 + β2 compared with Rat2 + β1 + β2.
Voltage dependence of activation and inactivation for Rat1 and Rat2 sodium channels. The voltage-dependence of activation (circles) and inactivation (squares) are shown for (A) α subunits alone, (B) α + β1 subunits, (C) α + β2subunits, and (D) α + β1 + β2subunits. Sodium currents were elicited by depolarizing pulses from a holding potential of −100 mV to potentials ranging from −90 to +30 mV in 5 mV increments. Conductance values were calculated by dividing the peak current amplitude by the driving force at each potential and normalizing to the maximum conductance, as described in Materials and Methods. Solid circles indicate Rat1, and open circles indicate Rat2. Values represent averages, and error bars indicate SDs. The data were fit with a two-state Boltzmann equation as described in Materials and Methods, and the parameters of the fits are shown in Table 1. Sample sizes were Rat1 α (8), Rat2 α (10), Rat1 α + β1 (5), Rat2 α + β1 (7), Rat1 α + β2 (6), Rat2 α + β2 (5), Rat1 α + β1 + β2 (6), and Rat2 α + β1 + β2 (5). The voltage dependence of inactivation was determined using a two-step protocol in which a conditioning pulse to potentials ranging from −90 mV to +5 mV was followed by a test pulse to −10 mV to measure the peak current amplitude. The peak current amplitude during the test pulse was normalized to the maximum current amplitude and is plotted as a function of the conditioning pulse potential. Solid squaresindicate Rat1, and open squares indicate Rat2. Values represent averages, and error bars indicate SDs. The data were fit with a two-state Boltzmann equation as described in Materials and Methods, and the parameters of the fits are shown in Table 1. Sample sizes were Rat1 α (6), Rat2 α (3), Rat1 α + β1 (6), Rat2 α + β1 (7), Rat1 α + β2 (6), Rat2 α + β2 (5), Rat1 α + β1 + β2(3), and Rat2 α + β1 + β2 (5).
With respect to the voltage dependence of inactivation, the curve for the Rat1 α subunit alone (Fig. 5A, filled squares) is shifted in the positive direction compared with that for the Rat2 α subunit alone (Fig. 5A, open squares). When β1 was coexpressed with the α subunits, the V1/2 values for both curves were shifted in the negative direction to a similar extent (Fig. 5B, Table 1). Coexpression of β2 with the α subunits did not significantly affect the V1/2 for either Rat1 or Rat2 compared with the α subunits alone (Fig. 5C, Table1). The combination of β1 and β2 shifted the V1/2 for Rat1 by −6 mV, and it shifted the V1/2 for Rat2 by −15 mV, so that the V1/2 for Rat2 + β1 + β2 was significantly more negative than that for Rat1 + β1 + β2 (Fig. 5D, Table 1). The slope factor (a) for the Rat1 α subunit was significantly smaller than that for the Rat2 α subunit alone. However, the addition of β1 or β1 + β2 decreased only the Rat2 slope factor, so that it was comparable to that of Rat1 with either β1 or β1 + β2. In summary, the voltage dependence of inactivation for Rat1 was significantly more positive than that for Rat2, and this effect was more pronounced when both the β1 and β2 subunits were present.
Rat1 sodium channel current is modulated by PKA phosphorylation
Electrical excitability of a neuron can be altered by modulating the activity of the sodium channels (Schiffmann et al., 1995). The Rat2 sodium channel is functionally modulated by PKA in Chinese hamster ovary cells (Li et al., 1992, 1993) and in Xenopus oocytes (Gershon et al., 1992; Smith and Goldin, 1996). Phosphorylation at PKA consensus sites in the cytoplasmic linker between domains I and II of the Rat2 channel reduces sodium current amplitude by 10–20% (Smith and Goldin, 1996, 1997). The I–II linker in Rat1 contains five consensus PKA sites at positions comparable to those in Rat2, although the amino acid sequences of the PKA sites in the two channels are not identical. It seemed likely that the Rat1 channel would also be modulated by PKA. To determine whether modulation occurred, sodium currents were measured during depolarizations to −10 mV before and 10 min after induction of PKA by perfusion with a mixture containing 25 μm forskolin, 10 μm cpt-cAMP, 10 μm db-cAMP, and 10 μl BMX. Application of the PKA-activating mixture reduced sodium current amplitude for Rat1 by 7 ± 4% (n = 8) and for Rat2 by 9 ± 5% (n = 9). Therefore, the PKA-activating mixture modulated the Rat1 sodium channel to cause a reduction in current amplitude comparable to that observed for the Rat2 channel.
DISCUSSION
We have constructed a full-length cDNA clone encoding the Rat1 sodium channel. Injection into Xenopus oocytes of RNA transcribed from this clone resulted in macroscopic currents that were sufficiently large for electrophysiological analysis, in contrast to the results with the original Rat1 clone (Noda et al., 1986b). The Rat1 clone that we constructed differs at four amino acid positions compared with the original Rat1 clone (Noda et al., 1986a). We do not know whether these differences represent actual polymorphisms in the rat population or are artifacts of cloning either this channel or the original Rat1 clone. Three of the differences are in cytoplasmic regions, and two of these are conservative changes. The notable exception is at position 979 in domain IIS6, where the Rat1 clone that we isolated contains a glycine and the clone isolated by Noda et al. (1986a) contains an arginine. All other voltage-gated sodium channels that have been sequenced thus far contain a glycine at the comparable position (Goldin, 1995). It is possible that this difference accounts for the fact that we observed significant sodium currents from Rat1 andNoda et al. (1986b) did not.
The level of current expressed from Rat1 in our studies was significantly less than that expressed from Rat2. A 500-fold greater amount of Rat1 RNA was injected to yield current amplitudes comparable to those observed for Rat2. This difference in current levels represents either a difference in the amount of protein (resulting from less efficient post-translational processing or insertion into the membrane or both), or it might represent a functional property of the Rat1 channel.
Rat1 sodium channels were modulated by the β1 subunit in a manner similar to that observed previously for the Rat2 channel (Isom et al., 1992). The primary effects were a faster time course of fast inactivation, accelerated recovery from fast inactivation, and a negative shift in the voltage dependence of inactivation. The effect of the β2 subunit on Rat1 was a slight acceleration of the kinetics of inactivation, similar to the effect of β2 on the Rat2 channel (Isom et al., 1995). The combination of β1 and β2 with either Rat1 or Rat2 resulted in properties similar to those observed when only β1 was coexpressed with the α subunit. It is likely that both Rat1 and Rat2 α subunits are associated with β1 and β2subunits in the CNS, so that the physiologically relevant properties are those determined in the presence of all three subunits.
The kinetics of fast inactivation for the Rat1 α subunit channels were slower than those for the Rat2 α subunit channels (Figs. 2A, 3A); however, coexpression of the β1subunit caused the two channels to have inactivation kinetics that were comparable (Figs. 2B, 3B), so that both channels probably have similar inactivation kinetics in vivo. In the presence of β1, Rat1 recovered from inactivation more quickly than Rat2, particularly at short recovery intervals of <10 msec (Fig. 4B). Therefore, on the basis of the kinetics of inactivation and recovery, Rat1 channels should be capable of transmitting higher frequencies of electrical impulses compared with the Rat2 channel.
The electrical excitability of sodium channels is affected by both the kinetics of inactivation and the voltage-dependent properties of the channels. When expressed in the absence of the β subunits, the Rat1 and Rat2 channels had similar voltage dependence of conductance curves (Fig. 5A); however, coexpression of both β1and β2 significantly shifted the V1/2, for Rat2 in the negative direction relative to the V1/2 for Rat1 (Fig. 5D, Table 1). Therefore, Rat1 channels with β1 and β2 would require a stronger depolarization for a comparable level of activation. With respect to the voltage dependence of inactivation, the V1/2 for Rat1 α subunit channels was more positive than that for Rat2 α subunit channels (Fig. 5A). This difference was magnified by the presence of β1 and β2 subunits, because their presence caused a larger negative shift in the V1/2 of Rat2 channels than for Rat1 channels (Fig. 5D, Table 1). The differences between Rat1 and Rat2 channels in the voltage dependence of activation and inactivation would have opposite physiological effects. The more positive V1/2 of activation for Rat1 would mean that Rat1 channels require a stronger depolarization to be activated. In contrast, the more positive V1/2of inactivation would make those channels less likely to be inactivated than Rat2 channels, resulting in more channels available to be activated. Because the difference in the V1/2 of inactivation is greater than that of activation, the net effect should be that Rat1 sodium channels are more available to transmit electrical impulses.
The excitability of sodium channels in neurons can be modulated by extrinsic factors such as phosphorylation. In particular, induction of PKA phosphorylation reduces current amplitudes through Rat2 sodium channels (Gershon et al., 1992; Li et al., 1992, 1993; Smith and Goldin, 1996), and this effect is observed in hippocampal neurons in which dopaminergic receptors are activated (Cantrell et al., 1997). In this study, the Rat1 sodium channel currents were similarly reduced by a PKA-activating mixture containing forskolin, cpt-cAMP, db-cAMP, and IBMX. We have demonstrated previously that this mixture reduces Rat2 sodium current amplitudes by induction of PKA phosphorylation of the channel, and that elimination of the consensus PKA sites in the I–II linker prevents that effect (Smith and Goldin, 1997). It is likely that the same mechanism is responsible for the reduction of Rat1 currents. Rat1 and Rat2 have the same number and positioning of the five consensus PKA sites in the cytoplasmic linker connecting domains I and II. Three of the five PKA sites have identical amino acid sequences (site 2: RRNS; site 3: RRDS; site 5: KRRSSS), and the other two sites differ by one amino acid [site 1: KRFSS (Rat2), KRYSS (Rat1); site 4: RRPS (Rat2), RRNS (Rat1)]. The second PKA site is of critical importance in PKA-mediated current reduction for Rat2 (Smith and Goldin, 1997) and is identical in both channel isoforms.
The electrical responses of neurons in the CNS are largely determined by the properties of the channels that are expressed in those cells. Sodium channels that inactivate rapidly cause the transient, inward currents representative of fast action potentials, whereas persistent or noninactivating sodium channels may be responsible for spontaneous action potentials and plateau potentials (Llinas, 1988; Taylor, 1993;Grill, 1996). These two distinct sodium channel conductances have been well characterized in cerebellar Purkinje cells (Llinas and Sugimori, 1980; Raman and Bean, 1997). Vega-Saenz de Miera et al. (1997)suggested that the inactivating and noninactivating sodium conductances result from expression of Rat1 and Rat6, respectively. Rat2 RNA was undetectable in Purkinje cells in that study (Vega-Saenz de Miera et al., 1997). The role of Rat6 has been examined by electrophysiological analysis of Purkinje cells from ataxic mice that either lack or have a mutant form of Scn8a, the murine ortholog of Rat6 (Burgess et al., 1995; Kohrman et al., 1995; Kohrman et al., 1996). These studies indicated that Scn8a channels are responsible for subthreshold currents, suggesting that the rapidly inactivating sodium current results from Rat1 channels (Raman et al., 1997). The electrophysiological properties that we have demonstrated for Rat1 sodium channels are consistent with this conclusion.
In summary, we have constructed a full-length cDNA clone for the Rat1 sodium channel and demonstrated that this channel is functional inXenopus oocytes. Some of the electrophysiological properties of Rat1 differ from those of Rat2, which would result in greater availability of Rat1 channels. These differences suggest that Rat1 channels are capable of more rapid firing of action potentials compared with Rat2 channels. In addition, the electrophysiological properties of the Rat1 channels are consistent with a role for these channels in mediating the rapidly inactivating transient current in cerebellar Purkinje cells. The availability of a functional clone for Rat1 will make it possible to evaluate the physiological role of this sodium channel isoform.
Footnotes
This research was supported by National Institutes of Health Grant NS 26729 to A.L.G. A.L.G. is an Established Investigator of the American Heart Association. We thank Drs. Linda Hall, Marianne Smith, Ted Shih, and Michael Pugsley for helpful discussions during the course of this work, and Dr. Linda Hall and Mike Wishingrad for the sequencing. We acknowledge Esther Yu and Mimi Reyes for excellent technical assistance.
Correspondence should be addressed to Dr. Alan L. Goldin, Department of Microbiology and Molecular Genetics, University of California, Irvine, CA 92697-4025.
