HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 276, Issue 52, 49427-49434, December 28, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/52/49427    most recent M109325200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Manganas, L. N. Articles by Trimmer, J. S. Search for Related Content PubMed PubMed Citation Articles by Manganas, L. N. Articles by Trimmer, J. S. Social Bookmarking                      What's this?

Episodic Ataxia Type-1 Mutations in the Kv1.1 Potassium Channel Display Distinct Folding and Intracellular Trafficking Properties^*

Louis N. Manganas, Sobia Akhtar^§, Dana E. Antonucci, Claire R. Campomanes, J. Oliver Dolly^§, and James S. Trimmer^¶

From the  Department of Biochemistry and Cell Biology, State University of New York, Stony Brook, New York 11794 and the ^§ Department of Biochemistry, Imperial College of Science, Technology, and Medicine, London SW7 2AY, United Kingdom

Received for publication, September 26, 2001, and in revised form, October 25, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Episodic ataxia type 1 (EA-1) is a neurological disorder arising from mutations in the Kv1.1 potassium channel -subunit. EA-1 patients exhibit substantial phenotypic variability resulting from at least 14 distinct EA-1 point mutations. We found that EA-1 missense mutations generate mutant Kv1.1 subunits with folding and intracellular trafficking properties indistinguishable from wild-type Kv1.1. However, the single identified EA-1 nonsense mutation exhibits intracellular aggregation and detergent insolubility. This phenotype can be transferred to co-assembled Kv1 - and Kv-subunits associated with Kv1.1 in neurons. These results suggest that as in many neurodegenerative disorders, intracellular aggregation of misfolded Kv1.1-containing channels may contribute to the pathophysiology of EA-1.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

A number of neurological channelopathies are derived from defective voltage-gated (Kv) potassium channels (1). Episodic ataxia type-1 (EA-1)¹ is a neurological syndrome that arises during infancy and persists throughout life. This disorder is characterized by continuous myokymia and brief and sometimes frequent attacks of cerebellar ataxia, resulting in imbalanced movements. Genetic linkage studies in EA-1 patients have identified at least 14 mutations in the mammalian KCNA1 gene, which encodes the Kv1.1 -subunit, as the underlying cause of EA-1 (2-4). Most of the KCNA1 mutations in EA-1 patients lead to Kv1.1 channels with altered biophysical characteristics, although certain mutations eliminate any detectable channel activity (5-13).

Mammalian Shaker or Kv1 potassium channels are membrane protein complexes composed of four transmembrane -subunits and up to four cytoplasmic Kv-subunits (14). All of the detectable Kv1.1-containing channel complexes in human and rodent brain are hetero-oligomeric complexes in which Kv1.1 is co-assembled with other Kv1 -subunits (e.g. Kv1.2 and Kv1.4) (15-19). These Kv1.1-containing channels are localized to axons and nerve terminals where they are key mediators of neurotransmitter release (16, 20-23). Pharmacological blockade of these channels with neurotoxins such as -dendrotoxin (-DTX) leads to hyperexcitability and enhanced transmitter release, which is the physiological basis for the epileptogenic activity of DTX (24, 25). Genetic ablation of murine Kv1.1 expression results in both ataxia and epilepsy, due to defects in both cerebellar and hippocampal excitability, respectively (26). Clearly, dysfunction of Kv1 channels containing Kv1.1 underlies the clinical manifestations observed in EA-1 patients.

All EA-1 missense point mutations identified to date are located in the region of the KCNA1 gene encoding the highly conserved core domain between transmembrane segments S1 and S6 (Fig. 1A). These Kv1.1 subunits can co-assemble with other Kv1 subunits forming heteromeric channels with altered gating properties and/or reduced currents (5, 6, 8-10, 13). Recently (13), a unique nonsense mutation in KCNA1 was identified in EA-1 patients which yields a Kv1.1 subunit with a truncated cytoplasmic C terminus (Fig. 1A). Patients with this mutation have a severe and remarkably drug-resistant form of EA-1, and this truncated Kv1.1 subunit does not form functional channels and acts as the most potent dominant-negative EA-1 mutant when co-expressed with wild-type Kv1.1 subunits (13).

We provide evidence that the nonsense mutation described in the patient with this novel drug-resistant form of EA-1 generates a Kv1.1 subunit with biochemical and cell biological properties distinct from wild-type Kv1.1 and from all other known EA-1 mutations. We have characterized the folding and intracellular trafficking of each of the 14 known EA-1 mutations. This was achieved by introducing each EA-1 mutation into the Kv1.1 cDNA and expressing the recombinant subunits in mammalian cell lines and cultured hippocampal neurons. We found that Kv1.1 with the EA-1 nonsense mutation exhibits a number of characteristic traits of misfolded membrane proteins and can confer this misfolded phenotype, with resultant intracellular retention, to co-assembled wild-type Kv1 - and Kv-subunits. The properties of this novel EA-1 mutant Kv1.1 polypeptide in many aspects resemble those of the mutant CFTR protein (27) encoded by the predominant mutation in cystic fibrosis patients, suggesting that diverse channelopathies may be based on similar defects in protein folding and intracellular trafficking. Moreover, these studies reveal that certain forms of EA-1 share phenotypic similarities with the large number of neurodegenerative disorders arising from protein misfolding (28).

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- All materials not specifically identified were purchased from Sigma or Roche Molecular Biochemicals. Sequencing was performed with Sequenase version 2.0 (United States Biochemical Corp., Cleveland, OH). Plasmid DNA used in all experiments was purified using the Qiagen purification kit (Chatsworth, CA). All enzymes used for molecular biology were from either Roche Molecular Biochemicals or New England Biolabs (Beverley, MA), with the exception of Pfu Turbo DNA polymerase from Stratagene (La Jolla, CA), which was used for all PCRs. All synthetic primers used for sequencing and PCRs were synthesized on a Beckman Oligo 1000 DNA synthesizer. PCRs were cycled on a GeneAmp PCR System 2400 from PerkinElmer Life Sciences.

All antibodies against potassium channel - and -subunit polypeptides and PSD-95 have been described previously (29-34). Anti-CD4 monoclonal antibody (18-46) was purchased form Santa Cruz Biotechnology. Anti-calnexin (8211-1) rabbit polyclonal anti-peptide antibody was generated against amino acids 578-591 of human calnexin. Immunoreactivity to calnexin was confirmed by immunoblot and peptide competition assay. Anti-vimentin (monoclonal, clone 9) was purchased from Sigma. Anti-ubiquitin (N-19) was purchased from Santa Cruz Biotechnology.

Analyses of Transfected Cells-- Transient transfection, immunofluorescence staining, and SDS-PAGE and immunoblotting analyses of transfected COS-1 cells were performed as described (32, 33, 35). For folding analyses, cells were incubated at 25 °C for 24 h after changing the medium or treated with 10% glycerol at 37 or 25 °C for 48 h after changing the medium (36). Culture and transfection of low density hippocampal neuronal cultures was as described (33, 37).

Generation of Kv1 Point Mutants and Chimeras-- Point mutants were generated using the Quick Change site-directed mutagenesis kit (Stratagene, La Jolla, CA), using two complementary oligonucleotides containing the desired mutation. The Kv1.1N (residues 1-389):1.4C (residues 543-654) Kv1 chimera was generated by making an (A-C) substitution at position 1414 of the Kv1.1 coding sequence and a (G-A) substitution at nucleotide 2129 of the Kv1.4 coding sequence. This substitution creates a new BamHI site that digests at nucleotide 1412 of Kv1.1 and 2127 of Kv1.4. Kv1.1(BamHI 1414) was digested with PstI/BamHI producing a Kv1.1N fragment corresponding to amino acids 1-388. Kv1.4(BamHI 2127) was digested with BamHI/EcoRI producing a Kv1.4C fragment corresponding to amino acids 543-654. A triple ligation was performed with these two fragments and pRBG4 digested with PstI/EcoRI giving rise to the Kv1.1N (residues 1-389):1.4C (residues 543-654) chimera.

Sucrose Gradient Sedimentation-- 0.5 mg of each protein standard (myosin, -galactosidase, phosphorylase, bovine serum albumin, ovalbumin, and carbonic anhydrase) and 50 µl of Kv1.1 COS-1 lysate were layered on top of separate 5-50% sucrose gradients (~2 ml) containing Tris-buffered saline, pH 8.0, 5 mM EDTA, 1% Triton X-100, 1 mM iodoacetamide, and protease inhibitor mixture. Samples were centrifuged for 4 h at 202,059 × g at 4 °C, and 10× 200-µl fractions were manually collected from the top of the gradient. Each 200-µl fraction was added to 800 µl of lysis buffer and immunoprecipitated with 1 µg/ml affinity-purified Kv1.1C antibody for 1 h at 4 °C. Protein A-Sepharose (30 µl) was used to precipitate antibody complexes for 30 min at 4 °C. Samples were washed three times in lysis buffer, and the final pellet was resuspended in sample buffer and analyzed by SDS-PAGE and immunoblot as described above.

Dendrotoxin Binding Assay-- Wild-type Kv1.1 and Kv1.1C79 were subcloned into the pSFV1 vector. cRNA was generated using Cap-Scribe system and electroporated into BHK cells. After 48 h of incubation at 37 °C and 5% CO₂, cells were harvested and used for toxin binding studies as intact cells or membrane fragments (38).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Analysis of EA-1 Mutations-- The positions of all known KCNA1 mutations that result in EA-1 are shown schematically in Fig. 1A. Each of the missense mutations occurs in the core domain consisting of the six transmembrane segments S1-S6 and the connecting extracellular and cytoplasmic segments. The bulk of the mutations are present in cytoplasmic domains connecting transmembrane segments S2-S3 and S4-S5. R417STOP is the single nonsense mutation that generates a truncated Kv1.1 -subunit lacking the final 79 amino acids (Kv1.1C79) of the cytoplasmic C-terminal tail (13).

View larger version (50K): [in this window] [in a new window]   Fig. 1.   Biochemical and cell biological analyses of all known Kv1.1 mutations in EA-1 patients. A, schematic depicting regions of the Kv1.1 -subunit harboring identified EA-1 mutations. B, COS-1 cells expressing wild-type or EA-1 mutant Kv1.1 subunits were harvested and permeabilized in lysis buffer containing 0.1% Triton X-100. Soluble (S) and insoluble (I) fractions were separated by centrifugation and analyzed by SDS-PAGE and immunoblot. AI, aggregation index, which represents the percentage of transfected cells exhibiting aggregated Kv1.1 staining. C and D, immunofluorescence staining of COS-1 cells (C) or primary hippocampal neurons (D) expressing Kv1.1 (left panel) or Kv1.1C79 (right panel). Arrows show aggregates formed by Kv1.1C79 subunits.

To investigate the molecular differences between Kv1.1 channels harboring these different EA-1 mutations, and how this contributes to the phenotypic variability in patients with this disease, site-directed mutagenesis was used to generate each of these mutations in rat Kv1.1. Rat Kv1.1 shares 98% amino acid sequence identity with human Kv1.1, and every position mutated in EA-1 patients is identical between wild-type rat and human Kv1.1. Each mutant Kv1.1 subunit was expressed in COS-1 cells and analyzed for solubility in buffers containing a 0.1% (v/v) concentration of the non-ionic detergent Triton X-100, a useful indicator of the folding state of membrane proteins (39). We found that wild-type Kv1.1 and each of the 13 missense mutations were soluble in 0.1% Triton X-100 (Fig. 1B), consistent with no obvious folding defects. In contrast, the Kv1.1C79 truncation mutant encoded by the R417STOP mutation exhibited a prominent detergent-insoluble fraction (Fig. 1B), indicative of misfolding. Also note that the bulk of the electrophoretic microheterogeneity of wild-type Kv1.1 is lacking in Kv1.1C79, presumably due to elimination of C-terminal phosphorylation sites (40).

Each of the EA-1 mutants was expressed in COS-1 cells and subjected to immunofluorescence staining. Wild-type Kv1.1 expressed in transfected mammalian cells is localized to the endoplasmic reticulum (33), such that immunofluorescence staining of cells expressing wild-type Kv1.1 yields a diffuse, reticular perinuclear pattern (Fig. 1C). In contrast, Kv1.1C79, while also perinuclear, is present in large membranous aggregates (Fig. 1C). A detailed quantitative analysis of the number of transfected cells exhibiting this aggregated staining pattern, which was defined as the aggregation index, revealed that the majority (56 ± 6.1%) of the cells expressing Kv1.1C79 had obvious intracellular aggregates as shown in Fig. 1C, whereas these were present in only 6.0% (± 1.0%) of the cells expressing wild-type Kv1.1. Each of the 13 missense EA-1 mutants resembled wild-type Kv1.1 in that aggregates were observed in 6% or fewer of the expressing cells (Fig. 1B).

To determine whether these differences between wild-type Kv1.1 and Kv1.1C79 existed in neurons, transfected cultured primary hippocampal neurons were subjected to immunofluorescence staining. Neurons expressing Kv1.1C79 formed perinuclear aggregates of the mutant protein similar to those observed in COS-1 cells (Fig. 1D, right panel), whereas cells expressing wild-type Kv1.1 did not (Fig. 1D, left panel). These results suggest that the mechanisms that mediate folding of Kv1 channels in COS-1 cells and cultured neurons are similar. These biochemical and immunohistochemical results together show that the 13 different missense mutations found in Kv1.1 from EA-1 patients resemble wild-type Kv1.1 and do not exhibit obvious characteristics of protein misfolding. In contrast, the truncated Kv1.1 generated by the R417STOP nonsense mutation found in patients with the drug-resistant form of EA-1 exhibits two hallmarks of misfolding, detergent insolubility and intracellular aggregation.

Truncation of the Kv1.1 Cytoplasmic C Terminus Results in Inefficient N-Linked Glycosylation and Loss of Neurotoxin Binding-- To investigate further the biochemical properties of Kv1.1C79 channels, transfected COS-1 cells were subjected to extraction in increasing concentrations of Triton X-100. Almost half of the total Kv1.1C79 pool was insoluble in 0.1% Triton X-100 (Fig. 2A). A 10-fold increase in Triton X-100 concentration did eventually solubilize the bulk of the Kv1.1C79 pool, consistent with effects of increased Triton X-100 on solubility of other misfolded proteins (39). Interestingly, the Kv1.1C79 polypeptide pool in COS-1 cells consisted of two populations, one form with an apparent mass of 55 kDa, and a second form of 50 kDa. These forms differed in their detergent solubility properties, such that the 50-kDa form was consistently enriched in the detergent-insoluble fraction (Fig. 2A).

View larger version (38K): [in this window] [in a new window]   Fig. 2.   Biochemical and cell biological analyses of Kv1.1C79. A, COS-1 cells expressing Kv1.1C79 were harvested and extracted with lysis buffer containing 0, 0.1, 0.4, 1.0, or 2.0% Triton X-100 (TX) detergent. Soluble (S) and insoluble (I) fractions were separated by centrifugation and analyzed by SDS-PAGE and immunoblot. B, COS-1 cells expressing Kv1.1C79 were harvested and extracted with lysis buffer containing 1.0% Triton X-100. Soluble fractions were treated (+) with PNGaseF (PF), EndoH (EH), or neuraminidase (NA), or buffer alone (). Samples were analyzed by SDS-PAGE and immunoblot. Arrow points to N-glycosylated subunits (high mannose type) and arrowhead points to unglycosylated subunits. C, COS-1 cells expressing Kv1.1C79 were fixed, permeabilized, and double-stained for Kv1.1C79 and calnexin. Arrow points to co-localized Kv1.1C79 and calnexin. D, BHK cells expressing Kv1.1 or Kv1.1C79 were harvested, and surface (S) or total (T) 125I--DTX toxin binding was measured. Bars represent mean ± S.D. from triplicate assays. E, COS-1 cells expressing Kv1.1 or Kv1.1C79 were harvested and solubilized with 1.0% Triton X-100. Extracts were fractionated on 5-50% linear non-denaturing sucrose gradients. Samples were analyzed by immunoprecipitation, SDS-PAGE, immunoblot, and densitometry.

The Kv1.1-deduced amino acid sequence contains a single consensus N-linked glycosylation site in the extracellular segment between the S1 and S2 transmembrane regions (41). Previously, we found that Kv1.1 is N-linked glycosylated at this site (42). To determine whether the two populations of Kv1.1C79 seen in COS-1 cells were different in N-glycosylation, membrane extracts were treated with PNGase F, an enzyme that cleaves N-linked sugars from glycoproteins. PNGase F treatment of Kv1.1C79 reveals that the higher molecular weight Kv1.1C79 pool is N-glycosylated, whereas the lower molecular weight pool is resistant to PNGase F digestion and presumably unglycosylated, such that the electrophoretic mobility of the higher molecular weight pool shifts to that of the lower molecular weight pool upon PNGase F digestion (Fig. 2B). Similar results were obtained with endoglycosidase H (Fig. 2B), consistent with the high mannose oligosaccharide present on the ER-localized Kv1.1C79 (Fig. 2B). Kv1.1C79 was insensitive to neuraminidase treatment (Fig. 2B), consistent with its ER retention. Although wild-type Kv1.1 expressed in heterologous cells exhibits substantial microheterogeneity in molecular weight on SDS gels (Fig. 1B), this is not due to differential glycosylation (43), and each of the components of the wild-type Kv1.1 pool is sensitive to both PNGase F and endoglycosidase H (42). These results together show that Kv1.1C79 is not as efficiently N-glycosylated as is wild-type Kv1.1. Moreover, the lower molecular weight unglycosylated pool of Kv1.1C79 exhibits greater resistance to detergent solubilization. These results are consistent with a model whereby the C-terminal truncation of Kv1.1C79 allosterically affects the folding of the S1-S2 segment and the subsequent addition of an N-linked oligosaccharide chain at this site.

To define the origin of the Kv1.1C79 intracellular aggregates, transfected COS-1 cells were double-labeled for the resident ER protein calnexin, and a marker of the Golgi apparatus, the lectin Lens culinaris agglutinin (44). Calnexin staining co-localized with the perinuclear aggregates containing mutant Kv1.1C79 channels (Fig. 2C) whereas L. culinaris agglutinin staining did not (data not shown). These immunofluorescence results suggest that the intracellular Kv1.1C79 aggregates are derived from ER.

To measure further the folding state of Kv1.1C79, we undertook binding studies with two neurotoxins (-DTX and DTX_K) that exhibit high affinity binding to Kv1.1. Binding sites for ¹²⁵I--DTX were detected on the cell surface of BHK cells expressing wild-type Kv1.1 (Fig. 2D). Exposure of total cellular binding sites by sonication yielded an 18-fold increase in ¹²⁵I--DTX binding in Kv1.1-expressing cells (Fig. 2D), consistent with the intracellular retention of the bulk of the expressed Kv1.1 protein (33). Negligible cell surface binding was observed in cells expressing Kv1.1C79, and cell fragmentation by sonication did not lead to an appreciable increase in total toxin binding (Fig. 2D). Immunoblot analyses performed on BHK cell membranes showed that Kv1.1 and Kv1.1C79 were expressed at similar levels in the transfected BHK cells, consistent with results from COS-1 cells (see Fig. 1B). Negligible ¹²⁵I-DTX_K binding was observed to intact or sonicated cells expressing Kv1.1C79 (data not shown). These results are consistent with the notion that misfolding of Kv1.1C79 alters the DTX-binding site, which is contributed by residues in the external mouth of the channel pore.

Unassembled and misassembled membrane protein complexes are retained in the ER and degraded by quality control mechanisms. To determine whether the distinct phenotype of Kv1.1C79 relative to wild-type Kv1.1 and the other EA-1 mutants is simply due to defects in subunit assembly, membrane extracts containing Kv1.1 or Kv1.1C79 channels were analyzed on nondenaturing continuous sucrose density gradients (45). Extractions were performed in 1% Triton X-100 to dissociate insoluble aggregates of Kv1.1C79 such that the bulk of both Kv1.1 and Kv1.1C79 were now soluble (Fig. 2A). We found that the peak of wild-type Kv1.1 immunoreactivity appeared in the gradient fraction containing proteins of apparent molecular mass of 165-285 kDa, with no evidence of pools of monomeric subunits. These results are consistent with the calculated molecular mass of Kv1.1 tetramers of 225 kDa, and with previous studies showing efficient assembly of Kv1.1 expressed in mammalian cells (43). The major peak of Kv1.1C79 co-migrated with wild-type Kv1.1, and peaks corresponding to unassembled monomeric subunits were not detected (Fig. 2E). These data suggest that the bulk of Kv1.1C79 that could be solubilized in 1% Triton X-100 was present as tetramers (the calculated molecular mass of Kv1.1C79 tetramers is 191 kDa). The Kv1.1C79 sample also contained higher molecular weight species, perhaps corresponding to Kv1.1C79 aggregates soluble in 1% Triton X-100. These results show that both wild-type Kv1.1 and Kv1.1C79 subunits are efficiently assembled, although higher molecular weight species, presumably corresponding to aggregated channel complexes, are also observed for Kv1.1C79 but not wild-type Kv1.1.

Kv1.1C79 Exhibits Characteristics Strikingly Similar to Misfolded CFTR Mutants-- Previously, it has been shown that the efficiency of intracellular trafficking of a number of integral membrane proteins, including mutant CFTR, is temperature-sensitive, implying that a polypeptide folding step was affected (46-48). Chemical chaperones, such as the polyhydric alcohol glycerol, have also been used as folding indicators, as glycerol can stabilize protein conformation and increase both the rate of in vitro protein refolding and the kinetics of oligomeric assembly (49-52). Moreover, a combination of lowered temperature and glycerol addition yields dramatic improvements in the trafficking of the misfolded F508 mutation of CFTR (36, 48). To demonstrate further that the Kv1.1C79 aggregates are composed of misfolded Kv1 channel protein, transfected cells were incubated at a lowered temperature of 25 °C and treated with 10% glycerol, or both. Incubation of transfected cells at 25 °C, or treating with 10% glycerol, resulted in minor increases in solubility of Kv1.1C79 in 0.1% Triton X-100 (Fig. 3, top panel) and moderate decreases in the number of cells forming ER-derived aggregates (Fig. 3, bottom panel). However, when these treatments were combined virtually all of the Kv1.1C79 pool was detergent-soluble (Fig. 3A, top panel), and the number of cells with intracellular Kv1.1C79 aggregates was dramatically reduced (Fig. 3A, bottom panel). Moreover, combining lower temperature and glycerol treatments resulted in a more efficient N-linked glycosylation of Kv1.1C79, such that the proportion of Kv1.1C79 in the lower molecular weight pool was greatly reduced (Fig. 3A). These results show that Kv1.1C79 behaves as a misfolded protein, similar to the prototypical F508 mutation of CFTR, in that treatments that promote proper folding rescue the detergent insolubility, aggregation, and defective N-linked glycosylation of Kv1.1C79. These rescue experiments are also consistent with a defect in protein folding, and not subunit assembly, as the basis of the unique phenotype of Kv1.1C79.

View larger version (38K): [in this window] [in a new window]   Fig. 3.   Characterization of Kv1.1C79 misfolding and assembly. A, COS-1 cells expressing Kv1.1 were incubated at 37 °C, and cells expressing Kv1.1C79 were incubated at 37 or 25 °C with or without 10% glycerol. Cells were harvested and permeabilized in lysis buffer containing 0.1% Triton X-100. Soluble (S) and insoluble (I) fractions were analyzed by immunoblot (top panel). COS-1 cells expressing Kv1.1 or Kv1.1C79 were treated with the same conditions, fixed, permeabilized, and stained. The percentage of expressing cells forming aggregates was measured and defined as the Aggregation Index. B, COS-1 cells expressing Kv1.1 (top panels) or Kv1.1C79 (bottom panels) were fixed, permeabilized, and double-stained for Kv1.1 and vimentin. Arrows point to collapsed vimentin near Kv1.1C79 aggregates. C, COS-1 cells expressing Kv1.1 (top panels) or Kv1.1C79 (bottom panels) were fixed, permeabilized, and double-stained for Kv1.1 and ubiquitin. Arrow points to aggregates containing ubiquitinated Kv1.1C79.

Recently, a novel cellular response to misfolded proteins was identified and termed the aggresome response (27). Aggresomes form when the intermediate filament protein vimentin sequesters misfolded protein aggregates, such as those formed by the F508 mutant of CTFR (27), by forming a ring-like structure around the aggregates at the centrosome. As such, a hallmark of the aggresome response is the collapse of the cellular network of vimentin. We determined whether the aggresome response occurred in cells expressing Kv1.1C79 aggregates by staining transfected COS-1 cells for vimentin and Kv1.1C79. Our immunofluorescence data show that whereas the filamentous vimentin network is not altered in cells expressing wild-type Kv1.1 (Fig. 3B, top panel), it collapses into aggresomes in cells expressing Kv1.1C79 (Fig. 3B, bottom panel). However, it is important to note that the aggregated Kv1.1C79 does not form the large structures ringed by vimentin that are typical of prototypical aggresomes (27).

Inefficient folding of membrane proteins results in ER retention and subsequent degradation by the ubiquitin-proteosome-mediated pathway (53). To determine whether this pathway was initiated in cells expressing wild-type Kv1.1 or Kv1.1C79, cells were double-labeled for Kv1 channels and ubiquitin. Aggregates composed of Kv1.1C79 subunits contained ubiquitin (Fig. 3C, bottom panel), whereas cells expressing wild-type Kv1.1 did not (Fig. 3C, top panel). These data suggest that cellular mechanisms known to respond to misfolding of other membrane proteins are also activated in cells expressing Kv1.1C79 but not wild-type Kv1.1 channels.

Kv1.1C79 Subunits Inhibit Trafficking of Heteromeric Kv1 Complexes-- The majority of Kv1 channels observed in vivo are heteromeric complexes containing other Kv1 family members and Kv-subunits (15-19). We next tested how heteromeric assembly of Kv1.1C79 with wild-type - and -subunit polypeptides, as would occur in the neurons of EA-1 patients, would affect the biochemical and cell biological characteristics of the resultant channel complexes. We co-expressed wild-type Kv1.1 or Kv1.1C79 with Kv1.2 and Kv2.1, as this specific heteromeric complex is present in myelinated axons of many central neurons and in cerebellar basket cell terminals (16, 22). Triple-label immunofluorescence staining of COS-1 cells expressing wild-type Kv1.1 and Kv1.2 -subunits and the Kv2.1 -subunit showed intracellular co-localization of these three subunits in the typical reticular ER pattern (Fig. 4A, top panels). In contrast, in cells co-expressing Kv1.1C79, Kv1.2, and Kv2.1, all three subunits were found predominantly in perinuclear aggregates (Fig. 4A, bottom panels). We also found that co-expression with Kv1.1C79 conferred detergent insolubility to co-expressed Kv1.2 and Kv2 (Fig. 4A). Interestingly, a small reduction in the amount of insoluble Kv1.1C79 was observed upon co-expression with Kv1.2 and Kv2.1 (Fig. 4A, bottom left panel), suggesting a partial rescue of some of the misfolded mutant subunits. Together these data suggest that Kv1.1C79 subunits can induce aggregation of other Kv1 - and -subunit polypeptides such that trafficking of other co-expressed Kv1 channel subunits to the cell surface is inhibited.

View larger version (58K): [in this window] [in a new window]   Fig. 4.   Effects of Kv1.1C79 on folding and trafficking of heteromeric Kv1 channel complexes. A, COS-1 cells expressing Kv1.1/Kv1.2/Kv2 or Kv1.1C79/Kv1.2/Kv2 subunits were fixed, permeabilized, and triple-labeled. Parallel cultures of cells expressing these same combinations of subunits were harvested and extracted with 0.1% Triton X-100. Soluble (S) and insoluble (I) fractions were collected and analyzed by immunoblot. Arrows point to the molecular weight of cell surface Kv1.2. B, COS-1 cells transfected with Kv1.1/1.2/2 or Kv1.1C79/1.2/2 at the described cDNA ratios were fixed and stained for cell surface and total Kv1.2, and the number of cells expressing Kv1.2 on the cell surface is defined as the Kv1.2 Surface Expression Index (Kv1.2 SEI). C, COS-1 cells expressing Kv1.1C79, Kv1.2, Kv2, and PSD-95 were fixed and stained for surface Kv1.1 channels (top panel) and then permeabilized and stained for PSD-95 (bottom panel). Arrow points to a surface Kv1.1 cluster. D, COS-1 cells expressing CD4 alone (left panels) or CD4 and Kv1.1C79 (right panels) were fixed and double-stained for CD4 and Kv1.1.

To determine the effects of Kv1.1C79 co-expression on the intracellular trafficking and surface expression properties of heteromeric channels, intact cells expressing Kv1.1C79, wild-type Kv1.2, and Kv2.1 were stained with an ectodomain-directed anti-Kv1.2 antibody. A significant dose-dependent decrease in Kv1.2 surface expression was observed upon Kv1.1C79 co-expression (Fig. 4B). These results are consistent with the dramatic decrease (arrows, Fig. 4A) observed for surface pools of Kv1.2 polypeptides that carry complex N-linked oligosaccharides (33).

A fraction of cells co-expressing Kv1.1C79, wild-type Kv1.2 and Kv2.1 did express cell surface Kv1.2. Given the negative effects of Kv1.1C79 on Kv1.2 surface expression, we next addressed whether these surface channels contained Kv1.1C79 subunits. Previously, we found that PSD-95 only clusters those Kv1 subunits expressed on the cell surface (34). The interaction of Kv1 subunits with PSD-95 is mediated via the last three C-terminal amino acids of the Kv1 subunits (54), such that Kv1.1C79 itself cannot interact with PSD-95. We found in cells expressing Kv1.2 and Kv2 that surface Kv1.1C79 can be clustered by PSD-95 (Fig. 4C). This could only result from co-assembly of Kv1.1C79 and Kv1.2 into heteromeric complexes. These data suggest that Kv1.1C79 subunits are competent to co-assemble with Kv1.2 and through direct subunit interaction affect the folding and surface expression level of co-expressed subunits.

Expression of Kv1.1C79 had no effect on the plasma membrane targeting of a non-interacting membrane protein, CD4. The number of COS-1 cells expressing CD4 on the cell surface was similar to that of cells expressing CD4 and Kv1.1C79 (Fig. 4D). These data suggest that the effects of Kv1.1C79 co-expression on the folding and trafficking of wild-type potassium channel subunits most probably arise from direct interaction.

Residues in the Proximal Region of the Cytoplasmic C Terminus Mediate Kv1.1 Folding-- As a first step in determining the region(s) within the cytoplasmic C terminus of Kv1.1 that are critical to proper folding, six other Kv1.1 C-terminal truncation mutants were generated: Kv1.1C73, Kv1.1C67, Kv1.1C60, Kv1.1C55, Kv1.1C33, and Kv1.1C13. Analyses of the solubility of these mutants in 0.1% Triton X-100 and of the percentage of expressing cells forming aggregates showed that none of these other truncation mutants exhibited the detergent insolubility and aggregation observed for Kv1.1C79 (data not shown). These results indicated that residues between amino acids 417 and 423 are critical for proper Kv1.1 folding (Fig. 5A). Furthermore, replacement of the Kv1.1 C terminus with that from Kv1.4 in the chimera Kv1.1N (residues 1-389):Kv1.4C (residues 543-654) rescues the misfolded phenotype. These results suggest that the C-terminal folding determinant may be conserved among Kv1 -subunits. The critical residues responsible for Kv1.1 folding lie just distal to the S6 transmembrane domain and are highly conserved among all mammalian Kv1 -subunits (Fig. 5B). To identify the key residues in this segment, alanine-scanning point mutagenesis was performed on each residue between amino acids 417 and 423. Mutation of each of these seven residues, with the exception of Gly-421, led to an increase in the number of cells with intracellular aggregates (Fig. 5C) and insolubility in 0.1% Triton X-100. However, no single mutation recapitulated the extent of misfolding seen for Kv1.1C79 subunits, implying that more than 1 residue in this segment contributes to the proper folding of Kv1.1 -subunits, and only through deletion of this entire segment, as occurs in EA-1 patients with the R417STOP missense mutation, is the misfolded phenotype expressed.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Analysis of sequences involved in folding of Kv1 channels. A, COS-1 cells expressing Kv1.1, Kv1.1C79, Kv1.1C73, or the Kv1.1N (residues 1-389):1.4C (residues 543-654) chimera were harvested and extracted in lysis buffer containing 0.1% Triton X-100. Soluble (S) and insoluble (I) fractions were collected and analyzed by immunoblot (left panel). COS-1 cells expressing these same constructs were fixed and immunofluorescence stained, and the number of expressing cells forming aggregates was measured and defined as the Aggregation Index (right panel). B, sequence alignment of Kv1.1 residues responsible for folding. C, COS-1 cells expressing Kv1.1R417A, E418A, T419A, E420A, G421A, E422A, and E423A were harvested and permeabilized in lysis buffer containing 0.1% Triton X-100. Soluble (S) and insoluble (I) fractions were collected and analyzed by immunoblot (left panel). The Aggregation Index in parallel cultures of COS-1 cells expressing these same constructs was also determined (right panel).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In the present study we carried out the first comprehensive analysis of the biochemical and cell biological properties of EA-1 causing mutants in the Kv1.1 potassium channel. We found that the 13 EA-1 missense mutations identified previously, although functionally diverse, have folding and subcellular localizations identical to wild-type Kv1.1 channels. However, a recently discovered nonsense EA-1 mutation, R417STOP (13) generates a truncated Kv1.1C79 polypeptide that exhibits biochemical and cell biological characteristics consistent with a misfolded phenotype. Patients harboring the R417STOP mutation have a severe and remarkably drug-resistant form of EA-1. As such, the unique biochemical and cell biological properties of the encoded mutant provide a plausible mechanism underlying the altered excitability and pathophysiology of patients harboring this particular EA-1 mutation.

Truncation of Kv1.1 by the R417STOP mutation yields Kv1.1C79 subunits with characteristics consistent with allosteric alteration of structure at two distinct sites. Glycosidase digestion of Kv1.1C79 revealed inefficient N-linked glycosylation when compared with wild-type Kv1.1 subunits. This indicates that removal of the Kv1.1 cytoplasmic C terminus induces structural alterations in the S1-S2 extracellular linker that contains the single N-linked glycosylation site. Interestingly, the unglycosylated Kv1.1C79 pools remain insoluble at detergent concentrations that effectively solubilize the bulk of the properly glycosylated mutant protein pool, suggesting that deficient glycosylation may further contribute to misfolding of Kv1.1 subunits, as has been suggested for the related Drosophila Shaker potassium channel polypeptide (55). Characterization of binding of two neurotoxins, -DTX and DTX_K, which exhibit high affinity binding to the pore region of Kv1.1 (56, 57), also suggest allosteric effects of the C-terminal truncation in Kv1.1C79 on protein folding. Tetrameric assembly is a prerequisite for high affinity DTX binding (57); however, as we have shown that Kv1.1C79 is competent for both homo- and heterotetrameric assembly, the lack of DTX binding to this mutant presumably reflects an allosterically transduced misfolding of the channel pore upon C-terminal truncation.

The segment identified in our mutational analysis that is required for proper folding of Kv1.1 (amino acids 417-423, RETEGEE) is absolutely conserved in Kv1.1, Kv1.2, and Kv1.3 in mammals, frog, and trout and highly conserved in the Drosophila Shaker -subunit and in other mammalian Kv1 potassium channel -subunits. The charged nature of this segment suggests that it is located near the surface of the protein. Secondary structure analyses (PHD Secondary Structure Prediction, EMBL, Heidelberg, Germany) of the Kv1.1-deduced amino acid sequence predict that this segment is helical in nature. It is attractive to speculate that such an exposed helical segment may play a role in intra- or intermolecular interactions required for proper folding of Kv1.1 channels. However, the lack of information on its location within the three-dimensional structure of the Kv1.1 subunit, or within the assembled tetrameric channel, precludes further predictions as to the precise role of this domain in Kv1.1 folding.

Previous studies (5, 6) have shown that certain EA-1 missense mutations express very low levels of functional channels. For example, the V174F, R239S, F249I, and E325D EA-1 mutant subunits express poorly in heterologous cells as homomeric channels and reduce the overall current when co-assembled with wild-type subunits. Potassium channel function is reduced in cells co-expressing missense mutants and wild-type Kv1.1 subunits through either dominant-negative effects or haploinsufficiency (6). The nonsense R417STOP mutation is exceptional in that the resultant Kv1.1C79 mutant cannot produce any functional channels and produces the most severe dominant-negative reduction of potassium current when co-expressed with wild-type subunits (13). The biochemical and cell biological phenotype of Kv1.1C79 is also distinct from less extensive C-terminal truncations, presumably due to the presence of the folding domain comprising amino acids 417-423. Interestingly, a previous electrophysiological study showed that the Kv1.1C72 mutant (truncated at amino acid 423) did not form functional homomeric channels (58). It is not known if this is due to lack of surface expression or altered gating of plasma membrane channels.

We show that the mechanism by which Kv1.1C79 subunits cause these effects is unique among all EA-1 mutations and is directly related to the misfolding of Kv1.1C79 subunits which influences any associated potassium channel - and -subunits. Although some Kv1.1C79 subunits co-assembled with wild-type Kv1 - and Kv-subunits are expressed on the cell surface, as shown here by interaction with PSD-95 and previously in electrophysiological studies (13), many heteromeric channels containing Kv1.1C79 are misfolded, and overall trafficking to the cell surface is dramatically reduced. As the vast majority of Kv1.1-containing channels in the mammalian central nervous system are heteromultimers containing co-assembled Kv1.2 and Kv1.4 -subunits, and Kv1.1 and Kv2.1 -subunits, the fact that overall misfolding of heteromeric channels containing Kv1.1C79 is observed is critical to the in vivo situation (16, 21). That these Kv1 channels localize to axons and synaptic terminals (23, 59) where they play a critical role in transmitter release is consistent with the hyperexcitability observed in EA-1 patients. Although all EA-1 mutations result in increased neuronal excitability, the mechanisms by which this is achieved may vary and underlie the phenotypic variabilities observed in EA-1 patients. For example, patients exhibit ataxia and myokymia to a different extent, and some also exhibit partial epilepsy (60). Furthermore, applicability of therapeutic agents and treatments may be dependent on the mechanism by which dysfunction is achieved. For example, patients harboring the R417STOP mutation exhibit resistance to conventional EA-1 therapeutics, such as carbamazepine and acetazolamide (13). Perhaps the intracellular aggregation of Kv1.1C79 subunits that is unique among EA-1 mutations leads to a more severe reduction in potassium channel expression, for example via effects on co-assembled wild-type Kv1 Kv1 - and Kv-subunits that are more difficult to overcome by treatment with these anticonvulsants.

Alternatively, the intracellular aggregation of Kv1.1C79 subunits could exert pleiotropic effects on neuronal function outside of direct effects on excitability. Much recent evidence suggests that many human diseases are caused by improper folding of nascent polypeptides as they achieve a final tertiary structure. Such proteins are either inactive or have altered activity as a result of inappropriate folding (e.g. Marfan syndrome and familiar hypercholesterolemia) or are mislocalized due to trafficking defects arising from abnormal folding (e.g. Tay-Sachs disease and 1-antitrypsin deficiency), as reviewed previously (61). Cystic fibrosis, the most common lethal autosomal genetic disease of Caucasians, is a well studied channelopathy that, in two-thirds of affected patients, arises from a mutation that generates misfolded F508 mutant CFTR polypeptides that are unable to exit the ER (62, 63). Our results show that the biochemical and cell biological properties of Kv1.1C79 are strikingly similar to CFTR F508. Like CFTR F508, Kv1.1C79 is insoluble in nonionic detergents and exists in perinuclear aggregates. Thus, misfolding and aggregation of mutant proteins like Kv1.1C79 could have consequences, such as dysregulation of cell cycle and apoptosis, beyond increased excitability and channel dysfunction that could contribute to the more severe phenotype and drug resistance in EA-1 patients harboring the R417STOP mutation. Like CFTR F508, we found that both the detergent insolubility and intracellular aggregation of Kv1.1C79 can be reversed by treating cells with chemical chaperones. As much effort in new therapeutics for cystic fibrosis is focused on using chemical chaperones to reverse the protein folding and trafficking deficiencies of CFTR F508 (64), our data suggest that similar efforts may be appropriate for certain forms of EA-1.

	ACKNOWLEDGEMENTS

We thank Drs. J. Engebrecht, M. N. Rasband, and K. J. Rhodes for critical reading of the manuscript and Dr. O. G. Shamotienko for assistance with the BHK cell expression.

	Note Added in Proof

Other investigators (65) have also found evidence of altered trafficking of the R417STOP mutation.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grant NS34383 (to J. S. T.) and by the Wellcome Trust (to J. O. D.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Dept. of Biochemistry and Cell Biology, State University of New York, Stony Brook, NY 11794-5215. Tel.: 631-632-9171; Fax: 631-632-9714; E-mail: james.trimmer@sunysb.edu.

Published, JBC Papers in Press, October 25, 2001, DOI 10.1074/jbc.M109325200

	ABBREVIATIONS

The abbreviations used are: EA-1, episodic ataxia type 1; -DTX, -dendrotoxin; BHK, baby hamster kidney; ER, endoplasmic reticulum; CFTR, cystic fibrosis transmembrane conductance regulator; PNGase, peptide N-glycosidase.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				L. Mckeown, M. P. Burnham, C. Hodson, and O. T. Jones Identification of an Evolutionarily Conserved Extracellular Threonine Residue Critical for Surface Expression and Its Potential Coupling of Adjacent Voltage-sensing and Gating Domains in Voltage-gated Potassium Channels J. Biol. Chem., October 31, 2008; 283(44): 30421 - 30432. [Abstract] [Full Text] [PDF]

				Y. Ogawa, I. Horresh, J. S. Trimmer, D. S. Bredt, E. Peles, and M. N. Rasband Postsynaptic Density-93 Clusters Kv1 Channels at Axon Initial Segments Independently of Caspr2 J. Neurosci., May 28, 2008; 28(22): 5731 - 5739. [Abstract] [Full Text] [PDF]

				D. P. Mohapatra, D. F. Siino, and J. S. Trimmer Interdomain Cytoplasmic Interactions Govern the Intracellular Trafficking, Gating, and Modulation of the Kv2.1 Channel J. Neurosci., May 7, 2008; 28(19): 4982 - 4994. [Abstract] [Full Text] [PDF]

				H. Vacher, D. P. Mohapatra, H. Misonou, and J. S. Trimmer Regulation of Kv1 channel trafficking by the mamba snake neurotoxin dendrotoxin K FASEB J, March 1, 2007; 21(3): 906 - 914. [Abstract] [Full Text] [PDF]

				Q. Gong, M. A. Jones, and Z. Zhou Mechanisms of Pharmacological Rescue of Trafficking-defective hERG Mutant Channels in Human Long QT Syndrome J. Biol. Chem., February 17, 2006; 281(7): 4069 - 4074. [Abstract] [Full Text] [PDF]

				H. Abriel and O. Staub Ubiquitylation of Ion Channels Physiology, December 1, 2005; 20(6): 398 - 407. [Abstract] [Full Text] [PDF]

				J. Wan, R. Khanna, M. Sandusky, D. M. Papazian, J. C. Jen, and R. W. Baloh CACNA1A mutations causing episodic and progressive ataxia alter channel trafficking and kinetics Neurology, June 28, 2005; 64(12): 2090 - 2097. [Abstract] [Full Text] [PDF]

				Q. Gong, D. R. Keeney, M. Molinari, and Z. Zhou Degradation of Trafficking-defective Long QT Syndrome Type II Mutant Channels by the Ubiquitin-Proteasome Pathway J. Biol. Chem., May 13, 2005; 280(19): 19419 - 19425. [Abstract] [Full Text] [PDF]

				R. Borgatti, C. Zucca, A. Cavallini, M. Ferrario, C. Panzeri, P. Castaldo, M. V. Soldovieri, C. Baschirotto, N. Bresolin, B. D. Bernardina, et al. A novel mutation in KCNQ2 associated with BFNC, drug resistant epilepsy, and mental retardation Neurology, July 13, 2004; 63(1): 57 - 65. [Abstract] [Full Text] [PDF]

				R. Khanna, E. J. Lee, and D. M. Papazian Transient calnexin interaction confers long-term stability on folded K+ channel protein in the ER J. Cell Sci., June 15, 2004; 117(14): 2897 - 2908. [Abstract] [Full Text] [PDF]

				P. Castaldo, P. Stefanoni, F. Miceli, G. Coppola, E. M. del Giudice, G. Bellini, A. Pascotto, J. R. Trudell, N. L. Harrison, L. Annunziato, et al. A Novel Hyperekplexia-causing Mutation in the Pre-transmembrane Segment 1 of the Human Glycine Receptor {alpha}1 Subunit Reduces Membrane Expression and Impairs Gating by Agonists J. Biol. Chem., June 11, 2004; 279(24): 25598 - 25604. [Abstract] [Full Text] [PDF]

				A. G. Ostermeyer, L. T. Ramcharan, Y. Zeng, D. M. Lublin, and D. A. Brown Role of the hydrophobic domain in targeting caveolin-1 to lipid droplets J. Cell Biol., January 5, 2004; 164(1): 69 - 78. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 276/52/49427    most recent M109325200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Manganas, L. N. Articles by Trimmer, J. S. Search for Related Content PubMed PubMed Citation Articles by Manganas, L. N. Articles by Trimmer, J. S. Social Bookmarking                      What's this?