Abstract
It has been hypothesized that R-type Ca currents result from the expression of the α1E gene. To test this hypothesis we examined the properties of voltage-dependent Ca channels in mice in which the α1E Ca channel subunit had been deleted. Application of ω-conotoxin GVIA, ω-agatoxin IVA, and nimodipine to cultured cerebellar granule neurons from wild-type mice inhibited components of the whole-cell Ba current, leaving a “residual” R current with an amplitude of ∼30% of the total Ba current. A minor portion of this R current was inhibited by the α1E-selective toxin SNX-482, indicating that it resulted from the expression of α1E. However, the majority of the R current was not inhibited by SNX-482. The SNX-482-sensitive portion of the granule cell R current was absent from α1E knock-out mice. We also identified a subpopulation of dorsal root ganglion (DRG) neurons from wild-type mice that expressed an SNX-482-sensitive component of the R current. However as with granule cells, most of the DRG R current was not blocked by SNX-482. We conclude that there exists a component of the R current that results from the expression of the α1E Ca channel subunit but that the majority of R currents must result from the expression of other Ca channel α subunits.
- dorsal root ganglia
- cerebellar granule cells
- pain
- synaptic transmission
- voltage-dependent calcium channels
- α1E knock-out mice
Voltage-sensitive Ca channels are of great importance in coupling excitability to Ca-dependent events within cells. It is clear that there are many different kinds of Ca channels. The properties of these channels make them suitable for performing different tasks in neurons and other cells (Catterall, 1998). Voltage-sensitive Ca channels are multisubunit structures consisting of a major pore-forming subunit (the α1 subunit) and several ancillary subunits (Hoffman et al., 1999). At this time 10 different α1 subunits are known to exist (Ertel et al., 2000). Seven of these (α1A–α1E, α1F, and α1S) code for high-threshold Ca channels (Catterall, 1998; Hoffman et al., 1999), whereas α1G–α1Icode for low-threshold Ca channels (Cribbs et al., 1998; Perez-Reyes et al., 1998; Lee et al., 1999). It has been important to try and establish which α1 subunits are responsible for the different types of Ca currents that can be recorded from particular types of cells. However, there are many splice variants of the various Ca channel subunits, and different combinations of these result in Ca currents with a wide range of biophysical properties when they are expressed in heterologous expression systems (Parent et al., 1997;Pereverzev et al., 1998; Hoffman et al., 1999). Thus, the potential diversity of Ca current types is very large. However, the situation has been facilitated by the availability of a number of drugs and toxins that selectively target different α1subunits. Thus, ω-conotoxin GVIA is diagnostic for α1B, dihydropyridines are diagnostic for α1C, α1D, and α1S, ω-agatoxin IVA is diagnostic for α1A, etc. (Catterall, 1998; Hoffman et al., 1999). One exception to this has been the Ca channel subunit encoded by the α1E gene. Until very recently no specific toxin or drug was known that targeted α1E Ca currents when these were expressed in heterologous expression systems (Newcombe et al., 1998). Consequently, the neuronal currents resulting from the expression of α1E were not known with certainty. There has been much speculation that neuronal Ca currents that are insensitive to other drugs and toxins represent α1E currents (Hilaire et al., 1997;Piedras-Renteria and Tsien, 1998; Tottene et al., 2000). Indeed, in certain instances these “residual” or “R-type” Ca currents do possess biophysical properties that resemble those resulting from the expression of α1E subunits (Randall and Tsien, 1995; Tottene et al., 1996; Hilaire et al., 1997). Nevertheless, there has been considerable controversy about the nature of α1E-based Ca currents in neurons, and very little is known about their functions (Bourinet et al., 1996; Meir and Dolphin, 1998). Recently, a toxin (SNX-482) has been described that does selectively block the Ca currents that result from the expression of the α1E subunit in heterologous expression systems (Newcombe et al., 1998). Interestingly, this toxin was unable to block neuronal R currents in several instances. On the other hand, experiments using antisense knock-out of α1Esubunits in neurons have supported the idea that R currents do indeed result from the expression of α1E subunits (Piedras-Renteria and Tsien, 1998; Tottene et al., 2000).
To answer questions further concerning the nature and functions of α1E Ca currents, we have generated mice in which the α1E gene has been deleted. We demonstrate that R currents are heterogeneous, consisting of elements that result from the expression of α1E as well as other non-α1E-dependent components.
MATERIALS AND METHODS
Construction of a Cacna1e null mutation. A bacterial artificial chromosome (BAC) containing the Cacna1egene was isolated by screening the Research Genetics CITB 129/Sv library with a Cacna1e cDNA. By the use of a probe specific for sequences encoding domain II of the Cacna1e protein, an ∼18 kb BamHI fragment was identified and cloned in the yeast shuttle vector pRS314. This vector had been modified previously to contain the thymidine kinase expression cassette. The DNA sequence of this fragment was determined and revealed to contain 11 exons that encode all of domain II.
The plasmid containing the 18 kb Cacna1e fragment was transformed into the yeast strain H1515 by LiCl. Yeast containing theCacna1e plasmid were transformed with a PCR fragment that had been generated using a template containing the Neo and URA3 expression cassettes. The PCR fragment was generated using chimeric primers that contain 50 bp of intron sequences of Cacna1eand 20 bp that correspond to the 5′ or 3′ ends of the selection cassette. The fragment produced encodes both selection cassettes and 50 bp arms, which are homologous to intron 4 and intron 8. After selection on growth plates lacking uracil, DNA was isolated from uracil prototrophs and transformed into bacterial strain DH10B. Plasmids were sequenced to confirm that the PCR fragment had integrated into the desired site.
The Cacna1e targeting vector was linearized usingNotI and electroporated into CJ7 ES cells. After selection with G418/FIAU, DNA was prepared from resistant clones and analyzed by genomic blot hybridization. An external 3′ probe was used to screen the NdeI-digested DNA for the presence of a rearranged 6.6 kb fragment and 6 kb endogenous fragment. Clones that exhibited the correct rearrangement were then analyzed using a 5′ internal probe to screen EcoRI-digested DNA for the presence of a 12 kb rearranged fragment and 8.7 kb endogenous fragment. Two clones were selected for blastocyst injection and produced several highly chimeric founder mice. Founders were then bred to C57BL/6J females.
Western blotting. Wild-type and α1E-deficient mice were decapitated, and the brains were removed on ice. The cerebellum was separated from the rest of the brain, and the brain regions were processed as follows. Tissue was homogenized for 20 strokes in an ice-cold buffer containing 0.32 m sucrose, 5 mm Tris, 5 μg/ml aprotinin, 5 μg/ml pepstatin, 5 μg/ml leupeptin, and 1 mm 4-(2 aminoethyl)-benzenesulfonyl-fluoride, pH 7.4. Homogenates were centrifuged at 1000 × g for 10 min. The resulting supernatant was diluted to 15 ml in sucrose buffer and centrifuged for 15 min at 20,000 × g. The supernatant was discarded, and the outer membrane portion of the pellet was gently washed and resuspended in 1 ml of sucrose buffer leaving the inner mitochondrial pellet intact. The suspended pellet was centrifuged in a bench top centrifuge at 14,000 rpm for 10 min at 4°C to remove the sucrose, and the resulting pellet was resuspended in 5 mm Tris and 2 mm EDTA, pH 7.4. Lysates were stored at −80°C until use.
Protein samples were loaded onto a 6% polyacrylamide gel and run at 120 V for ∼2 hr. Thirty micrograms of protein were loaded for brain homogenates, whereas 15 μg of α1E-transfected and untransfected human embryonic kidney (HEK) cell membranes was used. After transfer onto nitrocellulose, blots were blocked with 5% skim milk in Tris-buffered saline for 2 hr at room temperature. Blots were incubated with polyclonal α1E antibody at 1 μg/ml for 2 hr at room temperature.
Anti-human α1E antibody was generated by collaborators at Eli Lilly and Company. Amino acid sequences specific for α1E were prepared as glutathioneS-transferase (GST) fusion proteins in Escherichia coli. Antigens were derived from the IIS-6/IIIS-1 cytoplasmic loop region spanning amino acids 984–1099 of the human α1E subunit. The antibody was purified by protein A chromatography and affinity chromatography [for a detailed account of antibody preparation and purification, refer to Beattie et al. (1997)]. Antibody specificity was confirmed by ELISA and immunocytochemistry using human embryonic kidney cells transfected with appropriate α1 subunits (Volsen et al., 1995;Beattie et al., 1997). The specific use of the α1E antibody used in these studies has been reported previously (Volsen et al., 1995; Day et al., 1996; Beattie et al., 1997; Grabsch et al., 1999). After washing, blots were incubated with HRP-conjugated secondary antibody (1:20,000; Promega, Madison, WI) for 1 hr at room temperature. Bands were visualized by incubation in ECL chemiluminescent reagent and development on chemiluminescent film.
Dorsal root ganglion neuron culture. Dorsal root ganglia (DRGs) were rapidly removed from all spinal segments of 2- to 5-d-old neonatal knock-out (KO) and wild-type mice under aseptic conditions and digested sequentially in collagenase (Sigma, St. Louis, MO), collagenase and dispase (Boehringer Mannheim, Indianapolis, IN), and trypsin (Life Technologies, Gaithersburg, MD) in HBSS (Life Technologies) for 10 min, respectively, at 37°C. Ganglia were washed in DMEM three times and resuspended in F12 media (Life Technologies), supplemented with 10% FBS (Summit Biotechnology), N2 supplement (Life Technologies), 50 ng/ml nerve growth factor (Collaborative Biomedical Products), and penicillin and streptomycin (100 μg/ml and 100 U/ml, respectively). Single neuronal cells were obtained by trituration through a flame-polished pipette ∼10 times, and the cell suspension was centrifuged at 800 rpm for 5 min. The pellet was resuspended in F12 with the additives listed above, except that serum was reduced to 0.5%. Isolated DRG neurons were plated on polyornithine (Sigma)- and laminin (Collaborative Biomedical Products)-coated glass coverslips (25 mm in diameter) and incubated in the same medium that was replaced every 48 hr. Cultures were maintained at 37°C in a water-saturated atmosphere with 5% CO2 for up to 2 weeks.
Preparation of cultured cerebellar neurons. Cultured cerebellar neurons were prepared in a coculture system as described previously (Brorson et al., 1991). Briefly, 17- to 19-d-old embryos were removed from pregnant mice anesthetized previously with ether and killed by cervical dislocation. The cerebella were dissected out and incubated in trypsin (Worthington, Freehold, NJ; 100 mg/10 ml; 0.5 ml of trypsin in 4.5 ml of HBSS containing 25 mm HEPES) for 20 min at 37°C. After washing, tissue was resuspended in DNase and triturated through a flame-narrowed pipette until no visible tissue fragments remained. Cells were plated at a density of 3 × 105 cell/ml onto 15 mm round glass poly-l-lysine-coated coverslips. Neurons were suspended over a feeder layer of cortical astrocytes and maintained in serum-free defined medium containing 15 mm HEPES. After 2 d in culture at 37°C, in a humidified 5% CO2atmosphere, 5-fluoro-2-deoxyuridine (5 μm final concentration) was added to maintain astrocyte numbers at near confluency. The cultures were fed every 3–4 d by replacement of half of the medium. This preparation generated both cerebellar Purkinje cells and granule cells, which could be distinguished by their differences in size. Granule cells were used for electrophysiological analysis between 10 and 20 d in vitro.
Whole-cell patch clamp. The tight-seal whole-cell configuration of the patch-clamp technique (Hamill et al., 1981) was used to record Ba currents. Recordings were made at room temperature (21–24°C). Currents were recorded using the Axopatch 1D (Axon Instruments, Foster City, CA) or EPC-7 (List Electronics, Darmstadt, Germany) amplifier, filtered at 2 kHz by the built-in filter of the amplifier, and stored on the computer. Capacitative transients were cancelled at 10 MHz, and their values were obtained directly, together with the series resistance values from the settings of the Axopatch 1D or EPC-7 amplifiers. Leak corrections were performed using a P/N protocol. Command pulses were delivered at 30 sec intervals from −80 mV to +10 mV to elicit Ba current. Soft, soda-lime capillary glass was used to make patch pipettes. Before seal formation, the resistances of the recording electrodes were 4.5–6 ΩM for granule cells and 2.5–4 ΩM for DRG neurons. The extracellular buffer solution for whole-cell voltage-clamp experiments was composed of (in mm): 151 tetraethylammonium chloride, 5 BaCl2, 1 MgCl2, 10 HEPES, and 10 glucose; pH was adjusted to 7.4 with TEAOH. The standard internal solution consisted of (in mm): 100 CsCl, 37 CsOH, 1 MgCl2, 10 BAPTA, 10 HEPES, 3.6 MgATP, 1 GTP, 14 Tris2CP, and 50 U/ml creatine phosphokinase. The pH was adjusted to 7.3 with CsOH. The osmolarity of the pipette solution was 300 mOsm/1, and the osmolarity of the extracellular solution was between 325 and 340 mOsm/1. Toxins were applied by bath perfusion (cerebellar granule cells) or locally by a puff pipette (DRG neurons), always in the presence of 0.2 mg/ml chicken egg ovalbumin (Sigma), to block the low-affinity binding sites.
RESULTS
Disruption of the Cacna1e gene does not affect viability
A null allele of the Cacna1e gene was created using an 18 kb BamHI fragment that encodes all of domain II of the α1E protein. By the use of the homologous recombination system of yeast, the S4–S6 regions of domain II were replaced with a neomycin/URA3 selection cassette to generate a targeting vector (Fig. 1). We believed that by removing the pore-lining domain and its neighboring transmembrane domains that a null allele of Cacna1e would be created. This construct was introduced into embryonic stem cells, and the cells were screened for disruption of the Cacna1e locus. Approximately 110 neomycin-resistant clones were recovered and analyzed by Southern analysis. Of these 110 ES cells, 5 showed the desired disruption at the Cacna1e locus. Chimeric animals produced from blastocyst injection were then tested for the presence of mutant alleles. Two lines were established from independent ES cell clones, and intercross mating was used to generate homozygous mutant mice. Viable homozygous mutant animals were generated from both lines. Both of the lines generated gave similar results. There were no observed phenotypic differences between wild-type and homozygous knock-out mice.
Generation of a mouse with a disrupted allele of the Cacna1e gene. A, Targeting of theCacna1e gene is shown. Top, The genomic structure of the wild-type allele of the Cacna1e gene is shown with exons indicated by black boxes. Restriction sites used for determination of the disrupted allele are indicated.Bottom, A PGK neolURA3 expression cassette replaced the fourth through eighth exons of this fragment. The locations of probes used for genomic blot hybridization are shownbelow the disrupted allele. B, Genomic blot hybridization of DNA derived from neomycin-resistant ES cell clones is shown. ES cell DNA was probed with a 3′-flanking probe (left) and 5′ internal probe (right). Hybridization with the 3′ probe shows a 6 kb wild-type band and 6.6 kb mutant band, whereas the 5′ internal probe revealed an 8.7 kb wild-type band and 12 kb mutant band. C, Tail tip DNA from three intercross littermates was probed with the 3′-flanking probe and showed the expected size for wild-type, heterozygous, and homozygous mutant.ko, Knock-out.
Western blotting of α1E-deficient and wild-type mice
Western blot analysis was used to confirm deletion of the α1E protein in α1E-deficient mice compared with wild-type mice (Fig. 2). Cerebellum and whole brain minus cerebellum extracts from both 129/SV and C57BL/6J wild-type mice displayed a clear band at 230 kDa, the predicted size for the α1E subunit (Fig. 2). In contrast a corresponding band was not observed in either the cerebellum or whole brain minus cerebellum from α1E-deficient mice, indicating the absence of the α1E subunit from these mice (Fig. 2). Northern blot analysis revealed no detectable α1E transcript I in the homozygous knock-out mice using a probe located either within the deleted region or downstream from the deletion (+5387–5966).
Western blot analysis of α1E protein in wild-type and α1E-deficient mice. Brain lysates from 129/SV and C57BL/6J wild-type mice show the expression of the 230 kDa α1E protein, whereas the protein is clearly absent from brain tissue of α1E KO mice. Specificity of the antibody was confirmed by detection of a band in α1E-transfected HEK cell membranes and the lack of staining in untransfected HEK cell membranes.
Electrophysiology of Ba currents
We examined the properties of the voltage-sensitive Ba currents in cerebellar granule and DRG neurons cultured from α1E KO mice as well as their parental wild-type strains.
Ba currents in cerebellar granule and DRG neurons exhibited differential sensitivity to a number of toxins and drugs that have been shown previously to target different types of voltage-sensitive Ca channels (Figs. 3,4). In the case of cerebellar granule neurons (Fig. 3), components of the Ba current were blocked by the sequential addition of nimodipine (L-type blocker) ω-agatoxin VIA (P/Q-type blocker), and ω-conotoxin GVIA (N-type blocker), leaving a component of the current that we designated the R current on the basis of previous criteria in the literature (Fig. 3). We examined the effect of the α1E-selective toxin SNX-482 on the R current in these neurons. In both of the parental lines SNX-482 at a supramaximally effective concentration of 1 μm blocked ∼30% of the granule cell R current. However, it was clear that the majority of the R current was not blocked by this toxin. Although the overall magnitude of the whole-cell Ca current in α1E KO mice was the same as that in wild-type mice, the SNX-482-sensitive component of the Ba current was completely absent from α1E mice. However, a large R current still remained in these KO mice after sequential application of all of the drugs and toxins. We also noted that the size of the N-current was slightly larger in the granule neurons from α1E KO mice. The absolute magnitude of the currents recorded in these experiments can be seen in Table1.
Effects of drugs and toxins on the cerebellar granule cell IBa. A, Plot of the peak IBa versus time in an α1E KO mouse shows a resistant component of the high voltage-activated IBa remaining after consecutive bath application of specific drugs and toxins. SNX-482 had no effect on the toxin-resistant current. Inset,Representative traces at numbered points are shown. B, Plot of the peakIBa versus time from a wild-type C57BL/6J mouse is shown. Inset, Representativetraces are shown. SNX-482 inhibited a component of the R current remaining after nimodipine, ω-agatoxin IVA (ω-Aga IVA), and ω-conotoxin GVIA (ω-CTx GVIA) application. C, Plot of the peakIBa versus time from a wild-type 129/SV mouse is shown. SNX-482 also inhibited part of the R current in this case. Inset, Representative traces are shown. D, Comparison of inhibitory effects of specific drugs and toxins on the granule cell IBa in α1E KO and control mice is shown. Drugs and toxins were used in the following concentrations: nimodipine (Nimo), 500 nm; ω-conotoxin GVIA (CTx), 500 nm; ω-Aga IVA (Aga), 200 nm; SNX-482 (SNX), 1 μm; Cd2+, 100 μm. The data from the two parental control mouse strains were pooled. SNX-482 was only effective in inhibiting the IBa from control mice (**p < 0.01). The numbers inparentheses represent the number of experiments.E, The effects of the specific drugs and toxins on theIBa from the two control mouse strains did not differ significantly from each other. The numbers inparentheses represent the number of experiments.Resist, Resistant; wt, wild type.
Effects of drugs and toxins on theIBa in DRG neurons from wild-type and α1E KO mice. A, Plot of the peakIBa versus time in a DRG neuron from an α1E KO mouse. An R current still remained after sequential application of the different drugs and toxins. TheIBa from α1E KO mice was insensitive to SNX-482. Inset, Representative currents.B, Time course of the peakIBa from a wild-type DRG neuron exhibiting sensitivity to SNX-482. C, Time course of the peakIBa from a wild-type DRG neuron that was insensitive to SNX-482. D, Average inhibition (mean ± SEM) of the peak IBa by specific drugs and toxins. Drugs and toxins were used in the following concentrations: nimodipine, 2 μm; ω-conotoxin GVIA, 1 μm; ω-Aga IVA, 200 nm; SNX-482, 1 μm. DRG neurons from α1E KO mice (n = 7) and a population of DRG neurons from wild-type mice (n= 8) were insensitive to SNX-482 application. TheIBa inhibition by SNX-482 in sensitive DRG neurons (n = 10) was significant (p < 0.01). E, Interrelationship between sensitivity to ω-Aga IVA and SNX-482 in wild-type DRG neurons (n = 8). The correlation was significant (p = 0.0265) using the Spearman rank correlation test.
Ba2+ currents in cultured mouse neurons
In DRG neurons sequential addition of drugs and toxins also inhibited different components of the Ba current, leaving a substantial R current (Fig. 4). Interestingly, we observed two populations of DRG neurons from wild-type mice. In the first type, addition of SNX-482 was ineffective in blocking the R current. In a second population of neurons, addition of SNX-482 produced a clear inhibition of the R current although, as in the case of cerebellar granule neurons, the majority of the R current was not blocked by SNX-482. Both types of DRG neurons (i.e., SNX-482 sensitive and insensitive) were observed in both wild-type parental strains. Interestingly, in neurons that exhibited SNX-482-sensitive R currents, we found that ω-agatoxin IVA was relatively ineffective (Fig. 4E). DRG neurons from α1E KO mice exhibited Ba currents with the same magnitude as those from wild-type mice. The Ba currents in these neurons exhibited components that were blocked by nimodipine, ω-agatoxin IVA, and ω-conotoxin GVIA, leaving a substantial R current, but this was not blocked by SNX-482.
DISCUSSION
Voltage-sensitive Ca channels are a diverse family of molecules (Ertel et al., 2000). Of the nine genes whose expression produces channels of this type, the properties and functions of the α1E channel are the least well understood. Although α1E subunits appear to be widely distributed throughout the CNS and peripheral nervous system (Volsen et al., 1995; Day et al., 1996), it is not clear which types of Ca currents result from their expression. One of the problems has been the lack of a drug or toxin that specifically targets these channels, although the recent identification of the α1E-selective toxin SNX-482 has helped to alleviate this problem (Newcombe et al., 1998). Furthermore, because α1E subunits are structurally diverse (Pereverzev et al., 1998; Vajna et al., 1998; Spaetgens and Zamponi, 1999) and because their combination with different ancillary subunits results in a variety of Ca currents with differing biophysical characteristics (Parent et al., 1997; Nakashima et al., 1998), there is still considerable uncertainty about the types of Ca currents α1E subunits produce in neurons and other cells.
To overcome this problem some investigators have resorted to antisense approaches to examine the relationship between Ca currents and the expression of α1E subunits. For examplePiedras-Renteria and Tsien (1998) observed that antisense oligonucleotides for α1E partially reduced the size of the R current in rat cerebellar granule cells. They suggested either that more than one α1 subunit was responsible for the R current in these cells or that the apparent heterogeneity was caused by the expression of multiple forms of the α1E subunit that were differentially sensitive to antisense treatment. Furthermore, Newcombe et al. (1998) reported that the R current in cerebellar granule neurons was insensitive to SNX-482, suggesting a dissociation between this current and α1E expression. On the other hand, Tottene et al. (2000) reported that the R current in cerebellar granule neurons consisted of multiple components, some of which were sensitive to SNX-482 and some of which were resistant. However, these authors also demonstrated that antisense treatment was nearly completely effective in reducing both the toxin-sensitive and -insensitive components of the current. Thus, it is still unclear whether the R current, at least in cerebellar granule neurons, is a homogeneous entity or not and what its relationship to α1E expression might be.
To settle this issue, we have generated mice that completely lack the α1E transcript and protein. It is therefore clear that any Ca currents recorded from these animals cannot result from the expression of α1E. It is important to note therefore that there was a substantial R current in all of the neurons from α1E KO mice from which we made recordings. Therefore, we can conclude that the majority of R currents, at least in cerebellar granule and DRG neurons, do not result from the expression of α1E subunits. Having said this, it is also true that there appears to be a component of the R current in some cell types that does result from the expression of α1E, although this component is relatively small. The situation in mouse cerebellar granule cells therefore seems to be similar to that reported by Piedras-Renteria and Tsien (1998). Although the magnitude of α1E-dependent R current in their studies was larger than that in ours, their conclusion that the R current in granule neurons did not completely result from the expression of α1E seems similar to ours. Furthermore, our results are also consistent with the work of Newcombe et al. (1998), who observed SNX-482-resistant R currents in several types of neurons.
The identity of R currents in sensory neurons also seems to be complex. It is interesting to note that two different patterns of Ca current sensitivity were observed in these neurons. We only identified an SNX-482-sensitive component in those wild-type neurons with a relatively small ω-agatoxin IVA-sensitive current. We do not yet know what this phenotype means in terms of the overall classification of DRG neurons. However, as in the case of granule neurons, the magnitude of the SNX-482-sensitive current, in those DRG cells in which it is observed, was relatively small.
If the expression of α1E is not responsible for the majority of the R current in most neurons, then what is? It is interesting to note that Jun et al. (2000) have shown recently that there was a large decline in the magnitude of the cerebellar granule neuron R current in α1A KO mice [although, interestingly, Piedras-Renteria and Tsien (1998) found that α1A antisense did not reduce the R current in these neurons]; the decline in R current magnitude in these mice was much greater than that in the α1E KO mice reported here. Jun et al. (2000) demonstrated that ∼80% of the R current was absent in their α1A KO mice. Thus, it is conceivable that this portion, together with the SNX-482-sensitive portion described here, accounts for virtually all of the granule neuron R current, although exactly what the relationship is between the non-α1E-dependent R current and α1A is not clear at this time. Nevertheless, it is also interesting to note that in our studies on DRG neurons there was a negative correlation between the magnitude of the ω-agatoxin IVA-sensitive and the SNX-482-sensitive currents. It is also of interest that the SNX-482-sensitive current was restricted to a particular population of DRG neurons. Exactly what this population represents from the functional point of view awaits further identification but could indicate a role for α1E-based currents in the processing of a particular type of sensory information, such as pain (Saegusa et al., 2000).
Footnotes
This study was supported by Public Health Service Grants DA-02121, MH-40165, NS-33826, DK-44840, and NS-21442 to R.J.M. and grants from the National Cancer Institute Department of Health and Human Services to N.A.J. and N.G.C. We are grateful to Dr. G. Miljanich of Elan Pharmaceuticals for the gift of SNX-482.
Correspondence should be addressed to Dr. Richard J. Miller, Department of Neurobiology, Pharmacology, and Physiology, The University of Chicago, 947 East 58th Street (MC 0926), Chicago, IL 60637. E-mail:rjmx{at}midway.uchicago.edu.
