Portal de Periódicos da CAPES

Complex Subunit Assembly of Neuronal Voltage-gated K+Channels

1997; Elsevier BV; Volume: 272; Issue: 44 Linguagem: Inglês

10.1074/jbc.272.44.27577

1083-351X

Robert Koch, Siegmund G. Wanner, Alexandra Koschak, Markus Hanner, Christoph Schwarzer, Gregory J. Kaczorowski, Robert S. Slaughter, María L. García, Hans‐Günther Knaus,

Marine Toxins and Detection Methods

Neurons require specific patterns of K+ channel subunit expression as well as the precise coassembly of channel subunits into heterotetrameric structures for proper integration and transmission of electrical signals. In vivo subunit coassembly was investigated by studying the pharmacological profile, distribution, and subunit composition of voltage-gated Shaker family K+(Kv1) channels in rat cerebellum that are labeled by125I-margatoxin (125I-MgTX;K d, 0.08 pm). High-resolution receptor autoradiography showed spatial receptor expression mainly in basket cell terminals (52% of all cerebellar sites) and the molecular layer (39% of sites). Sequence-directed antibodies indicated overlapping expression of Kv1.1 and Kv1.2 in basket cell terminals, whereas the molecular layer expressed Kv1.1, Kv1.2, Kv1.3, and Kv1.6 proteins. Immunoprecipitation experiments revealed that all 125I-MgTX receptors contain at least one Kv1.2 subunit and that 83% of these receptors are heterotetramers of Kv1.1 and Kv1.2 subunits. Moreover, 33% of these Kv1.1/Kv1.2-containing receptors possess either an additional Kv1.3 or Kv1.6 subunit. Only a minority of the 125I-MgTX receptors (<20%) seem to be homotetrameric Kv1.2 channels. Heterologous coexpression of Kv1.1 and Kv1.2 subunits in COS-1 cells leads to the formation of a complex that combines the pharmacological profile of both parent subunits, reconstituting the native MgTX receptor phenotype. Subunit assembly provides the structural basis for toxin binding pharmacology and can lead to the association of as many as three distinct channel subunits to form functional K+channels in vivo. Neurons require specific patterns of K+ channel subunit expression as well as the precise coassembly of channel subunits into heterotetrameric structures for proper integration and transmission of electrical signals. In vivo subunit coassembly was investigated by studying the pharmacological profile, distribution, and subunit composition of voltage-gated Shaker family K+(Kv1) channels in rat cerebellum that are labeled by125I-margatoxin (125I-MgTX;K d, 0.08 pm). High-resolution receptor autoradiography showed spatial receptor expression mainly in basket cell terminals (52% of all cerebellar sites) and the molecular layer (39% of sites). Sequence-directed antibodies indicated overlapping expression of Kv1.1 and Kv1.2 in basket cell terminals, whereas the molecular layer expressed Kv1.1, Kv1.2, Kv1.3, and Kv1.6 proteins. Immunoprecipitation experiments revealed that all 125I-MgTX receptors contain at least one Kv1.2 subunit and that 83% of these receptors are heterotetramers of Kv1.1 and Kv1.2 subunits. Moreover, 33% of these Kv1.1/Kv1.2-containing receptors possess either an additional Kv1.3 or Kv1.6 subunit. Only a minority of the 125I-MgTX receptors ( 90% (data not shown). Anti-Kv1.2 antibody precipitated >95% of125I-MgTX receptors (n = 48), and the combination of any other anti-Kv1 antibody with anti-Kv1.2 did not increase the amount of immunoprecipitated material. These results indicate that in virtually all cerebellar MgTX receptors, Kv1.2 is an essential component. Using solely anti-Kv1.1, 83 ± 5% of receptors (n = 32) could be precipitated, whereas anti-Kv1.3 and anti-Kv1.6 recognized 21 ± 5 (n = 16) and 16 ± 4% (n = 11) of the sites, respectively. By using saturating concentrations of both anti-Kv1.3 and anti-Kv1.6, additive precipitation levels (33 ± 7%; n = 13) could be achieved, indicating that these two proteins are segregated into distinct MgTX-sensitive channel complexes (see "Discussion"). To determine whether Kv1.3 and/or Kv1.6 subunits are coassembled with Kv1.1, we investigated the extent of precipitation of 125I-MgTX receptors by either anti-Kv1.3 or anti-Kv1.6 in the presence of anti-Kv1.1. Neither antibody significantly increased the levels achieved by anti-Kv1.1 alone (83 ± 5%, see above). Thus, a saturating amount of anti-Kv1.1 together with anti-Kv1.3 precipitated 87 ± 4% (n = 20) of the receptors, whereas the anti-Kv1.1/anti-Kv1.6 combination yielded 86 ± 5% (n = 20) precipitation. In addition, a combination of anti-Kv1.1, anti-Kv1.3, and anti-Kv1.6 did not further increase the amount of precipitation. Taken together, these data indicate that Kv1.2 is the dominant subunit of all MgTX-sensitive K+ channels in rat cerebellum. In 80% of the receptors, Kv1.2 is assembled with Kv1.1, whereas in one-third of these receptors, either Kv1.3 or Kv1.6 seems to be an additional integral component of the complex. The remaining 20% of cerebellar MgTX receptors seem to be composed of homotetrameric Kv1.2 channels. To confirm the composition of the cerebellar MgTX receptor by independent means, we heterologously expressed Kv1.1, Kv1.2, and a combination of Kv1.1/Kv1.2 subunits transiently in COS-1 cells, and binding of 125I-MgTX or 125I-DTX was used to determine the pharmacological properties of the resulting complex. 125I-MgTX binds to Kv1.2 and Kv1.1/Kv1.2 membranes with aK d value of 0.08 pm (data not shown), a value identical to that determined with cerebellar membranes (see Fig.1). In contrast, no specific 125I-MgTX binding signal was observed for homotetrameric Kv1.1 channels. Due to this fact, 125I-DTX was used instead, because this ligand has equal affinity for both Kv1.1 and Kv1.2 channels (13Grissmer S. Nguyen A.N. Aiyar J. Hanson D.C. Mather R.J. Gutman G.A. Karmilowicz M.J. Auperin D.D. Chandy K.G. Mol. Pharmacol. 1994; 45: 1227-1234PubMed Google Scholar). 125I-DTX binds to Kv1.1, Kv1.2, and Kv1.1/Kv1.2 COS-1 membranes with a dissociation constant of 0.2 pm (data not shown). The pharmacology of 125I-DTX binding to Kv1.1, Kv1.2, Kv1.1/Kv1.2, and cerebellar membranes was investigated next (Fig. 5 C). Charybdotoxin inhibits125I-DTX binding to homotetrameric Kv1.1 channels with a K i value of ∼4000 pm, whereas it displays higher affinity for homotetrameric Kv1.2 channels (K i, ∼3 pm). However, in Kv1.1/Kv1.2 membranes, charybdotoxin displays a K i of 21 pm, in close agreement with the K i value determined with cerebellar membranes (18 pm). These data suggest that the observed pharmacological profile of Kv1.1/Kv1.2 membranes is due to the association of both subunits in a receptor complex and that this pharmacology is similar to that found with cerebellar membranes. To further validate the idea that Kv1.1 and Kv1.2 subunits are in fact associated in a complex after transient expression in COS-1 cells, immunoprecipitation experiments with anti-Kv1.1 and anti-Kv1.2 were performed (Fig. 5 B). For homomultimeric Kv1.2 channels,125I-MgTX or 125I-DTX receptors can be fully precipitated with anti-Kv1.2 but not anti-Kv1.1, whereas the opposite profile is observed when125I-DTX receptors from homomultimeric Kv1.1 channels are subjected to immunoprecipitation with these antibodies. However, both anti-Kv1.1 and anti-Kv1.2 are able to immunoprecipitate either 125I-MgTX or125I-DTX receptors from Kv1.1/Kv1.2 membranes (Fig. 5 B). In these experiments, anti-Kv1.2 precipitated close to 100% of125I-MgTX receptors, whereas anti-Kv1.1 precipitated ∼80%. These data imply that Kv1.1 and Kv1.2 subunits are indeed functionally associated in a receptor complex that very closely resembles the major constituent of the 125I-MgTX cerebellar receptor. The multiplicity of Kv channel genes in rat brain and their ability to form functional heterotetrameric channel complexes are two potential mechanisms for generating the enormous diversity of K+ channel currents observed in native neurons. Several lines of evidence suggest that Kv1 channel subunits form heterotetramers in vitro and in vivo. Coexpression of Kv1 subunits in Xenopus oocytes leads to the formation of heterotetrameric Kv channels with distinct properties when compared with those of the corresponding homomultimeric channels (9Wang H. Kunkel D.D. Martin T.M. Schwartzkroin P.A. Tempel B.L. Nature. 1993; 365: 75-79Crossref PubMed Scopus (519) Google Scholar, 10Sheng M. Liao Y.J. Jan Y.N. Jan L.Y. Nature. 1993; 365: 72-75Crossref PubMed Scopus (292) Google Scholar, 11Wang H. Kunkel D.D. Schwartzkroin P.A. Tempel B.L. J. Neurosci. 1994; 14: 4588-4599Crossref PubMed Google Scholar). Moreover, in situhybridization and immunocytochemical experiments have provided evidence concerning the existence of overlapping Kv channel expression, but the resolution of these techniques is too low to directly detect coassembly of different subunits into tetrameric structures. To address this issue, some studies have investigated the occurrence of heterotetrameric K+ channels in vivo by immunoprecipitation of Kv1 channel complexes, followed by cross-blotting with subtype-selective anti-Kv1 antibodies (9Wang H. Kunkel D.D. Martin T.M. Schwartzkroin P.A. Tempel B.L. Nature. 1993; 365: 75-79Crossref PubMed Scopus (519) Google Scholar, 10Sheng M. Liao Y.J. Jan Y.N. Jan L.Y. Nature. 1993; 365: 72-75Crossref PubMed Scopus (292) Google Scholar). Using this approach, heterotetrameric channel assembly of Kv1.2/Kv1.4 subunits in hippocampus (10Sheng M. Liao Y.J. Jan Y.N. Jan L.Y. Nature. 1993; 365: 72-75Crossref PubMed Scopus (292) Google Scholar) and Kv1.1/Kv1.2 subunits in basket cell terminals and juxtaparanodal regions (9Wang H. Kunkel D.D. Martin T.M. Schwartzkroin P.A. Tempel B.L. Nature. 1993; 365: 75-79Crossref PubMed Scopus (519) Google Scholar, 11Wang H. Kunkel D.D. Schwartzkroin P.A. Tempel B.L. J. Neurosci. 1994; 14: 4588-4599Crossref PubMed Google Scholar) has been demonstrated (28Sheng M. Tsaur M.L. Jan Y.N. Jan L.Y. Neuron. 1992; 9: 271-284Abstract Full Text PDF PubMed Scopus (400) Google Scholar). More recent studies have extended these findings and led to the identification of several subpopulations of defined Kv1 oligomers in bovine cerebral cortex (18Shamotienka O.G. Parcej D.N. Dolly J.O. Biochemistry. 1997; 36: 8195-8201Crossref PubMed Scopus (121) Google Scholar). An alternative way to address the issue of channel composition is to label native channels with a high-affinity ligand and establish their composition by quantitative immunoprecipitation. In the present study, we focused exclusively on the composition of MgTX-sensitive K+ channels in rat cerebellum. In cerebellar membranes 125I-MgTX binds to a single class of receptors with very high affinity (K d < 0.1 pm), slow ligand dissociation kinetics, and pharmacological characteristics that are not consistent with binding to a single homomultimeric Kv1 channel. Moreover, the pattern of Kv1 protein distribution in cerebellum is understood in some detail (Refs.27Veh R.W. Lichtinghagen R. Sewing S. Wunder F. Grumbach I.M. Pongs O. Eur. J. Neurosci. 1995; 7: 2189-2205Crossref PubMed Scopus (286) Google Scholar and 29McNamara N.M. Muniz Z.M. Wilkin G.P. Dolly J.O. Neuroscience. 1993; 57: 1039-1045Crossref PubMed Scopus (52) Google Scholar; this report). Several Kv1 subunits are expressed in this brain region at significant levels, such as Kv1.1, Kv1.2, Kv1.3, and Kv1.6. Kv1.2 and Kv1.3 are particularly interesting subunits, because these homotetrameric channels represent very high-affinity receptors for 125I-MgTX (30Helms L.M.H. Felix J.P. Bugianesi R.M. Garcia M.L. Stevens F. Leonard R.J. Knaus H.-G. Koch R. Wanner S.G. Kaczorowski G.J. Slaughter R.S. Biochemistry. 1997; 36: 3737-3744Crossref PubMed Scopus (50) Google Scholar). By a combination of autoradiography, immunocytochemistry, and immunoprecipitation studies, it has been possible to determine the composition of the 125I-MgTX receptor in cerebellum. These studies indicate that all receptors contain at least one Kv1.2 subunit and that ∼80% are heterotetramers of Kv1.1 and Kv1.2. In addition, some of these Kv1.1/Kv1.2 channels also contain an additional Kv1.3 or Kv1.6 subunit. It is most likely that all receptors found in basket cell terminals are Kv1.1/Kv1.2, because no other Kv1 subunit can be mapped to the compartment. However, at the molecular layer, homomultimeric Kv1.2, Kv1.1/Kv1.2, and Kv1.1/Kv1.2 with either Kv1.3 or Kv1.6 seem to be present. The unique pharmacological properties of the MgTX receptor should be mostly determined by the Kv1.1/Kv1.2 channels, because these are the major constituents in cerebellum. This idea has been confirmed in coexpression experiments in which Kv1.1 and Kv1.2 subunits were transiently expressed in COS-1 cells. Interestingly, these two subunits coassemble to yield a unique MgTX receptor phenotype that is very similar to that found in cerebellar membranes. Thus, heterotetrameric channel formation can occur in vitro and in vivo to the same extent and contributes to the diversity of K+ channels in the central nervous system. Those mechanisms controlling subunit assembly must play a very important role in determining the overall electrical activity in any given neuron. However, these processes remain to be characterized. We thank Maria Trieb, Emanuel Emberger, William Schmalhofer, and Jerry DiSalvo for technical contributions and recombinant toxin synthesis. Drs. Hartmut Glossmann, Jörg Striessnig, and Günther Sperk are gratefully acknowledged for continuous support and discussion.