Primary Structure and Functional Expression of the Human Cardiac Tetrodotoxin-Insensitive Voltage-Dependent Sodium Channel.

of 5

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
5 pages
0 downs
  Proc. Natl. Acad. Sci. USA Vol. 89, pp. 554-558, January 1992 Medical Sciences Primary structure and functional expression ofthe human cardiac tetrodotoxin-insensitive voltage-dependent sodium channel (complementary DNA/heart muscle/electrophysiology/antiarrhythmic) MARY E. GELLENS*, ALFRED L. GEORGE, JR.*t, LIQIONG CHENt, MOHAMED CHAHINEt, RICHARD HORNt, ROBERT L. BARCHIt§¶, AND ROLAND G. KALLENt§II Departments of*Medicine, tBiochemistry and Biophysics, ¶Neurology, and the §David Mahoney Institute of NeurologicalSciences, University of Pennsylvania, Philadelphia, PA 19104; and tDepartment of Neurosciences, Roche Institute of Molecular Biology, Nutley, NJ 07110 Communicated by EliotStellar, September 23, 1991 ABSTRACT The principal voltage-sensitive sodium chan- nel from human heart has been cloned, sequenced, and func- tionally expressed. The cDNA, designated hH1, encodes a 2016-amino acid protein that is homologous to other members of the sodium channel multigene family and bears >90 identity to the tetrodotoxin-insensitive sodium channel char- acteristic of rat heart and of immature and denervated ratskeletal muscle. Northern blot analysis demonstrates an =9.0- kilobase transcript expressed in human atrial and ventricular cardiac musclebut not in adult skeletal muscle, brain, myo-metrium, liver, or spleen. When expressed in Xenopus oocytes, hHl exhibits rapid activation and inactivation kinetics similarto nativecardiac sodium channels. The single channelconduc- tance of hHl to sodium ions is about twicethat of the homol-ogous rat channel and hHl is more resistant to block by tetrodotoxin (ICso = 5.7 pM). hHl is also resistant to Iu-cono- toxin but sensitive to block by therapeutic concentrations of lidocaine in a use-dependent manner. Voltage-dependent sodium channels (NaChs) form a multi- gene family, with six isoforms currently identified in the rat  1). Although structurally very similar, these isoforms can be distinguished by their kinetics  2), single-channel conduc- tance  3), toxin sensitivity (4, 5), and reactivity with specific immunoreagents  6). For example, a NaCh isoform ex- pressed in rat cardiac and in immature or denervated adultskeletal muscle differs dramatically from the NaCh in inner- vated skeletal muscle in its relativeinsensitivity to block by tetrodotoxin (TTX), saxitoxin, and ,u-conotoxin  4, 5, 7). Tetrodotoxin-insensitive (TTX-I) voltage-dependent NaChs are critical for the initial rapid upstroke of the cardiacactionpotential and are responsible for mostof the Na+ current that occurs in mammalian heart  8,9). In addition, this TTX-I NaCh isoform is the principal target of class I (cardiac) antiarrhythmic agents (10). Although NaCh block- ing agents of this type are widely used in the acute therapy of ventricular tachyarrhythmias, their precise molecular mech- anism of action remains unclear, in part due to thedifficulty of studying human cardiac NaChs in isolation. An invitro expressionsystem for studying isolated human cardiac NaChs would have great utility. However, previous attempts to functionally express cardiac TTX-I NaChs in oocytes using RNA from mammalian heart have metwith variable success (11-13). While functional expressionof the rat muscle TTX-I isoform (rSkM2) from its cloned cDNA has beenaccomplished (14), theapplicability of these findings to the homologous human cardiac NaCh is unknown. We report here thecloning, sequencing, and functional expression of the TTX-I NaCh from human heart, designated hHl.** MATERIALS AND METHODS Screening of a Human Cardiac cDNA Library. A size- selected [>1 kilobase (kb)] oligo-(dT) and random-primed adult human cardiac cDNA library constructed in AZAPII (Stratagene) was screened with cDNA probes derived from rSkM2 [nucleotides (nt) 1-4385 and 5424-7076; ref.15]. Hybridizations were performed at 420C for 18 h in 50% (vol/vol) formamide/5x SSPE/5X Denhardt's solution/ 0.1% SDS/salmon sperm DNA (0.15 mg/ml)/random- primed 32P-labeled probe (1.5 x 101 dpm/ml)]and filters were washed with 6x standard saline citrate (SSC)/0.1% SDS at 650C. (lx SSPE = 0.18 M NaCl/10 mM sodium phosphate, pH 7.4/1 mM EDTA.) Plaque-purified clones wererescued as pBluescript phagemids and sequenced as described  15). Northern BlotAnalysis. Human tissues wereobtained as frozen surgical pathology specimens or procured from ca- daveric transplant organ donors by the National DiseaseResearch Interchange (Philadelphia, PA). Total cellular RNA was isolated by the method of Chirgwin et al. (16). Samples (10 gg) were size-fractionated, electroblotted, and prehybrid- ized as described (15). A subtype-specific hHl antisense complementary RNA probe derived from clone C92, repre- senting 0.9 kb of the 3'-untranslated (UT) region (nt 7494- 8491) was transcribed in vitro fromBamHI-linearized tem- plate DNA andused in Northern blot hybridizations at 5 x 106 dpm/ml. After hybridization (550C, 18 h),blots were washed at a final stringency of O.lx SSC/0.1% SDS, 750C. Functional Expression in Xenopus Oocytes. A full-length hH1 construct was made in pBluescript by sequential ligation of S14 EcoRI-Sac II (nt +1 to +252), C75 Sac II-Kpn I (nt +253 to +4377), and C92 Kpn I-EcoRI (nt +4378 to +8491) fragments and the full-length hH1 insert moved into a mod- ified pSP64T vector (14). nt -151 to -8 of the 5'-UT region were deleted from the construct using exonuclease III and mung bean nuclease (14). A 901-base-pair (bp) Kpn I-Eag I fragment from clone C21 (nt 4378-5279) was exchanged forthe corresponding fragment in pSP64T-hHl to eliminate a cloning artifact in C92. Synthetic sense mRNA was tran- scribed in vitro from Spe I-linearized hH1 template as de- scribed (14). Stage V or VI oocyteswere microinjected with 30-50 ngof synthetic hH1 mRNA and studied after 3-10 days. Na+ currents were measured either by two-microelectrode volt- age clamp (14) or by patch clamp with outside-out patches  3). The bath solution contained 116 mM NaCl, 2 mM KCI, 1.8 mM CaCl2, 2 mM MgCI2, and 5 mM Hepes(pH 7.6). For Abbreviations: NaCh, sodium channel; TTX, tetrodotoxin; TTX-I, TTX-insensitive; UT, untranslated; ID, interdomain; I-V, current- voltage; nt, nucleotide(s). 'To whom reprint requests shouldbe addressed at: 233 Anatomy Chemistry, 36th & Hamilton Walk, Philadelphia, PA 19104-6059. **The sequence reported in this paperhas been deposited in the GenBank data base (accession no. M77235). 554The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked  advertisement in accordance with 18 U.S.C. §1734 solely to indicate thisfact.  Medical Sciences: Gellens et at. Proc. Natl. Acad. Sci. USA 89 (1992) 555 HH1I MAN F - -  LPGSFRTELAERAKAGSTESELPEARLLASKPLGPELGPELPYQTFWKKTFEFSATNALYVLS 11 5Sk M2MAU L - - LRTSRFRSAIKMEORGASERGQEARODOSKPLGPRLGPELPYYKFVNGTFRFSATNALYVLS 1 16 SkM1 MASSSLPHLVPPGPHCLRPFTPESLAAIEQRAVEEEAR---LORNKOMEIEEPERKPRSOLEAGKNLPL IYGDPPPEVI&IPLEDLDPYYSDKKTFIVLUKGKAIF&FSATPALYLLS 115 0 H H1 P F HPVRR AAVKILVHSL FN4LI1MC T I L TCV FMAQHO PPPWTKYVEYT FTA[IY T FEStVK ILARA FCLHAFT FL RrPWNWLDF SV I IM1AY T TEFVDLGNVSAL RTIFEVL RALKTISVISG 23 5 56kM 2 P FHPVR RAAVK I LVHS LFSM4L IMCT I L TNCVFMAA0HDPPPUWTCVVEYT FTA I OT FE SLVKI LAR GFCLHMAFTFLRDPWMULD FSVIVNAYTTEFVDLGNVSAL RT FRVLRALKT IZSVI SG236 Sk M1 PFSIVRRVAIKVL IHALFSMF IN TILTNCVFMTMSNPPSWSKHVEYTFTGIYTFESLIKILARiGFCIDDFTFLRDPLIMULDFSVITMAYVTEFVOLGUISALRTFRVLRALrTITVIPG 23 5 [Si 152- 153 1S4 0   0 H HI LKTIVGALIOSVKKLADVKVLTVFCLSVFALIGLOLFMGNLRHKCVRMFTALMGTN0GSVEAOGL--------- VWESD   ---------LYLSDPEHYLLKIISTSDVLLCG 327 S k M2 LKTIVGALIOSVKKLADHMAVLTVFCLSVFALIGLQLFM4GNLRHKCVRMFTELMGTMGSVE&DGL   VWMLD ---------VYLNDPAHTLLKNGTTDVLLCG 328 Sk M1 LKTIVGAIO KLDMLVCSFALVLL GLOCRPPNTTWGOWSDWGDTYNTNOSANTDEYNENYLGNALG 35 5 I55 0 KH1H NSSDAGTCPEGYHCLKAGENPDHGYTSFDSFAWAFLALFRLMTQDCWERLYOQTLRSAGK TMI FFNLVI FLOSFYLVNLI LAVVANAYEEONDATIAETEEKEKRFQEAMEMLKKEHEA 447 S k M2 NSSDAGTCPEGYRCLKAGENPDHGYTSFDSFAWAFLALFQLMTQDCWERLYQQTLRSAJGK[YMI FFNLVIFLGSFYLVNLILAVVANAkYEEQNQATIAETEEKEIKRFOEAMEMLKKEHEA 448 S kMl NSSDAGHCPEGYECIKAGRNPNYGYTSYDTFSWAFLALFRLMTQDYWENLFOLTLRAAGKTYMIFFVVI IFLGSFYL1HLILAVVAMAYAEQMEATLAEDOEKEEEFQQMLEKYKKHOEE 4 75 SB HK1 LT IRGVDTVSRSSLENSPLAPVNSHERRSKRRKRMSSGTEECGODRLPKSDSEDGPRAMNHLSLTRGLSRTSNKPRSSRGS1 FTFRRRDLGSEADFADDENSTARESESHHTSLLVPWPL 567 S k M2 IT IRGVOTVSRSSLEN4SPLAPVTNHERKSKR-KRLSSGTEDGGDORLPKSOSEDGPRALNOLSLTHGLSRTSMRPRSSRGSI FTFRRRDOGSEADFADDENSTAGESESHRTSLLVPWPL 56 7 SkM1 LE KAKEAAALESGEE  490 H HI RR TSAOGOPSPGITSAPGHAL HGKKNSTVDCNGVVSLLGAGDPEATSPGSHLLRPVMLEHAPPOT TTPSEEPGGPQMLITSOAPC VDGFEEPGARQRALSAVSVLTSALEEL EESRHKCPPCW 687SkM2 R HPSAQGQPGPGASAPGYVL NGKRNSTVDCP*GVVSLLGAGDAEATSPGSYLLRAPMVLDR PPDT TTPSEEPGGPOML1TPOAPCAQGFEEPGARQRALSAVSVLTSALEELEESHRKCPPCU 687SkM ----A000PTHNK---------DCNG--SLDASGEKGPPRPS   --------SADSAISDAMEELEEAHQKCPPWW 542 0 0 HH1 NRLAQRYLIWECCPLWMSIKQGVKLVVMOPFTDLTITNCIVLNTLFMALEHYNMTSEFEEHLOVGNLVFTGIFTAEMTFKI 1ALDPYYYFOQOWNIFDSIIVILSLKELGLSRMSNLSVL 807 5 k MZ NRFAOHYLIWECCPLUHSIKQKVKFVVMDPFADLTITMCIVLNTLFM4ALEHYNMTAEFEEMLOVGNJLVFTGIFTAEMTFKI IALDPYYYFQgGWNIFDSIIVILSLMELGLSRN4GNLSVL 8 07 SkMl VKCAHKVLIWNCCAPWVKFKHIIYLIVMDPFVDLGITICIVLNTLFMAMEHYPMTEHFDNVLSVGHLVFTGIF1'AEMVLKLIAMDPYEYFOOGWNIFDSFIVTLSLVELGLANVQGLSVL 662 MI~ 1S2 11S3 0 H H1 RSFRLLRtVFKLAKSWPTLMTLIKIIGNSVGALGNLTLVLAIIVFIFAVVQ.MOLFGKHYSELRO--SDSGLLPRWHMMDFFHAFLIIFRILCGEWIETNWDCMEVSGQSLCLLVFLLVMVI 92 5 5 k Z RS F RL LRV F KLAKSWPT LN4TL IK I IGNS VGALGM L TLVLAI I VF I FAVVGMOLFGKHY SE L R H RI$0SDSL LPRWHMMDF F HAFL II FR I LCGEWI ETNWDCM4E VSGOSL CLL VF L LVMVI1 92 7 S kMi1 RSFRLLRVFKLAKSWPTLNMLIKIlGNSVGALGNLTLVLAI IVFIFAVVGQOLFGKSYKECVCKIASDCNLPRWJHMNDFFHSFLIVFRILCGEWIETH4WDCM4EVAGOAMCLTVFLMVMVI 78 2 11IS4 11S5 H H l G LNVVLNI FLAL LL S SFSADHSLTAPOED REMN NLQL A LAR I ORG L RFVK RT TWOF CCGL LR HRP0KPAALAAQGO LPSC A T PTSP PPPE T EKVPPTRtKE TQ F EOGEQ PGOG T PGDPEPV 1 04 5 5kHZ GNLVVLNLFLALLLSSFSADNLTAPOEDGEMNNLOLALAR IQRGLRFVKRTTWOPCCGILRRRPKKPAALATHSOLPSCI TAPRSPPPPEVEKVPPARKETRFEEOKRPGOGTPGOSEPV 1047 SkMl GNLVVLNLFLALLLSSFSADSLAASOEDGEHMJMLOIAIGRIKWGIGFAK ---- -TFLLGLLRGKI LSPKEI ILSLGEPGGAGENAEESTPEDEKKEPPPEDKELKDNHILNHVGLTD0PRS 898 H Hl CVP IAVA ESD TDDE EDOE ENSLOGTEE ESSKOOE SOP VSGWPR GP PD SR T SOVSA TASSEAEASA SOAOWR OOWKA   E POAPGCGE TPED SCSE GSTADMT T AELLEQ IPDLGQDVKD 1 163 SkH2 CVPIAVAESDTEDOEEDEENSLGTEEESSKO-ESQVVSGGHEPYQEPRASOQVSETTSSEAGASTSQADWQQEOKT- -EPOAPGCGErPEDSYSEGSTADMTNTADLLEQIPDLGEDVKD 1164 SkMl SI-----ELDHLNF1NNPYLT1OVPIASEE ----------SDLEMPTEEETDAFSEPEDIKKPLOPLvDGNSSVC-------STADYKPPEEDPEEQAEENPEGEO 982 HHl PEDCFTEGCVRRCPCCAVDTTOAPGKVWWRLRKTCYHIVEH4SUFETFIIFMJLLSSGALAFEDIYLEERKTIKVLLEYADKMFTYVFVLEMLLKWVAYGFKKYFTNAkWCWLDFLIVDVSL 1283 0kM? PEDCFTEGCVRRCPCCMVDTTOSPGKVWWRLRKTCYR1VEHSWFETFIIFMILLSSGALAFEDIYLEERKTIKVLLEYADKMFTYVFVLEMLLKWVAYGFKKTFTNAWCWLDFLIVDVSL 1284SkM1 PEC~ECKCCYDSGGMWLRCKVHWEFVMLSGLFDYFORITLYDVTIIELKVYFVFNWWDLV I1 102 11151 111S2 11153   0 0 0 661 VSLVANTLGFAEHGPIKSLRTLRALRPLRALSRFEGHRVVVNALVGAIPSI HHVLLVCLI FWLI FSIMGVNLFAGKFGRCINQTEG0LPLNYT1VNNKSDCESLNLTGELYWTKVKVNFD 1403 0kM2 VSLVANTLGFAEMGPI KSLRTLRALRPLEZALSRFEGHRVVVNALVGAIPSIMNViLVCLIFWLIlFSIMGVNLFAOKFGRCI NQTEG0LPLWTTIVNNKSECESFNVTGELYWTKVKVNFD 1404 S kM1 SLIIANWLGYSEtLGP IKSLRTLRALRPLRALSRFEGHRVVVNALLGAIPSIMNVLLVCLI FWLIFSIM0VNLFAGKFYvCVNTTTSE -RED ISVVNNKSESESLMYTGQVRWMNVKVN*YD 1221 111S4 111S5- HHl NVGAGYIALLOVATFKGWMDIMYAAVOSRGYEEPOPWEYNLYMYIYEVIEI IFGSFfTIHLIEIVIIDNENOOKKKIGOODIEMTEEOKKYYNAI4KKIGSKKPQKPIPRPLNKYQGFIFD 1523 S k M2 NVGAGYLAIIOVATFKGWHDIHYAAV050GTEEDPOWEONIYMYIYFVVEJIEFOSEETLNFIEIVIIDENFNDKKKLIQODIFHTEEOKKYYNAI4KKLGSKKPOKPIPRPLNKYOGFIFD 1 52 4 SkM1 NVGLGYLSILLVATEKGWHDIHYAAVOSREKEE0PHYEVNLYHYLYFVIEI IEOSEETLNLIGIVIIOHFNDDKKKFGGKDIEHTEEDKKTYNAI4KKLGSKKPQKPIPRPQNKIOGMVYO 1341 11156 HH1I I VT KOADVT1M4FL I CLNHVTMM4VE TDOOSPEEK I NIIAKINL LEfvA IETQGE CI VK LAAIRHYTF TNSwNHiEDEFVVV ILS I VGTVLSD I IOQKY FESPIT LERV IRLAR IGR I R IIRGAKGI1 16430 kM?2 IVTKQAEDVTIHEFLICINHVTMHVETDOOQSPEKVN1IAKINLIEVAIFTGECIVKM~A&IR~IYTTTNSWNIFDFVVV1ISIVGTVISDI IQKYFFSPTLFRV1RLARIGRILRLIRGAKGI 1644SkMl FVTKQVEDISRMILZCINMVTHMVETD0000LKVDIIYHNHNVF IIEFTGECVLKMFAIRHTTETIGWNIEDEVVVILSIVGLALSDLIQKYFVSPTIFRVIRLARIGRVLRLIRGAKGI 1461 IVS1IVS2IVS3 0IVS4 661 RTLLFALIMMSLPALENIGLLLFLVMEIYSI FOHANEAYVKWEAGlDDMFNFO'IEANSHICLFQITTSAGWDOII0PILNTGPPYCDPTIPNSMGS-RGDCGSPAVGIIPETTYII ISELI 1762 0 kM 2 RTLLEAI*IHISLPALIEHIOIIIEVME1TSJ FGHANEAYVKWEAGIDDMENEOTFANSHICLEOITTSTOUDGIISPIINTGPPYCDPNIPNSNGS-RGNCGSPAVGILEETTYIIISFLI 1 76 3 SkMl RTLLEAIMHSIPALFNIGILLEILVMEIYSIEOHSNEAVVKKESGIDDMENEETFGISIICIEEITTSAGWDGLLNPILNSGPPDCDPTLENPGTHVRGDCGNPSIGICFFCSY IISFII 1581 11155 ~IVS ~ 66H1 VVNMYJAIIEF TETPSDFMY EFPAQIESLDAASPRAPOSIML1VGRICDLATRLEG1DLINEFA1 82 HHl NPSKI STEP TTTLRRKHEEVSAMVIQRAFRRHLLQRSLKHASFLFRQQAG-SGIOEEDAPEREGI IATVMSENE0RELGPPSSSSISSTSEPPSYDSVTRATSDNIOVRGSDYSHSEDL 2001 0kHZ NPSKI SYEPITTTLRRKHEEVSATVIQRAERRHILORSVKHASFLFRQQAGGSGLSDEDAPEREGL IATMMMNENSRRORPLSSSS1SSTSFPPSYDSVTRATSDNLPVRASDYSRSEDL 2003 Okml MPSKVSYEPI TTTLKRKOEEVCA1KIQRAYRRHLLORSVKOASYMYRHOQDG- -- NDDGAPEKEGLLANTMNKMYGHEKEGDGVQSQGEEEKASTEOAGETVEPE - -PTSSSDTALTPSP 1816 HHl ADEEPOPSRD------RESIV 2016 0kHZ ADFPPSPDRO------RESIV 2018 SKM1 PELPPSSSEEOOOTVRPGVKESLV 1840 FIG. 1. Amino acid sequence of hH1, deducedfrom its cloned cDNA, is compared to sequences of rSkM1 (13) and rSkM2 (15). Identical amino acidresidues are shaded. The suggested locations of the six transmembrane a-heliceswithin each domain are designated by horizontal lines. Potentialextracellular consensus glycosylation sites are indicated by circles(solidcircles indicate sites conserved in all known NaChs). patch clamp experiments, the vitelline membrane was re- contained150 mM NaCl, 2 mM KC1, 1.5 mM CaC12, 1 mM moved by hyperosmotic stripping (17). The bathing solution MgC12, 10 mM glucose, and 10 mM Hepes(pH 7.4). The pipet  556 Medical Sciences: Gellens et al. solution contained130 mM CsF, 10 mM CsCl, 5 mM EGTA, and 10 mM Cs-Hepes (pH 7.3). For single-channel record- ings, capacity transients were eliminated by averaging rec- ords without openings and subtracting this average from all records. Measurementswere made at 20-220C. RESULTS Isolation and Characterization of Human Heart NaCh cDNAs. A cDNA library from adult human cardiac muscle was screenedwith probes corresponding to6.5 kb of the rSkM2 cDNA. Ninety positive clones were identified (6 x 105 recombinants screened), and 16 were plaque-purified. Four overlapping clones weresequenced (S14, 3.6 kb; C75, 4.3 kb; C21, 1.5 kb; and C92, 4.5kb), and these clones collectively encompassed 8491 bp of the cDNA designated as hH1. The identity, orientation, andapproximate position of these cDNAs were defined by comparisonswith the nucleotide sequence of rSkM2. Designationof individual clones was as follows (nt 1 is defined as the first nucleotide in the most 5' clone): S14 was 3517bp, nt 1-632 and 762-3646(5'-UTof 150 bp and amino acids 1-161 and 205-1165); C75 was 4366 bp, nt 152-4518(amino acids 2-1456); C21was 1450 bp, nt 4257-5707 (amino acids 1369-1852); and C92 was 4530 bp, nt 3980-8491 (amino acids 1343-2016, and 3'-UT of 2293 bp). The initiation codonwas at position 151-153 and the termi- nation codon was at 6199-6201 in the complete hH1 se- quence. A 129-bp segment (nt 633-761; amino acids 162-204) that was deleted from S14and an insertion of 19 bp that introduced a premature termination codon in C92 (insertion follows nt 4962) werediscoyered andappeared to be cloning artifacts. These alterations were not present in the corre- sponding regions ofoverlapping clones C75 and C21. Full- length constructs containing one or both of thesedefects were nonfunctional in oocyte expression studies. The complete nucleotide sequence of hH1 consists of 150 bp of 5'-UT sequence,an open reading frame of 6048 bp, and a 3'-UT region of 2293 bp that contains neither a polyade- nylylation signal sequence nor a poly(dA) region. The pre- dicted initiation site of hH1 resembles the consensus se- quence for eukaryotic initiationsites only in the presence of a purine nucleotide (adenosine) at position -3 and a guano- sine at position +4 relative to the start codon. An out-of- frame ATG is present at relative positions -8 to -6 and this is a constituentfeature of all previously cloned NaChs. An extensive 3'-UT is present but bears no significantnucleotide sequence homology with other NaChs. Primary Structureof hH1. The primary structure of hH1 (Fig. 1) consists of2016 amino acids with a calculated molecularweightof 227,159. The sequence is comparable to other NaChs, with fourlarge (226-288 residues) homologous domains, each containing at least six potential membrane spanning a-helical segments including a positively charged amphipathicsegment (S4) (18). Comparisonsbetween hH1 and each of the cloned NaChs reveal a consistent pattern of primary structure homology within the repeat domains and the interdomain (ID) 3-4 region. ID 1-2 and ID 2-3 are much less conserved; only rSkM2 exhibits a high degree (>80%) of amino acid sequence identity with these regions of hHl. In addition, the ID 1-2 region is 296 residues long, more similar to brain NaCh isoforms than theadult skeletal muscle isoform (rSkMl) or eel electroplax NaCh. There are 14 potential sites for N-linked glycosylation in regions of hH1 predicted to be extracellular, all of which are conserved from rSkM2; 5 sites (residues 214,291,328, 1365, and 1380)are found in all cloned NaChs. Most of these sites are located within the S5-S6 interhelical regions of D1 and D3. Within predicted cytoplasmic domains, there are six consensus sites for cyclic nucleotide-dependent phosphory- lation (Ser-483, -571, and -593; Thr-17, -977, and -1026) althoughnone are universally present in other NaChs. Tissue Distribution of hH1. The steady-state levels of hH1 RNA transcripts in various adult human tissues wereexam- ined. An hHl-specific antisense cRNA probe hybridizes with an -9.0-kbtranscript present in total RNA isolated from right atrium and left ventricle but not in RNA isolated from human adult skeletal muscle, brain, myometrium, liver, or spleen (Fig. 2). These results are consistent with the tissue-specific expression of hH1 in adult cardiac muscle. Functional Expression of hHl. Xenopus oocytes were in- jected with synthetic mRNA generated from a full-length hH1 construct lacking the 5'-UT region (Figs.3-5). Typical Na' currents were observed 3-10 days later with either two- microelectrode or outside-out patch recording (Fig. 3A). The normalized peak current-voltage (I-V)relationship for six patches is shown in Fig. 3B. The maximum inward currents in these patches ranged between 89 and 1100 pA and acti- vated at potentials more positive than -60 mV (maximum at -10 mV). The absence of reversal of the current at voltages as high as +90 mV in these patches indicates that the channels are highly selective for Na' over Cs', with a selectivity ratio >34:1. Data describing the steady-state voltage dependence of inactivation were fit by a Boltzmann distribution with a midpoint of -61.9 + 0.3 mV and a slopefactor of 7.7 ± 0.3 mV. The kinetics of inactivation during a voltagepulseusually exhibited a single exponential with a time constant  rh) that decreased =e-fold/53 mV with depolarization. Inactivationkinetics were rapid and voltage-dependent, unlikethe currents from either rSkM1 (19) or rat brain IIA (20), which show abnormally slow inactivation in Xenopus oocytes. The hH1 currents were insensitive to TTX (Fig. 4A), with an IC50 value of 5.7 ± 0.8 AM. ,-Conotoxin (100 nM), which blocks TTX-sensitive NaChs from skeletal muscle  5, 19), does not block the Na+ current of hH1 when expressed in oocytes. Lidocaine (10 AM) produces a frequency-dependent block (Fig. 4B). In five oocytes,10 ,uM lidocaine blocked 2 ± 2%, 21 ± 3%, and 37 ± 3 of the peak current at 0.5, 5, and 10 Hz, respectively. At 100 ,uM lidocaine, the block was 29 ± 7%, 54 ± 7%, and 70 ± 7 for frequencies of 0.5, 5, and 10 Hz (n = 3). Single-channel currents were obtained for hH1 from out- side-out patches (Fig. 5). The kinetics of activation and inactivation are comparable to those seen in patches from mammalian heart (21, 22). The first latencies to openings and the probability of late openings decreasewith depolarization; the openings also may occur in bursts. A plot of the ampli- tudes of the single-channel currents as a function of voltage shows inward rectification. Between -20 and +10 mV, where the I-V relationship is relatively linear,the slope conductance is 22 pS. This value is intermediate between the conductance of rSkM1 (32 pS) and rSkM2 (10 pS) measured under identical conditions(Fig. 5). b 1 b 9.5 7.5   4 4 2.4   FIG. 2. Northern blot analy- sis of the tissue distribution of hH1 transcripts. Locations of RNA size standards  in kb) elec- trophoresed on the same gel are indicated to the left. The size of the hH1 transcript in atrium and ventricle is -9.0 kb. Proc. Natl. Acad. Sci. USA 89 (1992)  Proc. Natl. Acad. Sci. USA 89 (1992)557 A a:   -4 C 1.0 0.I 0.6 IA 5 ms -120-100 -80 -60 -40 -20 Voltage (mV) B D 0 .1 e id 0 -. 0 DISCUSSION At leastsixdistinct isoforms of the voltage-dependent NaCh a subunit exist inrat brain, skeletal muscle, and heart. Based upon previouslypublished observations (23, 24), it appears likely that a similar array of NaCh subtypes exists in the human genome. We characterize here a member of the human NaCh a-subunit multigene family, hH1. The most striking aspect of the hH1 channel primary structure is its high level of similarity with rSkM2 (15) and the rat cardiac NaCh RHi that appears to be identicalto rSkM2 (25). The degree of primary structure identity that exists between humanhH1 and rSkM2 (93.8 ) is significantlygreater than that between the two rat isoforms rSkM1 and rSkM2 (59 ), which are coexpressed in skeletal muscle. Thesecomparisons are particularly striking in the ID 1-2 and ID 2-3 regions whereamino acid sequence identity between rSkM1 and rSkM2 is much lower (14 and 23%) than that between human hH1 and rSkM2 (88% for ID 1-2 and 84 for ID 2-3). Thisobservation suggests thatstructural differences between NaCh isoforms havebeenconserved during evolu- tion and are likely to beimportant physiologically. The S5-S6 interhelical region in D1 has been postulated to contribute to the binding site for TTX. A comparison of the primary structure of this segment in TTX-I and TTX- sensitive NaChs is particularly interestingsince a marked decrease in the affinity for TTX has been reported in rat brain II NaChs subjected to site-specific mutation at position 387 (Glu -- Gln) inthis region (26). Although theneutralization of negative charge producedby this mutation might interfere with binding of the cationic neurotoxins TTX and saxitoxin, a glutamate corresponding to Glu-387 is conserved in all known NaCh sequences including hH1 (Glu-375) and, there- fore, cannot be a determinantof toxin binding. However, the adjacentasparagine residue that is present in all TTX- sensitive NaCh isoforms is replaced by arginine in both hHl(Arg-376) andrSkM2. This charge difference could con- voltage2 (m0) -20 20 40 6080 FIG. 3. Activation and inacti- T.   kitvation of Na' currents of hH1 expressed i oocytes. (A) A family   \ 6 2 r I; of Na' currents in a large outside- T/I   . ;i ,, out patch. The holding potential T.t s F # was -120 mV, and capacity tran- 4, / a U sients were removed by a P/8 pro- Ir   ; X cedure from the same holding po- tential. The currents were elicited by 26-ms pulses from -70 to +70 mV in 10-mV increments. (B) The 5 normalized peak I-V relationship 4 (mean + SD) for six patches. (C) T The steady-state inactivation 3 A\ h curve forthe data (mean + SD) obtained with large outside-outpatches from seven oocytes. The 2 test potential was -30 mV, and the 45-ms prepulses ranged from 1 1 9>-fold/53 mV -120 to 0 mV. The theoretical curve is anonlinear least squares fit of the data by the formula   I/Imax = 1/{1+exp[(V-Vo.5)/kv]}, 0.9 where Imax is the current mea- 0.8 sured from -120 mV. (D) Voltage 0.7 dependence of Th, measured by 0.6 fitting the decay of the current 0.5 L I after the peak witha single expo- nential. Data (mean ± SEM) were V (mY) analyzed from six patches.tribute to the decreased affinity for TTX exhibited by these channels. Four additional residues in this region are con-served in the TTX-I isoforms but differ in charge from the consensus of the TTX-sensitive NaCh sequences; these include two additionalpositive residues (Lys-317 and Arg- 340) andtwo negative residues (Glu-346 and Asp-349). In spite of the structural and electrophysiological similar- ities between hH1 and rSkM2, there are two distinct func- tional differences.  i) hH1 is less sensitive to TTX block; with an apparent affinity (IC50 = 5.7 tM) almost 3 times lower than that found for rSkM2 (14).  ii) The amplitudes of the single- channel currents of hH1 are different thanthose found in rSkM2. The hH1 single-channel I-V curve sh- svs inward rectification witha conductance near 0 mV of -22 pS (Fig. 5), whereas equivalent measurements in rSkM2 (Fig. 5; ref. 27) and in developing rat skeletal muscle (3) yield a linear I-V relation with a slope =10 pS. The simplest hypothesis forthedifference between the two types of TTX-I channels is that the hH1 channel has more negative charge near the extra-cellular mouth of the porethan rSkM2. This charge could increase thelocal concentrationof Na' (thus increasing the conductance) and of Ca2+, known to block open Na' chan- nels in a voltage-dependent manner (28). Calcium ions can produce the curvature in the I-V trace at negative potentials, where the block is enhanced due to the higher probability of Ca2' residing in a blocking site in the pore Four negatively charged residues in hH1 that replace neutral amino acids at corresponding positions of rSkM2 and that may lie at the extracellular mouth of the channel are located in the S5-S6 interhelical regions of domains 1 (Glu-302 and Glu-312), 2 (Asp-870), and 4 (Asp-1741). The inactivation kinetics of expressed hH1 currentsare rapid and closely resemble sodium currents observed in intact cardiac tissue. In contrast, functional expressionofother cloned NaChs (rSkM1 and rat brainIIA) in Xenopus oocytes results in abnormally slow inactivation (19,20). The Medical Sciences: Gellens et al.
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks