GENOMICS

12, 729-737

(19%)

Molecular Cloning, Characterization,

and Genomic Localization

of a Human Potassium Channel Gene MARK E. CURRAN,*,’

GREGORY M. L,wxs,t

AND MARK T. KEATING*~

*Department of Human Genetics and Eccles Program in Human Molecular Biology and Genetics; *Division of Cardiology, University of Utah Health Sciences Center, Salt Lake City, Utah 84112; and tDepartment of Human Genetics, Integrated Genetics, Inc., Framingham, Massachusetts 07 701 Received June 27, 1991; revised November 27. 1991

Potassium (K+) channels are critical for a variety of cell functions, including modulation of action potentials, determination of resting membrane potential, and development of memory and learning. In addition to their role in regulating myocyte excitability, cardiac K+ channels control heart rate and coronary vascular tone and are implicated in the development of arrhythmias. We report here the cloning and sequencing of a Kt channel gene, KCNAl, derived from a human cardiac cDNA library and the chromosomal localization of the corresponding genomic clone. Oligonucleotides based on a delayed rectifier Kt channel gene were used in PCR reactions with human genomic DNA to amplify the 54-56 regions of several different K+ channel genes. These sequences were used to isolate clones from a human cardiac cDNA library. We sequenced one of these clones, HCKl. HCKl contains putative S2-S6 domains and shares -70% sequence homology with previously isolated Shaker homologues. HCKl was used to screen human cosmid libraries and a genomic clone was isolated. By sequencing the genomic clones, a putative Sl domain and translation initiation sequences were identified. Genomic mapping using human-rodent somatic cell panels and in situ hybridization with human metaphase chromosomes have localized KCNAl to the distal short arm of human chromosome 12. This work is an important step in the study of human cardiac K+ channel structure and function and will be of use in the o 1992 Academic PWS, IIIC. study of human inherited disease.

INTRODUCTION

Potassium (K+) channels are important in a variety of cellular functions and critical for the regulation of excitable cells. In the heart, K+ channels regulate the rate of spontaneous depolarization of pacemaker cells, thereby controlling heart rate (Hille, 1984). K+ channels may also regulate coronary vascular tone and have been implicated in the origin of cardiac arrythmias (Fish et al., 1990). The identification and molecular characterization of Shalzer, a rapidly inactivating voltage-sensi-

’ To

whom

correspondence should

be addressed.

tive K+ channel in Drosophila (Papazian et al., 1987) has facilitated the study of K+ channels in mammalian systems. Much of this work, however, has focused on the molecular physiology of the central nervous system in rodent models (Hoger et al., 1991; McCormack et al., 1990; Swanson et aZ., 1990; Temple et aZ., 1988), and one K+ channel has been isolated from human brain (HBKS, Grupe et al., 1990). Although relatively little attention has been paid to the heart, several cardiac K+ channels have recently been isolated from rat (RKlRK5; Roberds and Tamkun, 1991) and human sources (HKl and HK2; Tamkun et al., 1991). The long QT syndrome (LQT) is a dominantly inherited disorder that can cause sudden death from Cardiac arrhythmias (Ward, 1964; Romano, 1965). The biochemical basis of LQT is not known, but presumably involves mutations in a gene that is important for the regulation of myocyte repolarization. Two circumstantial pieces of evidence have implicated K+ channels in the pathogenesis of LQT. First, pharmacological agents that block K+ channels (for example, type-l antiarrhythmic agents like quinidine, tricyclic antidepressant agents, cesium) and certain metabolic abnormalities (particularly hypokalemia) can induce a phenotype that is similar to LQT in purportedly normal individuals. Like their counterparts who have inherited this disorder because of a single mutant gene, individuals with acquired LQT frequently have prolongation of the QT interval on electrocardiogram and may develop cardiac arrhythmias (Moss and Schwartz, 1979). Second, K+ channels activators, like pinacidil, have been shown to reduce the incidence of repolarization-related ventricular arrhythmias in animal models (Davidenko et aZ., 1989). These data support the hypothesis that LQT is caused by mutations in a cardiac K+ channel gene. As no K+ channel markers that would be useful for linkage analyses of disease families have been reported, it was not possible to test this hypothesis. In this study we report the molecular cloning, characterization, and genomic localization of a human K+ channel gene, KCNAl, and describe a polymorphic marker associated with this locus.

729 Copyright 0 1992 All rights of reproduction

ofm-7543/92 $3.00 by Academic Press, Inc. in any form reserved.

730

CURRAN,

MATERIALS

AND

LANDES,

METHODS

Polymerase chain reactkms and isolation of human S4-S6 domains. Two oligonucleotide primers were based on the sequence of the human brain K+ channel HBK2 (Grupe et al., 1990) and used to amplify the S4-S6 domains of K+ channel sequences from total human genomic DNA. The sequence of primer S4 contains a 24-bp recognition sequence, 5’ CAGGCCTCCATGAGGGAGCTGGGG 3’ spanning nucleotides 1107-1131 of HBKZ, a 6-bp EcoRI site, 5’ GAATTC 3’, and an additional 6 bp of 5’ sequence to allow complete digestion of the PCR products with EcoRI. The sequence of S6 contains a 22-bp recognition sequence, 5’ CGATGGCACACAGCGAGCCCAC 3’ spanning nucleotides 1311-1333 of HBK2, a 6-bp Hind111 site, 5’AAGCTT 3’, and an additional 6 bp of 5’ sequence to ensure digestion with HindIll. A total of 200 pg of total human DNA was amplified in a 100~~1 reaction volume as described (Saiki et al., 1988). Amplifications were carried out in a Perkin-Elmer/Cetus DNA thermal cycler as follows: initial denaturation at 94OC for 10 min, followed by 25 cycles of denaturation at 94°C for 1 min, 62°C for 1 min, 74°C for 2 min. PCR products were subsequently digested with EcoRI and Hind111 and cloned into pUC19 for sequence analysis (Maniatis et al, 1982). Isolation of cDNA clones. Approximately 1 X lo6 independent recombinants from a human cardiac cDNA library (Clontech No. HL1038b) were lifted and fixed in duplicate to Biotrans membranes and screened with PCR clone K2 as described by the manufacturer (ICN biochemicals). Radiolabeled probe was prepared by PCR as follows: 25 ng of K2 DNA was amplified in a 20 ~1 reaction containing 100 PM dGTP, 100 pM dATP, 100 PM dTTP, 8 /LM dCTP, 1.65 /.&f [a-32P]dCTP, 1 PM S4 primer, 1 PM S6 primer, 10 n&f Tris (pH 8.3 at 20°C), 50 mM KCl, 1.5 m&f MgCl,, 0.01% gelatin, and 2.5 units Taq polymerase. Filters were hybridized overnight at 65°C washed twice for 20 min in 2~ SSC/O.l% SDS at 22°C and twice in 0.2X SSC/O.l% SDS at 65°C for 30 min. Positive plaques were isolated to homogeneity, phage DNA purified by the plate lysate method, and the insert subcloned into pUC18 for sequence analysis as described (Maniatis et al., 1982). Isolation of genomic clones. Approximately 1 X lo6 independent recombinants from a total human cosmid library (Stratagene NO. 951202) were lifted and fixed in duplicate to Biotrans membranes and screened with cDNA clone HCKl under conditions described by the manufacturer (ICN Biochemicals). Probe DNA was prepared by the random hexamer method (Feinberg and Vogelstein, 1983) and hybridized to filters overnight at 65°C. Washes were carried out in 0.1X SSC/O.l% SDS at 65°C for 45 min and filters exposed to Kodak XAR film overnight at -70°C. DNA sequence analysis. DNAs were sequenced by dideoxy chain termination methods (Sanger et al., 1977) using either modified T7 polymerase (Sequenase, US Biochemicals) or Taq polymerase (TaqTrack, Promega). Sequence was determined for both strands using a primer walking strategy. Oligonucleotide primers were synthesized by the University of Utah Department of Human Genetics DNA synthesis facility. RNA preparation and analysis. RNA was extracted from tissues as described by Chomczynski and Sacchi (1987). cDNA was made

AND

KEATING

through reverse transcription of 1 gg of total RNA, with synthesis primed by random hexamers (Perkin-Elmer/Cetus Corp., Norwalk, CT). PCR was then performed on the cDNA using one primer from the Sl domain, 5’ GAGACCCTGCCTGAGTTCAG 3’, and a second primer from the S2 domain, 5’ GGGTCCTGGGCAGGAGCGGTG 3’. PCR was as described except that annealing was at 67°C. To control for potential contamination by genomic DNA, PCR was performed on the cDNA samples using primers derived from a human skeletal muscle sodium channel that is not expressed in brain, heart, or colon tissues (L. Ptacek personal communication, data not shown). Chromosomal localization. The human/rodent somatic cell panel number 1, GM/NA07300, GM/NA07301 and the murine, human, and Chinese hamster control DNAs were obtained from the National Institute of General Medical Sciences Human Genetic Mutant Cell Repository, Coriell Institute for Medical Research (Camden, NJ). The human chromosomal content of each cell line is shown in Table 2 (Taggart et al., 1985; Warburton et al., 1990). PCR analysis of the panel DNAs was used to establish the chromosomal location of KCNAl. Two oligonucleotide primers were designed against nontranscribed, single-copy sequence from the KCNAl locus: Primer 1, 5’ AGGAAAAATACCTGATCCTGACTGGC 3’; primer 2,5’ TATGAACATGAAAATCAGTATGTG 3’. PCR products were fractionated on 3% nusieve/l% agarose gels, stained with ethidium bromide, and scored for presence or absence of the proper PCR product. In situ hybridization. Metaphase spreads were prepared from normal cultured lymphocytes (46,XY) by standard procedures of colcimid arrest, hypotonic treatment, and acetic acid-methanol fixation. Probe preparation and fluorescent in situ hybridization was carried out as described by Lichter et al. (1988) with the following changes. Hybridizations were performed using 5-10 pg/ml of biotinylated probe, 200 Fg/ml total human Cot 1 DNA (GIBCO BRL, Gaithersburg, MD) 800 rg/mL salmon sperm DNA in a cocktail of 50% formamide, 6X SSC (pH 7.0), and 10% dextran sulfate. Probe containing cocktail was applied to sample slide, overlayed with a coverslip, and sealed. The probe and sample DNAs were then simultaneously denatured at 80°C for 7 min. Following hybridization and washing, a counterstain solution consisting of 100 rig/ml 4’,6-diamidino-2-phenylindole-dihydrochloride (DAPI), and propidium iodide (PI) in 20 mMTris-HCl (pH 8.0)/ 90% glycerol was pipeted onto the slide, overlayed with a coverslip, and sealed. Preparations were visualized on a Zeiss photomicroscope equipped with DAPI, PI, and FITC epifluorescence optics. Photographs were taken with Kodak Ektachrome 400 film. Genetic mapping. Probe DNAs were labeled to high specific activity and hybridized to Southern blots of restriction enzyme-digested genomic DNA from six unrelated individuals. Twenty-seven different restriction enzymes were analyzed for each probe. Linkage analysis was performed using the LINKAGE software suite (Lathrop and Lalouel, 1984).

RESULTS Amplification

of conserved K+

channel sequences. To human cardiac cDNA libraries, specific oligonucleotides based on the DNA se-

generate

probes

for

screening

HBK2 Kl K2

TACTTCGCAGAGGCTGACGATGACGArPCGCTTTTTCCCAGCATCCCGGATGCCTTCTGGTGGGCAGTGGTTACA ------------------A-CC-G-G~-C-A---CT-T--------T--C--------------------C--C ------------------A-CC-G-G~-C-A---C--T--------T--C--------------------C--C

HBK2 Kl K2

ATGACCACGGTAGGTTACGGGGACATGTACCCCATGACTGTGGGGGG~GATC~GGCTCGCT~ --------T--G--C------------AGG---C--C-----T-----C----------------------------------T--G--C------------AGG---C--C-----T-----C---------------------------

FIG. 1. Nucleotide alignment of HBKZ and PCR derivatives. PCR primers, based on DNA sequences from the S4 and S6 transmembrane domains of the human brain K+ channel gene HBKB, were used to amplify total human DNA. Although specific primers were used for PCR, two clones that were related but not identical to HBK2 were obtained. Nucleotide sequences from these clones are shown with the S4 and S6 primer sites underlined. Kl is 88% and K2 is 90% identical to HBK2 in this region.

HUMAN

CARDIAC

K+

CHANNEL

731

KCNAl

CAGGGCCAAGCGAGGGGATCGCGCCAGCAACCCCAGTCTCCCCAGA~GGGGCCGGCC~CGCT~AGCG~GCCTGACGCCA~CGCC~GGAGCGTG AGTAGGGGGCGCGGGAGCCGGTCAGCTGGGGCGCAGCATGCCCTCT~TCCC~GCCA~GAGA~GCCCTGGTGCCCCTGGA~CGGCGGTGCCATG

-58 42

MEIALVPLENGGAM

14

ACCGTCAGAGGAGGCGATGAGGCCCGGGCAGGCTGCGGCCAGGCCACAGGGG~GAGC~CAGTGTCCCCCGACGGCTGGGCTCAGCGA~GGCC~G LQCPPTAGLSDGPK TVRGGDEARAGCGQATGGE

141 47

GAGCCGGCGCCAAAGGGGCGCGGCGCGCAGAGAGACGCGGACTCGG~GTGC~CCCTTGCCTCCGCTGCCGGACCCGGGAGT~GGCCCTTGCCTCCG EPAPKGRGAQRDADSGVRPLPPLPDPGVRFLPP

240 80

CTGCCAGAGGAGCTGCCACGGCCTCGACGGCCGCCGGC LPEELPRPRRPPPEDEEEE

339 113

GDPGLGTVEDQALG

ACGGCGTCCCTGCACCACCAGCGCGTCCACATCAACATCTCCGGGCTGCGCT~GAGACGCAGC~GGCACCCTGGCGCAGTTCCCC~CACACTCCTG SGLRFETQLGTLAQFPNTLL TASLHHQRVHINI *

438 146

GGGGACCCCGCCAAGCGCCTGCGCTAC~CGACCCCCTGAGG~CGAGTACTTCTTCGACCGC~CCGGCCCAGCTTCGACGGTATCCTCTACTACTAC DRNRPSFDGILYYY GDPAKRLRYFDPLRNEYFF

531 179

CAGTCCGGGGGCCGCCTGCGGAGGCCGGTCAACGTCTCCCTGGACGTGTTCGCGGACGAGATAC~TTCTACCAGCTGGGGGACGAGGC~TGGA~GC QSGGRLRRPVNV SLDVFADEIRFYQLGDEAMER *

636 212

TTCCGCGAGGATGAGGGCTTCATTAAAGAAGAGGAGARGCCCCTGCCCCGC~CGAGTTCCAGC~CAGGTGTGGC~ATCTTCGAGTA~CGGA~GC FREDEGFIKEEEKPLPRNEFQRQVWLI FEYPES Sl TCTGGGTCCGCGCGGGCCATCGCCATCGTCTCGGTCTTGG~ATCCTCATCTCCATCATCACCTTCTGCTTGGAGACCCTGCCTGAGTTCAGGGATG~ SGSARAIAIVSVLVILISI ITFCLETLPEFRDE

735 245

CGTGAGCTCGTCCGCCACCCTCCGGCGCCCCACCAGCCTCCCGCGCCCGCCCCTGGGGCC~CG~AGCG~GTCATGGCCCC~CCTCTGGCCCTACG RELVRHPPAPHQPPAPAPGANGSGVMA PPSGPT s2 GTGGCACCGCTCCTGCCCAGGACCCTGGCCGACCCCCTTCTTCATCGTGGAGACCACGTGCGTCATCTGGTTCACCTTCGAGCTGCTCGT~GCTTCTTC VAPLLPRTLADPFFIVETTCVIWFTFELLVRFF 53 GCCTGCCCCAGCAAGCCAGGGTTCTCCCCGG~CATCATG~CATCA~GATGTGGTGGCCATCTTCCCCTACTTCATCACCCT~GCACCG~CT~CA ACPSKAGFSRNIMNIIDVVAIFPYFITLGTELA

933 311

834 278

1032 344 1131 371 S4-

GAGCAGCAGCCAGGGGGCGGAGGAGGCGGC~CCAG~TGGGCAGCAGGCCATGTCCCTGGCCATCCTCCGAGTCATCCGCCTGGTCCGGGTGTTCCG~TC EQQPGGGGGGQNGQQAMSLAILRVIRLVRVFRI

1230 410 55

TTCAAGCTCTCCCGCCACTCCAAGGGGGCAGATCCTGGTCATCTTCTTCCTCTTCATC FKLSRHSKGLQILGKTLQASMRELGLLIFFLFI

1329 443

GGGGTCATCCTCTTCTCCAGTGCCGTCTACTTCTACTTCGCAGAGGCTGAC~CCAGG~CCCATTTCTCTAGCATCCCTGACGCCTTCTGGTG~CAGT~TC GVILFSSAVYFAEADNQGTHFSSIPDAFWWAVV

1428 476

ACCATGACCACTGTGGGCTACGGGGACATGACATCATCGTGGGCTCGCTGTGTGCCATCGCCGGGGTCCTCACCATTGCC TMTTVGYGDMRPIT~GGKI~GSLCAIAGVLTIA

1521 509

CTGCCTGTGCCCGTCATCGTCTCCAACTTCAACTACTTCTGCACTCAG LPVPVIVSNFNYFYHRETDHEEPAVLKEEQGTQ

1626 542

AGCCAGGGGCCGGGGCTGGACAGAGGAGTCCAGCGGAAGGGT SQGPGLDRGVQRKVSGSRGSFCKAGGTLENADS

1725 575

GCCCGAAGGCAGCTGCCCCCTAGAG~GTG~CGTC~~CC~~GC~CGTGGACTTGCG~GGTCCCTTTAT~CCTCTGCCTG~CACCAGCCG ARRQLPPREV

1824

GGAAACAGATTTGTGAAAGGAGATTCAGGCAGACTGGTGG~GTGGAGTAGG~TGG~GGCTTGCTG~CATGGATATCTACATTATACCGCA~GT ATTTGAAGTCACACTGTAACCTCAGTCTACCCCTCTACCCCTCTCCTTTCACTCCTTTCCTCCCTCCCTCGATCCCCCCATTTTCTCTATTCTTTCCATGACACCCA AGGGTCGCCTATTPTTAAAAAGTACCACATTCCATGACGCA~AGC~TGG~TGGT~GCGCTGTGAGATGGATGTA~TGTAGCCAGTCTCCTATA CCCAGCAGAGGGATAACCCAC~TGACTCTAAATAT ATGATTGTAmGTGTATAGTATTATTPTTATGCCTGGT~GTGGCTT~~ACTGTAGTTCAGATAGA~TATT~GGGTATATT~C~ GATACATGTTGTATTTATGGAAGAAAGAGTTGTCCTGATG~T~C~TGTTACTTATATTAGAGTCAGA~TC~GGTATGG~TGTTCTGTTTCCTG TGTCTCCAAGCCTCT~CTCTGGGATGTGGTATTGGTGC GCCTlWACAATTCTTGTAACmCTT CAAAAAGCATTTTAATGATATTGGAGGAATACTTCTGATAATTTATTGTCTTCCC

1923 2022 2121 2220 2319 2418 2517 2598

585

FIG. 2. Nucleotide and putative amino acid sequence of human cardiac K+ channel gene, KCNAl. Nucleotides are numbered +l through f2595 from the first base of the putative initiator methionine. A single open reading frame encodes 584 amino acids and predicts a protein with a molecular weight of 64,290. Six putative membrane spanning domains are overlined and numbered consecutively Sl through S6. Two consensus sites for N-linked glycosylation are indicated by asterisks (*) at asparagine 125 and 190. A single consensus sequence for phosphorylation by CAMP-dependent Kinase A is present at serine 556 and is marked with a solid circle (0).

732

CURRAN,

LANDES,

AND

quence of HBKB, a human brain K+ channel gene (Grupe et aZ., 1990), were used as primers to amplify human genomic DNA. Two PCR products were isolated, cloned, and sequenced; nucleotide alignment of HBK2 and these PCR-derived clones is shown in Fig. 1. Although specific oligonucleotides were used in these experiments, the DNA sequences of these clones were not identical to HBKB. As only a small number of PCR prod-

KEATING

ucts were characterized, it is possible that a clone identical to HBK2 was present. Isolation and sequence analysis of cDNA and cosmid clones. One PCR-derived clone (clone K2) was radiolabeled and used to screen a human cardiac cDNA library. Initially, eight plaques were identified and purified. Oligonucleotides based on vector-specific sequences were used in PCR experiments to amplify the inserts con-

KCNAl HK2 HBK2 HKl SHAKER

MEIALVPLEN _____-----

GG --

--V-M-SA-S

SGCNSHMPYG Y-AQA-ARER

KCNAl HK2 HBK2 HKl SHAKER

SGVRPLPPLP DPGVR PLPPLPEE ____-----------_--_-MSEKS-TLA CTSHDPQSSR GSRR-RRQRS EKKKAHYRQS SFPHCSDLMP SGSEEKILRQKEQLEQKEE QKKIAERKLQ LREQQLQRNS LIX

KCNAl HK2 HBK2 HKl SHAKER

GTVEDQALGT __-_____-FPEAGG-G YYS--DHGDE GGGPQHFEPI

ASLHH QRVHI ----_____ E-LVGCCSS C-YTDLLPQD EGGGGYSSVR YSDCCE--VCE--VPHD-DF

NISGLRFETQ LGTLAQFPNT LLGDPAKRLR ---------_----___-------__--___--____ -RS-C&--D-----GR-Vm------E____- E-RTQ -V--------R--N---J,---e--R---V--e-----

KCNAl HK2 HBK2 HKl SHAKER

YFDPLRNEYF ___------~-em-----__--______ ___---____

FDRNRPSFIX ------------------A ---------A ---s-----A

VFADEIRFYQ --__----__ I-LE---e-v I-TE-m----SE--K--E

AMTVRGGDE ARAGCGQATG GELQCPPTAG LSDGPKEPAP KGRGAQRDAD -------------------------------------- ------

ILYYYQSGGR --____---_ ___--____--________ __-__--___

E-LAHSR-AA

LRRPVNVSLD ---G-e--------e-p--K-----PF-----m-p--

AAAVAAA-AA M&Iv--

VEGSGGSGGG SHHHH-SRGA -YGLGEDRQH RKKQQ-QQQH LPRPRRPPPE -----_---A-GEV-G-EG -SEEEEDEEYGSLPKLSSQ

LGDEAMERFR -----_--------La---E--LLK----Q-INK--

DEEEEGDPGL ---------EQQDA-E----EEGRF ---GGAGHGF

EDEGFIK ----------CLPEG -----TJRE -----__

Sl

EEEKPLPRNE FQRQVWLIFE YPESSG@ ---------- ____ ------mTe-re -------SQp -------L-m --____ --G ---m--E--KK-I--L----e-S e-G ---R---D-K--K---L----e-Q --V

KCNAl HK2 HBK2 HKl SHAKER

PAPHQPPAPA PGANGSGVMA PPSGPTVAPL LPRTLAD _-________ -VSRGSQEEE EDEDDSYTFH H-ITP-E-GT SHYKVFNT TTN-TKIEED EVPDIT-

KCNAl Hi;

CPSKAGFSRN IMNIIDWAI

FPYFITLGTE

L

-------------PA-F--

_-____---_ L--------D I------A-V

_ _

---Q-L-FK-

SHAKER

IAIVSVLVIL ----------__--_-___ --_-----__ V--I--F---

LPEFR ISIITF+ET ------or ---VI------QFRVLXRG ---VI-__ ---L--VI-__ --e-K

KCNAl HK2 HBK2 HKl SHAKER

--N-LN-C-D

-ll:;iE;;TLm m-----L--------I-SV--V--II--

AE ?!z

QQQ-

:;?fg

A

-----_______ D,,,,SL-p--e

FNYFYHRETD HEE

KCNAl HK2 HBK2 HKl SHAKER FIG. 3. channels Computer alignment. potential sequence

GLDRGVQRKV SGSRGSFCK _____-__---------_ APDL RATDNGLGK Y-PSNLLK-F R-STSS-LGD A-GQHL~ L-E-SSDIMD

AGGTLEN ------PD FPE KSEYLEM LDDGIDATT~

DERELVRHP -----L--GNNGGVS-VS D-RDLVMAL

ADSARRQ ------ANRERRP EEGVKES GLTDHTGRHM

LPPREV* --LEKCNVKA SYLPTPHRAY LCAKEEKCQG VPFLRTQQSF

Comparison of KCNAl to four other K+ channels. Predicted amino acid sequences of (HKl and HKZ), a human brain K+ channel (HBKB), and the Drosophila K+ channel Group, Inc., Pileup algorithm. Amino acid identity is indicated as a dash and sequence Six putative transmembrane domains, conserved in all five channels, are indicated by site of phosphorylation by CAMP-dependent Kinase A, approximately 40 amino acids is conserved in all but the brain channel, HBKZ.

--__--___--_______-

PAVLK EEQGTQSQGP

KSNVDL RRS LYALC AEKRMLTEVV KGDDSETDKN NCSNAKAVET EKQQLQLQLQ LQQQSQSPHG KCNAl, two additional human cardiac K+ Shaker were compared using the Genetics gaps were introduced as spaces to optimize boxes and labelled Sl through S6. A single downstream from S6, is underlined. This

HUMAN

CARDIAC

K+

tained in these clones, and the size of each insert was determined by agarose gel electrophoresis. The clone containing the largest insert was subcloned into pBluescript for further characterization. This gene has been designated KCNAl by the Human Gene Nomenclature Committee. This designation stands for potassium voltage-gated channel, Shaker-related subfamily, member 1. The cDNA clone has been named HCKl. HCKl contains 1602 nucleotides with -70% sequence identity to HBK2 (data not shown). An open reading frame of 811 bp was identified. No poly (A)+ tail or polyadenylation signal sequence was found in the 791 bp of putative 3’ untranslated sequence. Comparison of this sequence to previously studied K+ channel genes suggested that HCKl is a partial cDNA clone lacking the 5’ sequences responsible for translation initiation and the putative Sl transmembrane domain. To identify the complete coding sequence of KCNAl, we speculated that this gene, like several previously identified mammalian K+ channel genes, lacked introns. HCKl was radiolabeled and used to screen a human genomic library. Approximately 1 X lo6 cosmid clones were screened and 22 clones were isolated. These 22 cosmids were subdivided into six distinct families based on EcoRI digestion profiles and Southern hybridization to HCKl (data not shown). The EcoRI fragments that hybridized were subcloned and the genomic clone corresponding to KCNAl was identified from this group by DNA sequencing. Complete nucleotide homology between the cDNA and the genomic clone was found and no intervening sequences were identified. The complete open reading frame of KCNAl was identified by sequence analysis of the genomic clone (Fig. 2). The putative coding region of KCNAl spans 584 amino acids and encodes a protein with a predicted molecular mass of 64,290. A potential initiation methionine

KCNAl HK2 XCNAl HK2 KCNAl HK2 KCNAl HK2

CHANNEL

733

KCNAl

codon was identified by its position 78 nucleotides downstream from an inframe stop codon and by consensus to the initiation sequence defined by Kozak (1986). A hydrophilicity plot shows six hydrophobic regions numbered Sl-S6 in Fig. 2. These hydrophobic domains are thought to be membrane spanning domains, a structure consistent with other K+ channels that have been studied. The fourth putative transmembrane domain (S4) has a charged residue at every third position, a motif that is thought to be important in conferring voltage gating properties to the channel (Papazian et al., 1991). One potential site for phosphorylation by CAMP-dependent protein kinase (R/K-R/K-X-(X)-S/T; Krebs and Beavo, 1979) was identified at residue Serine 556. Two potential sites for N-linked glycosylation (N-X-S/T; Hubbard and Ivatt, 1981) were identified at Asparagine residues 125 and 190. Comparison

of KCNAl

and other K+ channel genes.

To identify evolutionarily conserved regions of KCNAl we compared its predicted amino acid sequence to that of Shaker, a K+ channel isolated from Drosophila melanogaster. As shown in Fig. 3, significant regions of sequence conservation were identified. Amino acid homology is most apparent in the putative transmembrane domains, particularly in the S4, S5, and S6 domains where amino acid identity is 100, 96, and 93%, respectively. The region separating the fifth and sixth membrane spanning domains is also highly conserved. This is of functional significance as this region is thought to be important for ion conductance and selectivity (Yellen et al., 1991). One potential site of phosphorylation by CAMP-dependent protein kinase A at Serine 556 is also conserved, suggesting that this site may be important for channel regulation. The most divergent regions were in the amino terminus, a region thought to be involved in the inactivation of voltage-gated K+ channels (Van-

Sl CTGGGTCCGCGCGGGCCATCGCCATCGTCTCGGTCTTGGTTATCCTCATCTCCATCATCA :::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::: CTGGGTCCGCGCGGGCCATCGCCATCGTCTCGGTCTTGGTTATCCTCATCTCCATCATCA CCTTCTGCTTGGAGACCCTGCCTGAGTTCAGGGATGAACGTGAGCTCGTCCGCCACCCTC :::::::::::::::::::: :::::::::::::::::::::::::: : : : : :::::::: CCTTCTGCTTGGAGACCCTGCCTGAGTTCAGGGATGAACGCCTC CGGCGCCCCACCAGCCTCCCGCGCCCGCCCCTGGGGCCAAGGCCC :::: :::::::::::::::::::::::::::::: :::::: ::::::::::: CGGCCCCCCACCAGCCTCCCGCGCCCGCCCCTGGG-CCAACGGGGTCA-----CGCCCTCTGGCCCTACGGTGGCACCGCTCCTGCCCAGGACCCTGGCCGACCCCTTCTTCA ::::::::::::::::::::::::::::::::::::::::::::::::::::: -------TGGCCCTACGGTGGCACCGCTCCTGCCCAGGACCCTGGCCGACCCCTTCTTCA 52

KCNAl HK2

TCGTGGAGACCACGTGCGTCATCTGGTTCACCTTCGAGCTGCTCGTGCGC~CTTCGCCT : : :::::::::::::::::::::::::::::::::::::::::::::::::::::::::: TCGTGGAGACCACGTGCGTGATCTGGTTCACCTTCGAGCTGCTCGTGCGCTTC~CGCCT

FIG. 4. DNA alignment of the Sl-S2 domains of KCNAl and HK2. The divergent region of two highly related K-t channel genes, KCNAl and HK2, are shown. The putative Sl and S2 transmembrane domains are overlined. Nucleotide identity is indicated by dots (:) and gaps in the alignment are indicated as spaces. KCNAl contains an additional 15 bp in the extracellular loop spanning Sl and S2. Several base changes are also evident between the two channels.

734

CURRAN,

LANDES,

AND

northern analysis as the conservation of sequence between different K+ channels did not allow specific detection of the KCNAl transcript by direct hybridization, First-strand cDNA from cardiac, brain, placenta, and normal colon were amplified as described under Materials and Methods. The results of this experiment are shown in Fig. 5. A PCR product of the size associated with KCNAl was detected in all tissues except for colon. In addition to the expected PCR fragment, this figure shows numerous additional PCR products, these products are presumably conserved K+ sequences, which are also targets for the primer pair. Thus KCNAl appears to be expressed in brain, heart, and placenta. Identification of a restriction fragment kngthpolymorphism associated with KCNAl. A two-allele EcoRV restriction site polymorphism (RFLP) with allele sizes of 15 and 12 kb was identified through Southern blot screening of six unrelated individuals using HCKl as probe. Observed heterozygosity for this marker was calculated from the genotypes of 110 unrelated individuals. Allele 1 (15 kb) was present at a frequency of 0.69 and allele 2 (12 kb) was present at a frequency of 0.31.

FIG. 5. Expression of KCNAl. Total RNA from brain, cardiac, colon, and placental tissues were analyzed by PCR for presence of KCNAl as described under Materials and Methods. All tissues except colon show a product of the size associated with KCNAl suggesting that expression of this channel is not restricted to cardiac tissues. Lane 1, total human DNA; Lane 2, brain; Lane 3, cardiac; Lane 4, placenta; Lane 5, colon; Lane 6, KCNAl cosmid DNA.

Dongen et al., 1990). Figure 3 also shows comparisons of KCNAl with three additional human K+ channels. HKl and HK2 are cardiac K+ channels while HBK2 was isolated from brain. KCNAl is highly homologous to all three sequences. Again, the highest degree of sequence conservation is apparent in the putative transmembrane domains and while the greatest divergence, both in length and amino acid composition, is seen in the regions separating transmembrane domains and in the amino and carboxyl termini. KCNAl is closely related to HK2. They share nearly identical amino termini and transmembrane domains, but differ significantly in the carboxyl terminus and in the domain that separates the putative Sl and S2 transmembrane domains. The sequence divergence in the SlS2 loop is illustrated in Fig. 4. The SI-S2 loop of KCNAl contains several base changes and is 15 bp longer than the same region of HK2. As a result, the predicted amino acid sequence of these genes diverge dramatically in the middle of an extremely conserved region (see Fig. 3). The sequence conservation between HK2 and KCNAl is so significant, however, that it is possible that these two sequences represent polymorphic alleles of the same gene or represent recent duplication of the same channel gene. Isolation and genomic localization of the HK2 genomic clone should help resolve this question. of KCNAl expression. To examTissue distribution ine the tissue distribution of KCNAl, RNA samples from several different tissues were examined for presence or absence of a PCR product derived from the KCNAl coding sequence. PCR analysis was chosen over

To determine Chromosomal localization of KCNAl. the chromosomal location of KCNAl, HCKl was used to genotype LQT kindred 1532 (Keating et al., 1991). These genotypes were then compared to other markers that had been used in the analysis of K1532. The results of these studies are shown in Table 1. A maximum lod score of +2.72 at a recombination fraction of 0.05 was identified between KCNAl and the von Willebrand Factor VIII (FSVWF) locus, which has been mapped to distal 12p (Ginsburg et al., 1985). Within this disease family, when both parents were genotyped, five matings had at least 1 parent who was heterozygous for both markers. These matings had an average of 3.8 offspring. To confirm the tentative assignment of KCNAl to the short arm of chromosome 12, the chromosomal location was determined by two independent methods: analysis of a human/rodent somatic cell mapping panel and fluorescent in situ hybridization of a KCNAl cosmid to human metaphase chromosomes. PCR amplification was used to analyze a human-rodent somatic cell hybrid mapping panel for presence of KCNAl. To ensure that we did not amplify sequences from K+ channel genes at other loci, oligonucleotide primers were designed from nontranscribed, single-copy sequence derived from the KCNAl cosmid subclone. To test these primer sets, they were used in PCR experiments with human genomic DNA. After the amplification of a single product of the correct length was confirmed (data not shown), primers were used to screen the somatic cell panel. The results of these studies are shown in Table 2. KCNAl was local-

TABLE Pairwise

Linkage

Analysis

between

KCNAl

LOD

1

and Von Willebrands Recombination

KEATING

Factor

VIII

(FSVWF)

Locus

in Kindred

1532

Maximum

fraction

0.001

0.01

0.025

0.05

0.075

0.1

0.2

0.3

2

0

1.52

2.42

2.66

2.72

2.66

2.56

1.95

1.16

2.72

0.05

HUMAN

+

I

I

I

I

I

I

+

I

++

I

+

CARDIAC

+

I

++

1++1

+

I

+

I

+

I

++++

I

+++

I

I

I

I

I

I

I

I

I

I

I

I

++

I

I

I

I

I

I

I

I

I

I

+

+++

I

+

I

I

I

I

I

I

I

+

I

I

I

I

I

I

I

t

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

+

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

+

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

++

+

I

I

I

I

+

I

+++

I

+

I

t++++++

I

++++I+1

I+

+

I

I

I

+

I

I

I++ltl

I

++t

I

I

I

I

ItI+

++I+1

I

I

t

t

I

+++

I

I

I

I

I

+++

+++

I

+

+

I

I

++

I

I

tt+t

I

I

I

I

I

t

tt

I

t

Itttl

I

I

I

I

++

+

I

++t+

It

+I++++

++

t

I

I

++

I

I

+

I

I

I

+I

I

I

I

+t

++++I+1

I

1++1

t+++

++t++

I

I

+t

I

I

I

ttttt

t

+++

I

I

+++++

+t++++

I

I

I

I

t

I

If

+++i

I

I

++

t

I

I

++++tti

I

++

tt++++

++I

+t+

I

+

++++t++t+

+t

735

KCNAl

I

I

+++++

+++

CHANNEL

++

I+++++

I

+++

K+

If

I

I

tt++

I

I

+

It+++

I

+

I

t

+

t

I

t

t

I

+

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

736

CURRAN,

LANDES,

ized to human chromosome 12 with a concordance ratio of 100%. The chromosome 12 localization was confirmed and refined by annealing biotinylated KCNAl cosmid and a known chromosome 12 marker to human metaphase chromosome spreads. The chromosome 12 marker selected was cosHcol1 (Cheah et aZ., 1985), isolated from the COL2Al locus on chromosome 12q. Figure 6 shows that both KCNAl cosmid and cosHcol1 clearly hybridize to the same chromosome on opposite sides of the centromere. When hybridized separately (data not shown), each probe exhibits a single chromosomal origin with cosHcol1 mapping to proximal 12q and KCNAl cosmid to distal 12~. These experiments clearly demonstrate that KCNAl is located on the short arm of chromosome 12. DISCUSSION

We have identified a K+ channel gene that is expressed in the human heart. This gene is highly homologous to K+ channel genes isolated from other species. Areas of sequence homology and divergence will be useful for planning mutagenesis experiments designed to explore the functional significance of specific K+ chan-

AND KEATING

nel domains. For example, a consensus site of phosphoryiation by CAMP-dependent kinase A has been conserved in the human cardiac K+ channels described here but is absent from the human brain channel, HBKB. This site is approximately 40 amino acids downstream from the sixth transmembrane domain. As this domain is thought to be intracellular, differential phosphorylation could play a role in regulation of KCNAl. We have identified a potential site for N-linked glycosylation in KCNAl and other previously studied K+ channels. It is not known, however, if these sequences are functionally important. As glycosylation of calcium channels appears to change their functional properties, it is possible that this modification will also be important for K+ channels (Catterall, 1988). The amino acid sequences separating putative Sl and S2 transmembrane domains show remarkable variation, both in composition and in length. By contrast, the regions immediately flanking this variable domain are highly conserved. There are two possible explanations for this observation. First, this region may be important, possibly affecting channel regulation by altering interactions with the extracellular environment. As KCNAl and HK2 are highly conserved in all domains except this

FIG. 6. Fluorescence in situ hybridization of KCNAl cosmid. KCNAl cosmid and the known chromosome 12 marker, cosHcol1, were simultaneously hybridized to human metaphase chromosomes as described under Materials and Methods. Photographs were taken directly from the photomicroscope and are unenhanced. Both probes hybridized to the same pair of C group chromosomes. When used separately (data not shown), cosHcol1 hybridized specifically to proximal 12q and the KCNAl cosmid localized solely to the distal end of the short arm of chromosome 12. Fifty independent metaphase spreads were analyzed using the KCNAl cosmid as probe, 43 showed hybridization to both chromatids of each pair of chromosome 12s, 4 showed hybridization to both chromatids of a single chromosome 12, and 3 did not hybridize.

HUMAN

CARDIAC

K+

region, this hypothesis can be tested by generating chimerit channels for this region. Second, the variable region may have no functional significance and thus is not conserved. We have physically mapped KCNAl to chromosome 12p and identified a genetic marker associated with this gene. This is the first reported human K+ channel gene to be mapped. As this region of chromosome 12 has not been well characterized, the use of KCNAl as a polymorphic DNA marker will be important for general linkage studies. KCNAl will also be useful as a candidate gene for studies of inherited cardiovascular and neurologic disorders. ACKNOWLEDGMENTS We thank Donald Atkinson and Christine Dunn for technical assistance and Ed Meenan for oligonucleotide synthesis. We thank Mark Leppert, Richard Lifton, Bob Weiss, Louis Ptacek, and Steven Prescott for their experimental advice and critical reading of the manuscript. Steve Gerkin assisted with sequence analysis. This work was supported in part by the NIH, the American Heart Association, the American Cancer Society, the Howard Hughes Medical Institute, and a Syntex Scholars Award.

REFERENCES Catterall, W. A. (1988). Structure channels. Science 242: 50-61.

and function

of voltage-sensing

Cheah, K. S. E., Stoker, N. G., Griffin, J. R., Grosveld, Solomon, E. (1985). Identification and characterization man type II collagen gene (COLZAl). Proc. N&l. Acad.

ion

F. G., and of the hu-

Sci. USA 82:

2555-2559. Chomczynski, P., and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162: 156-159. Davidenko, J. M., Cohen, L., Goodrow, R., and Antzelevitch, C. (1989). Quinidine-induced action potential prolongation, early af&depolarizations, and triggered activity in canine Purkinje fibers.

Circulation 79: 674-686. Feinberg, A. P., and Vogelstein, B. (1983). A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132: 6-13. Fish, F. A., Prakash, C., and Roden, D. M. (1990). Suppression of repolarization-related arrhythmias in vitro by low-dose potassium channel activators. Circulation 82: 1362-1369. Ginsburg, D., Handin, R. I., Bonthron, D. T., Donlon, T. A., Bruns, G. A. P., Latt, S. A., and Orkin, S. H. (1985). Human von Willebrand Factor: Isolation of cDNA clones and chromosomal localization. Science 228: 1401-1406. Grupe, A., Schroter, K. H., Ruppersberg, J. P., Stocker, M., Drewes, T., Beckh, M., and Pongs, 0. (1990). Cloning and expression of a human voltage-gated potassium channel: A novel member of the RCK potassium channel family. EMBO J. 9: 1749-1756. Hille, B. (1984). “Ionic Channels of Excitable Membranes,” pp. 99116, Sinauer Associates, Sunderland, MA. Hoger, J. H., Walter, A. E., Vance, D., Yu, L., Lester, H. A., and Davidson, N. (1991). Modulation of a cloned mouse brain potassium channel. Neuron 6: 227-236. Hubbard, S., and Ivatt, R. (1981). Synthesis and processing of asparagine-linked oligosaccharides. Annu. Rev. Biochem. 50: 555-583. Keating M., Atkinson, D., Dunn C., Timothy K., Vincent G. M., and M. Leppert. (1991). Linkage of a cardiac arrhythmia, the long QT syndrome, and the Harvey ras-1 gene. Science 252: 704-706. Kozac, from

M. (1984). Compilation the translational start

Acids Res. 12: 857-872.

and analysis of sequences site in eukaryotic mRNAs.

upstream Nu&ic

CHANNEL

737

KCNAl

Krebs, E., and Beavo, of enzymes. Annu.

J. (1979).

Phosphorylation-dephosphorylation

Rev. Bhchem. 48: 923-959.

Lathrop G. M., and Lalouel J-M. (1984). Easy calculations of lod scores and genetic risks on small computers. Am.J. Hum. Genet.36: 460-465. Lichter, P., Cremer, T., Borden, J., Manuelidis, L., and Ward, D. C. (1988). Delineation of individual human chromosomes in metaphase and interphase cells by in situ suppression hybridization using recombinant DNA libraries. Hum. Genet. 80: 224-234. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982). “Molecular Cloning: A laboratory Manual,” pp. 1-541, Cold Spring Harbor Press, Cold Spring Harbor, NY. McCormack, T., Vega-Saenz de Miera, and Rudy, B. (1990). Molecular cloning of a member of a third class of shaker-family K+ channel genes in mammals. PFOC.Natl. Acad. Sci. USA 87: 5227-5231. Moss, A. J., and Schwartz, P. J. (1979). Sudden death and the idiopathic long QT syndrome (editorial) Am. J. Med. 66: 67. Papazian, D. M., Schwarz, T. L., Tempel, B. L., Jan, Y. N., and Jan, L. Y. (1987). Cloning of genomic and complementary DNA from Shaker, a putative potassium channel gene from Drosophila. Science 237: 749-753. Papazian, D. M., Timpe, L. C., Jan Y. N., and Jan, L. Y. (1991). Alteration of voltage-dependence of Shaker potassium channel by mutations in the S4 sequence. Nature 349: 305-310. Roberds, S. L., and Tamkun, M. M. (1991). Cloning and tissue-specific expression of five voltage-gated potassium channel cDNAs expressed in rat heart. PFOC.Natl. Acad. Sci. USA 88: 17981802. Romano, 659.

C. (1965).

Congenital

cardiac

Luncet 1: 65%

arrhythmia.

Saiki, R. K., Gelfand, D. H., Stoffel, S., Scharf, S. J., Higuchi, R., Horn, G. T., Mullis, K. B., and Erlich, H. A. (1988). Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239: 487-491. Sanger, F., Nicklen, S., and Coulson, with chain-termination inhibitors. 5463-5467.

A. R. (1977).

DNA

sequencing

PFOC.Natl. Acad. Sci. USA 74:

Southern, E. M. (1975). Detection of specific fragments separated by gel electrophoresis. 517.

sequences J. Mol.

among

DNA

Biol. 98: 503-

Swanson, R., Marshall, J., Williams, J. B., Boyle, M. B., Folander, K., Luneau, C. J., Antanavage, J., Oliva, C., Buhrow, S. A., Bennett, C., Stein, R. B., and Kaczmarek, L. K. (1990). Cloning and expression of cDNA and genomic clones encoding three delayed rectifier potassium channels in rat brain. Neuron 4: 929-939. Taggart, R. T., Mohandas, T. K., Shows, T. B., and Bell, G. I. (1985). Variable numbers of pepsinogen genes are located in the centromerit region of human chromosome 11 and determine the high frequency electrophoretic polymorphism. PFOC.Natl. Acad. Sci. USA 82: 6240-6244. Tamkun, M. M., Knoth, K. M., Walbridge, J. A., Kroemer, H., Roden, D. M., and Glover, D. M. (1991). Molecular cloning and characterization of two voltage-gated K+ channel cDNAs from human ventricle. FASEB J. 5: 331-337. Temple, B. L., Jan, Y. N., and Jan, L. Y. (1988). potassium channel gene from mouse brain.

Cloning

of a probable

Nature 332: 837-839.

VanDongen, M. J., Frech, G. C., Drewe, J. A., Joho, R. H., and Brown, A. M. (1990). Alteration and restoration of K+ channel function by deletions at the N- and C-termini. Neuron 5: 433-443. Warburton, D., Gersen, S., Yu, M.-T., Jackson, C., Handelin, B., and Houseman, D. (1990). Monochromosomal rodent-human hybrids from microcell fusion of human lymphoblastoid cells containing an inserted dominant selectable marker. Gerwmics 6: 358-366. Ward, 0. C. (1964). A new familial Irish Med. Assoc. 64: 103-106.

cardiac

syndrome

in children.

Yellen, G., Jurman, M. E., Abramson, T., and MacKinnon, R. (1991). Mutations affecting internal TEA blockade identify the probable pore-forming region of a K+ channel. Science 251: 939-941.

J.