Multiple epithelial Na channel domains participate in subunit assembly - 《American Journal of Physiology》美国生理学杂志干细胞之家



免疫细胞治疗专区	欢迎关注干细胞微信公众号

12 3 4 5 6 7 8 9 10 ... 24 下一页

返回列表

查看: 549006\|回复: 233	go Multiple epithelial Na channel domains participate in subunit assembly [复制链接]

轻羽

新手上路

Rank: 1

积分: 0
威望: 0
包包: 0

楼主

发表于 2009-4-21 13:47 |只看该作者 |倒序浏览 |打印

作者：James B. Bruns, Baofeng Hu, Yoon J. Ahn, Shaohu Sheng, Rebecca P. Hughey,  Thomas R. Kleyman,作者单位：1 Departments of Medicine and Cell Biology and Physiology, University ofPittsburgh, Pittsburgh 15261; and Department ofMedicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104
   % G0 g- l9 P- s6 l( B+ Y


                     ' D8 d/ n& v, R5 w7 E* g* U6 n: R




                  ! q: |% t, J' z9 M- H* `
      , B. Y5 G8 U  Z
      5 B+ b) h) K% S4 M; h
      7 d3 U& c! L* J, D. i$ i
      【摘要】
   Epithelial sodium channels (ENaCs) are composed of three structurally related subunits that form a tetrameric channel. The Xenopus laevis oocyte expression system was used to identify regions within the ENaC -subunit that confer a dominant negative phenotype on functionalexpression of -ENaC to define domains that have a role insubunit-subunit interactions. Coexpression of full-length mouse -ENaC with either 1 ) the -subunit firstmembrane-spanning domain and short downstream hydrophobic domain( -M1H1); 2 ) -M1H1 and its downstream hydrophilicextracellular loop ( -M1H1-ECL); 3 ) the membrane-spanningdomain of a control type 2 transmembrane protein (glutamyl transpeptidase; -GT) fused to the -ECL ( -GT- -ECL); 4 ) theextracellular domain of a control type 1 transmembrane protein (Tac) fused tothe -subunit second membrane-spanning domain and short upstreamhydrophobic domain (Tac- -H2M2); or 5 ) the -subunitcytoplasmic COOH terminus ( -Ct) significantly reducedamiloride-sensitive Na currents in X. laevis oocytes. Functional expression of Na channels was not inhibited when full-length -ENaC was coexpressed with either 1 )the -ECL lacking a signal-anchor sequence, 2 ) -M1H1 and -Ct expressed as a fusion protein, 3 ) full-length -GT,or 4 ) full-length Tac. Furthermore, the expression of ROMK channelswas not inhibited when full-length ROMK was coexpressed with either -M1H1-ECL or -Ct. Full-length FLAG-tagged -, -, or -ENaC coimmunoprecipitated with myc-tagged -M1H1-ECL, whereaswild-type -GT did not. These data suggest that multiple sites withinthe -subunit participate in subunit-subunit interactions that arerequired for proper assembly of the heterooligomeric ENaC complex. 2 }/ ~  x5 S; H+ d$ L
      【关键词】 amiloridesensitive sodium channel dominant negative mutants
   THE EPITHELIAL NA CHANNEL (ENaC) has a key role in theregulation of urinary Na reabsorption, extracellular fluid volumehomeostasis, control of blood pressure, and airway fluid levels( 14, 25, 57, 63 ). ENaCs are composed ofthree homologous subunits, termed,, and, that assembleinto a tetrameric structure with a subunit stoichiometry of2 :1 :1 ( 16, 34 ), although an alternativesubunit stoichiometry has been proposed( 54 ). Mutations leading toclinical disorders have been identified in all three subunits.Gain-of-function mutations in the - and -subunits have beenidentified in patients with Liddle's syndrome, a disorder characterized byvolume expansion and hypertension( 22, 26, 51, 59, 67 ), whereas ENaCloss-of-function mutations have been identified in patients with type Ipseudohypoaldosteronism, a disorder characterized by volume depletion,hypotension, and hyperkalemia( 10, 58 )., n7 @/ t* z; |8 J9 q2 ?  z

Mouse ENaC subunits are composed of between 638 and 699 amino acids andshare a common topology ( 3 ).Each subunit has two membrane-spanning domains (M1 and M2) and adjacentextracellular hydrophobic domains (H1 and H2) separated by a largeextracellular loop (ECL) ( 8 ),whereas the NH 2 and COOH termini of each subunit are intracellular( 7, 43, 55 ). The combined application of site-directed mutagenesis and oocyte expression has yielded importantfindings regarding the structure/function relationships of ENaC( 5, 32, 44, 52 ). Two amiloride-bindingsites have been identified within the ENaC subunits, one within the pore region of each subunit and a second within the ECL of the -subunit ( 27, 33, 46 ). The main selectivityfilter has been localized to a three-residue tract [(G/S) X S] withinthe H2 region (or equivalently termed as pre-M2 region)( 29 - 31, 47, 56 ). Mutations that alter ENaCgating have been found in the NH 2 termini( 20, 21 ), ECL( 50 ), pore regions( 48, 53 ) and nearby regions( 18 ), M2 domains( 19 ), and COOH termini( 12 ).
; I& r4 S! l' ]8 N
Limited information is available regarding the identification of domainswithin the heterooligomeric Na channel complex that participate insubunit-subunit interactions and promote subunit oligomerization.Na channel subunits are thought to assemble in the endoplasmicreticulum (ER) where they undergo N-linked glycosylation( 2, 7, 11, 13, 23, 40, 41, 45, 55, 65 ). The interactions of newlysynthesized Na channel subunits within the ER must be of asufficient affinity to promote Na channel assembly. A previousstudy by Adams and co-workers( 2 ) showed that theNH 2 -terminal domain of -ENaC interacted with and preventedfunctional expression of full-length -ENaC. Similarly, ourprevious work indicated that the NH 2 -terminal domain of -ENaC interacted with full-length -ENaC andprevented the expression of amiloride-sensitive currents in Xenopuslaevis oocytes ( 3 ),whereas results published by Chalfant and co-workers ( 9 ) showed that NH 2 terminally deleted -, -, or -ENaC prevented functionalexpression of full-length -ENaC. We have extended theseobservations and have identified multiple domains throughout -ENaCthat, when coexpressed with full-length -ENaC, confer adominant negative phenotype. These domains may have an important role infacilitating subunit-subunit interactions that allow for proper assembly andfunctional expression of the heterooligomeric Na channel.

MATERIALS AND METHODS
% G) a" J3 }6 u, S( N) n
Materials. All chemicals were purchased from Sigma (St. Louis, MO)or as specified otherwise.

Mouse ENaC constructs. Mouse (m) -, -, and -ENaC cDNAs were previously cloned into pBluescript SK(-)(Stratagene, La Jolla, CA) ( 3 ).The following -mENaC domains were generated by PCR: -M1H1-ECLcontaining S102-S568; -M1H1 containing S102-R164; -M1H1-cytoplasmic COOH terminus (Ct) containing S102-R164 fused toH613-L699; -Ct containing H613-L699; -glutamyl transpeptidase (GT)/ECL containing the short cytoplasmic tail and transmembrane domain of rat -GT (M1-T28) ( 28 ) fusedto the -subunit's extracellular loop (Y165-S568); and Tac/M2H2containing the extracellular domain of the human interlukin-2 receptor -subunit (Tac, M1-Q240) ( 39 ) fused to the -subunit's second hydrophobic and transmembrane domains (V569-R618).COOH terminally truncated -ENaCs were generated by placingstop codons at H613, R564, and R583 of -, - and -,respectively. The FLAG epitope was added to the COOH terminus of the -, - and -cDNAs, the V5 epitope was added to the COOH terminus ofthe -subunit and -Ct cDNA, and the myc epitope was added to theCOOH termini of -M1H1-ECL and -GT-ECL cDNAs. All epitope-taggedconstructs were cloned into pcDNA3.1 (Invitrogen, Carlsbad, CA). Allconstructs were confirmed by automated DNA sequencing at sequencing facilitiesat the University of Pennsylvania or at the University of Pittsburgh. cRNAswere synthesized from the above constructs in linearized pBlue-script orpcDNA3.1 using the T3 or T7 mMESSAGE mMACHINE kit (Ambion, Austin, TX). cRNAswere stored at -80°C and diluted in nuclease-free water beforeinjection.

Oocyte preparation and injection. Oocytes were obtained from adultfemale X. laevis using protocols approved by Institutional AnimalCare and Use Committees at the University of Pennsylvania and the Universityof Pittsburgh. Stage V and VI X. laevis oocytes were enzymaticallydefolliculated in 2 mg/ml type IV collagenase and then maintained at 18°Cin modified Barth's saline [MBS; (in mM) 88 NaCl, 1 KCl, 2.4NaHCO 3, 15 HEPES, 0.3 Ca(NO 3 ) 2, 0.41CaCl 2, and 0.82 MgSO 4, as well as 10 µg/ml sodiumpenicillin, 10 µg/ml streptomycin sulfate, and 100 µg/ml gentamicinsulfate, pH 7.4]. Oocytes were microinjected with mouse -, -, and -ENaC cRNAs (2 ng/subunit) with or without test cRNAs (20 ng, unlessotherwise indicated) in 50 nl nuclease-free H 2 O. Oocytes were alsocoinjected with H613X, R564X, and R583X cRNAs (0.5ng/subunit) with or without -Ct cRNA (5 ng). Whole cell currents weremeasured 1-2 days after cRNA injections.6 M' T: {! s7 j+ i  W% B

Whole cell current measurements. The two-electrode voltage-clamp technique was used to measure the amiloride-sensitive whole cell inwardcurrents with a DigiData 1200 interface (Axon Instruments, Foster City, CA)and a TEV 200A Voltage Clamp Amplifier (Dagan, Minneapolis, MN). Dataacquisition and analyses were performed using pClamp 7.0 software (AxonInstruments) with a Pentium II-based PC (Gateway 2000, N. Sioux City, SD).Recording pipettes were pulled from borosilicate glass capillaries (WorldPrecision Instruments, Sarasota, FL) and filled with 3 M KCl. Pipettes withtip resistances of 0.5-3 M in a 110 mM NaCl bath solution werechosen for experiments. Oocytes were bathed in a solution containing (in mM)96 NaCl, 2 KCl, 1.8 CaCl 2, 1 MgCl 2, and 5 HEPES-NaOH, pH7.4. All measurements were carried out at room temperature(22-25°C), and the bath solution was continuously perfused at5-6 ml/min by gravity. Oocytes were typically incubated in the bathsolution for at least 5 min before the current was recorded to allow currentsto stabilize. Oocytes were clamped at -100 mV for 900 ms. Currents weremeasured at 600 ms after initiation of the clamp potential. Betweenrecordings, the circuit was opened, allowing the oocytes to rest at thereversal potential. To minimize the contribution of variability in expressionlevels from different batches of oocytes, we measured the whole cell currentsin oocytes coinjected with -mENaC cRNAs and in oocytescoinjected with -mENaC cRNAs plus an -subunitconstruct cRNA, using oocytes obtained from a single batch. Oocytes from thetwo groups were clamped in an alternating manner so that the contribution of expression time to the measured currents was similar for each group.Measurements from repeated experiments, using oocytes from at least two frogs,were pooled to obtain a sufficient number of observations for statisticalanalysis.
  }3 w+ [( j8 T8 L2 X: X* q, M
Transient expression in Chinese hamster ovary cells. Wild-type Chinese hamster ovary (CHO) cells were cultured as previously described( 4 ). Transient expression ofmENaC subunits, dominant negative constructs, and/or -GT was carriedout in CHO cells by infection with recombinant vaccinia (vT7CP) expressing T7 RNA polymerase, followed by transfection with expression vectors containing T7promoters at the 5'-end of the cDNAs as previously described( 42, 64 ). The full-length -GT used contained a substitution of asparagine for threonine atposition 380. This mutation prevents the autocatalytic cleavage of -GTand eliminates enzymatic activity without affecting processing or trafficking (Hughey R, unpublished observations). vT7CP was a kind gift from Bernard Moss(National Institute of Allergy and Infectious Disease, Bethesda, MD).Transfections were carried out 30 min postinfection with lipofectamine reagentas described by the manufacturer (Invitrogen, Carlsbad, CA). After overnightincubation, 1 x 10 6 cells were solubilized at roomtemperature in 300 µl of a detergent solution (50 mM Tris · HCl, 1%Nonidet P-40, 0.4% deoxycholate, and 62.5 mM EDTA, pH 8.5) supplemented with1% Protease Inhibitor Cocktail Set III (Calbiochem, La Jolla, CA). Insolublematerial was removed by centrifugation in a microcentrifuge at 20,000 g for 7 min at 4°C. Supernatants were recovered and incubatedovernight at 4°C after the addition of protein G-Sepharose (Sigma) andmouse monoclonal antibodies against either Flag (Anti-Flag M2, Sigma) or Myc (9E10, Santa Cruz Biotechnology, Santa Cruz, CA). Immunoprecipitates wererecovered by brief centrifugation, and each was washed once with 1 ml each of1% Triton X-100 in HEPES-buffered saline (HBS; 10 mM HEPES-NaOH, pH 7.4, 150mM NaCl), 0.01% SDS in HBS, and finally HBS. Where indicated,immunoprecipitates were treated with N -glycosidase (PNGase) F asdescribed by the manufacturer (New England BioLabs, Beverly, MA). Proteinswere recovered by heating for 3.5 min at 95°C in Laemmili SDS-samplebuffer containing fresh 0.14 M -mercaptoethanol. Samples were subject toSDS-PAGE (all reagents were from Bio-Rad, Richmond, CA) on 7.5% Criterion gelsand then electrophoretically transferred from the gel to Immobilon-NC filters(Millipore, Bedford, MA). The blot was blocked overnight at 4°C in PBS (8mM sodium phosphate, 2 mM potassium phosphate, 140 mM NaCl, 10 mM KCl, pH 7.4)containing 5% nonfat dry milk. Subsequently, the blot was incubated in PBScontaining 1% nonfat dry milk and the indicated mouse monoclonal antibody atroom temperature for 3 h. All blots were then washed extensively with PBSfollowed by a 45-min incubation with horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (KPL, Gaithersburg, MD) in PBS containing 1% nonfat drymilk. After an extensive washing in PBS, HRP bound to proteins was detectedusing Western Lightning Chemiluminescence Reagent Plus (PerkinElmer, Boston,MA) and BioMax MR film (Eastman Kodak, Rochester, NY) as directed by themanufacturer.

mENaC surface expression in X. laevis oocytes. Full-length -mENaC was coexpressed with or without -M1H1-ECL in X.laevis oocytes. The -subunit was tagged with the V5 epitopeat its COOH terminus. The vitellin membrane was manually removed with fineforceps and, after being washed in MBS, the oocytes were incubated twice for20 min with 1.5 mg/ml sulfo-NHS-SS-biotin (Pierce, Rockford, IL) in MBScontaining 10 mM triethanolamine, pH 9.0, on ice. Free biotin reagent wasquenched by four 5-min washes in MBS containing 5 mM glycine. After threewashes with MBS, 15 oocytes were solubilized on ice in 300 µl of a detergent solution {25 mM MES, pH 6.4, 200 mM NaCl, 1% Triton X-100, 60 mM n -octyl glucoside, 0.1% SDS, 0.5% Nonidet P-40, 0.02% sodiumdeoxycholate, 1% digitonin, 0.5% Tween 20, 0.02% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate, and 2 mM EmpigenBB}. Insoluble material was removed by centrifugation for 15 min at 15,000 g at 4°C. Supernatants were recovered and incubated overnight at4°C with immobilized streptavidin (Pierce). Streptavidin precipitates werethen washed and subjected to SDS-PAGE and immunoblot analyses as above.Anti-V5 antibody (Invitrogen) was used to detect the ENaC -subunit.

Statistical analysis. Data are expressed as means ± SE. Unpaired Student's t -tests were used to assess significance. P
$ G* n. q- h9 ~; k8 d6 @
RESULTS
0 k9 y- A8 s  _& v$ c& u: M
The X. laevis oocyte expression system was used to identify regions within the -subunit that confer a dominant negative phenotypeon -ENaC expression. -ENaC topology and the -ENaC constructs used for these studies are illustrated in Fig. 1. We previously observedthat coexpression of full-length mouse -ENaC with thecytoplasmic NH 2 terminus of the -subunit inhibited functional Na channel expression based on amiloride-sensitive currents in X. laevis oocytes( 3 ). Similarly, in this study we observed that coexpression of -ENaC with -M1H1also led to an inhibition of functional channel expression( Fig. 2 A ). Thisfunctional inhibition was dose dependant( Fig. 2 B ), suggesting that -M1H1 was competing with full-length subunits for binding sitesthat promote Na channel assembly. Although -M1H1 inhibited functional ENaC expression, expression of a fusion protein of -M1H1 and -M1H1-Ct did not inhibit functional ENaC expression ( Fig. 2 C ), suggestingthat the presence of the COOH terminus within the ER lumen may prevent orlimit interactions between -M1H1-Ct and full-length subunits.
3 {# Z6 F4 e9 ?- g$ i5 w+ @  {
Fig. 1. Topology of -subunit of the epithelial Na channel( -ENaC) and the -ENaC domains and chimeras tested for theability to confer a dominant negative phenotype. The topology of full-length -mouse (m) ENaC is illustrated ( top ), followed by the -ENaC domains and chimeras used in this study. M1 and M2, hydrophobic1st and 2nd membrane-spanning domains; H1 and H2, largely hydrophobic regionsimmediately following M1 and immediately preceding M2, respectively. The shortcytoplasmic and transmembrane domains of -glutamyltranspeptidase( -GT), a type II integral membrane protein, and the extracellulardomain of the IL-2 receptor -chain (Tac), a type I integral membraneprotein, were fused to the -subunit extracellular loop ( -ECL)and to -H2M2, respectively, to ensure that -ENaC domains wereproperly located in the lumen or membrane of the endoplasmic reticulum(ER).1 f2 ~6 n1 r7 i* I
' X9 K" K7 y$ c, O
Fig. 2. -M1H1 inhibits functional ENaC expression in a dose-dependentmanner. A : oocytes were coinjected with wild-type -mENaC cRNAs (2 ng/subunit) with or without a 10-foldexcess (20 ng) of -M1H1 cRNA. Whole cell Na currents( I Na ) were recorded 24-48 h after injection, usingthe 2-electrode, voltage-clamp technique at a clamp potential of -100 mVin the absence or presence of 100 µM amiloride. Amiloride-sensitiveNa currents were significantly lower in oocytes coinjected with -mENaC and -M1H1 ( n = 20). * P B : oocytes were coinjected with wild-type -mENaC cRNAs (2 ng/subunit) with increasing amounts of -M1H1 cRNA (0, 0.1, 0.3. and 1 ng). Whole cell amiloride-sensitiveNa currents were recorded as in A ( n =25-28). * P -M1H1  -ENaCvs. -ENaC alone. C : oocytes were coinjected withwild-type -mENaC cRNAs (2 ng/subunit) with or without a10-fold excess (20 ng) of -M1H1 fused to the -COOH terminus( -M1H1-Ct cRNA). Whole cell amiloride-sensitive Na currentswere recorded as in A [ n = 19; P = not significant(NS)].

We observed that coexpression of -ENaC with -M1H1-ECL also inhibited functional Na channel expression( Fig. 3 A ). Further experiments were carried out to determine whether -ECL, independently of -M1H1, had a role in conferring a dominant negative phenotype. Toensure proper orientation of the ECL within the ER lumen, we generated afusion protein composed of the membrane-spanning domain of a control type 2integral membrane protein, -GT, fused to -ECL ( -GT-ECL).Coexpression of -ENaC with -GT-ECL resulted in aninhibition of functional Na channel expression ( Fig. 3 B ). Incontrast, coexpression of -ENaC with either the -ECL (lacking a signal-anchor sequence) or full-length -GT did not inhibit functional channel expression( Fig. 3, C and D ).$ M; r3 [; ]; q( n3 A0 H4 A) ]2 Z

Fig. 3. -M1H1-ECL and -GT-ECL inhibit functional ENaC expression. A : oocytes were coinjected with wild-type -mENaCcRNAs (2 ng/subunit) with or without a 10-fold excess (20 ng) of -M1H1-ECL cRNA. Whole cell amiloride-sensitive Na currentswere recorded 24-48 h after injection at a clamp potential of -100mV ( n = 60). * P B : oocytes werecoinjected with wild-type -mENaC cRNAs (2 ng/subunit) withor without a 10-fold excess (20 ng) of -GT-ECL cRNA. Whole cellamiloride-sensitive Na currents were recorded as in A ( n = 49-53). * P C : oocytes werecoinjected with wild-type -mENaC cRNAs (2 ng/subunit) withor without a 10-fold excess (20 ng) of -ECL cRNA (lacking asignal-anchor sequence). Whole cell amiloride-sensitive Na currents were recorded as in A ( n = 22-25; P = NS). D : oocytes were coinjected with wild-type -mENaC cRNAs (2 ng/subunit) with or without a 10-foldexcess (20 ng) of -GT cRNA, a type II integral membrane protein. Wholecell amiloride-sensitive Na currents were recorded as in A ( n = 39; P = NS).
' i% Y! S& f# e$ A3 r0 A( S" C. f
-ECL has six consensus sites for N-linked glycosylation( 7 ). To confirm that the ECLwithin the -M1H1-ECL or -GT-ECL constructs was properlytransferred into the ER lumen, we transiently expressed myc epitope-tagged -M1H1-ECL or -GT-ECL in CHO cells. The expressed constructs wereimmunoprecipitated from cell lysates, and N-glycans were cleaved by treatmentwith PNGase F. Both -M1H1-ECL and -GT-ECL displayed a shift to alower apparent molecular weight after treatment with PNGase F( Fig. 4 ), confirming theaddition of N-glycans to these polypeptides and indicating that the ECLswithin these constructs were present within the ER lumen.( h# X" g% q) e
8 L. d) W9 |" P
Fig. 4. -M1H1 and the short cytoplasmic tail and transmembrane domain of -GT are functional signal-anchor sequences allowing for ERtranslocation of -ECL. myc-Tagged -M1H1-ECL or -GT-ECLcDNAs were transiently expressed in Chinese hamster ovary (CHO) cells asdescribed (see MATERIALS AND METHODS ). After overnight incubation,cells were solubilized and the expressed proteins were immunoprecipitated withan anti-myc antibody (9E10). Immunoprecipitates were treated with or without N -glycosidase (PNGase) F, followed by SDS-PAGE and immunoblotanalysis. Both constructs showed a significant shift in apparent molecularweight after PNGase F treatment.( i) Y  N; R' z- L6 J6 h

Previous studies have shown that residues within M2 and H2 form the pore ofthe channel( 29 - 31, 36, 37, 47 - 49 ). We examined whether -H2M2 conferred a dominant negative phenotype. TheNH 2 -terminal ectodomain of a control type 1 integral membrane protein, the human interleukin-2 receptor -subunit (commonly referredto as Tac), was fused to -H2M2 (Tac- -H2M2) to ensure properorientation of -H2M2 in the ER membrane. Coexpression of -ENaC with Tac- -H2M2 inhibited functionalNa channel expression ( Fig.5 A ), whereas full-length Tac did not confer a dominant negative phenotype ( Fig.5 B ).) {7 |* i3 O8 _

Fig. 5. Tac- -H2M2 inhibits functional ENaC expression. A : oocyteswere coinjected with wild-type -mENaC cRNAs (2 ng/subunit)with or without a 10-fold excess (20 ng) of Tac- -H2M2 cRNA. Whole cellamiloride-sensitive Na currents were recorded 24-48 h afterinjection at a clamp potential of -100 mV ( n = 24). * P B : oocytes were coinjected with wild-type -mENaC cRNAs (2 ng/subunit) with or without a 10-foldexcess (20 ng) of full-length Tac cRNA. Whole cell amiloride-sensitiveNa currents were recorded as in A ( n = 20; P = NS).+ x! G5 R; j' x% F9 s+ x

We also observed that coexpression of -Ct with full-length -ENaC inhibited the expression of functionalNa channels ( Fig. 6 A ). In contrast, coexpression of ENaC subunits lackingtheir COOH termini with -Ct did not alter functional Na channel expression ( Fig.6 B ), suggesting that direct interactions exist betweenENaC intracellular domains. As simultaneous overexpression of several proteinsmay overload the translation machinery of a cell, leading to a false dominantnegative phenotype, we performed similar experiments with an unrelatedchannel, ROMK. The expression of ROMK channels was not inhibited when full-length ROMK was coexpressed with either -Ct or -M1H1( Fig. 6C ), suggesting thatmultiple domains within the -subunit confer a dominant negativephenotype, presumably by specifically associating with full-length ENaCsubunits and preventing the assembly of tetrameric channels composed offull-length subunits.

Fig. 6. -COOH terminus inhibits functional expression of full-length, butnot COOH terminally truncated, -ENaC. A : oocyteswere coinjected with wild-type -mENaC cRNAs (2 ng/subunit)with or without a 10-fold excess (20 ng) of -COOH terminal (Ct) cRNA.Whole cell amiloride-sensitive Na currents were recorded24-48 h after injection at a clamp potential of -100 mV( n = 24). * P B : oocytes were coinjectedwith COOH terminally truncated -mENaC(ENaC C; -H613X, -R564X, -R583X) cRNAs(0.5 ng/subunit) with or without a 10-fold excess (5 ng) of -Ct cRNA( n = 20; P = NS). C : oocytes were coinjected withwild-type ROMK cRNA (2 ng) with or without a 10-fold excess (20 ng) of -M1H1 or -Ct cRNA. Whole cell K currents wererecorded 24-48 h after injection using the 2-electrode voltage-clamptechnique at a clamp potential of -100 mV in the absence or presence of5 mM BaCl 2. For these experiments, bath NaCl was replaced with KCl.Ba 2 -sensitive K currents were unchanged inoocytes coinjected with either -M1H1 or -Ct ( n =15-16; P = NS).5 F7 U, n' O/ J! W4 T
! z- P# r# j' z; e0 C/ W9 |
Experiments were performed to examine whether -M1H1-ECL binds tofull-length ENaC subunits with sufficient affinity to allow forcoimmunoprecipitation. -M1H1-ECL with a COOH-terminal myc tag wastransiently coexpressed with full-length, COOH-terminal FLAG-tagged -, -, or -mENaC, or full-length -GT (control) in CHO cells.Coimmunoprecipitation was observed when full-length ENaC subunits wereimmunoprecipitated, and subsequent immunoblots were probed for -M1H1-ECL ( Fig.7 A ). We also observed coimmunoprecipitation when -M1H1-ECL was immunoprecipitated, and subsequent immunoblots wereprobed for full-length ENaC subunits ( Fig.7 B ). -M1H1-ECL did not coimmunoprecipitate withfull-length -GT ( Fig.7 A ). We examined whether the dominant negative phenotypeobserved with coexpression of -M1H1-ECL was associated with aninhibition of surface expression of Na channels. Full-length -ENaC were coexpressed with or without -M1H1-ECL in X. laevis oocytes. The -subunit was tagged with the V5 epitopeat its COOH terminus. The vitellin membrane was removed, and surface proteinswere labeled by treatment with membrane-impermeant sulfo-NHS-SS-biotin. Oocytes were homogenized/solubilized, and biotinylated (i.e., cell surface)proteins were precipitated with streptavidin-agarose. Precipitated proteinswere subjected to SDS-PAGE and immunoblot analyses to detect the ENaC -subunit using anti-V5 antibody. Coexpression of -ENaC and -M1H1-ECL in X. laevis oocyteswas associated with a marked reduction in surface expression of -ENaC( Fig. 8 ). We also examinedwhole cell expression of selected dominant negative constructs in oocytescoinjected with -ENaC cRNAs and either -M1H1-ECL, -GT-ECL, or -Ct cRNA to confirm that these constructs wereexpressed in oocytes. Figure 9 illustrates that -M1H1-ECL, -GT-ECL, and -Ct were readilydetected by immunoprecipitation followed by immunoblotting.6 a' X( J$ D! Y  F7 p$ U
! o1 s) f( R6 P* J/ r
Fig. 7. -M1H1-ECL coimmunoprecipitates with -, -, and -ENaC, but not with -GT. A : myc-tagged -M1H1-ECLand individual COOH terminally FLAG-tagged ENaC subunits or -GT weretransiently coexpressed in CHO cells. ENaC or -GT wasimmunoprecipitated (IP) from cell lysates, and subsequent immunoblots (IB)were probed for -M1H1-ECL. Top : -M1H1-ECLcoimmunoprecipitated with -, -, and -ENaC, but not with -GT. Bottom : immunoblots of cell lysates verifying expressionof the transfected constructs. B : myc-tagged M1H1-ECL andindividual COOH terminally FLAG-tagged ENaC subunits were transientlycoexpressed in CHO cells. -M1H1-ECL was immunoprecipitated from celllysates, and subsequent immunoblots were probed for ENaC. -, -,and -ENaC coimmunoprecipitated with -M1H1-ECL.

Fig. 8. -M1H1-ECL inhibits surface expression of the -subunit whencoexpressed with -mENaC in Xenopus laevis oocytes. Oocytes were coinjected with -mENaC cRNAs (2ng/subunit) with or without -M1H1-ECL cRNA (20 ng) or with water alone.The -subunit was tagged with the V5 epitope at its COOH terminus. Aftera 36-h incubation, the vitellin membrane was removed and surface proteins werelabeled with sulfo-NHS-SS-biotin. Fifteen oocytes werehomogenized/solubilized, and biotinylated proteins were precipitated withstreptavidin-agarose. Precipitated proteins were subjected to SDS-PAGE andWestern blot analyses to detect the -subunit. This experiment wasperformed twice with similar results.
6 s9 m6 z$ ?6 B2 f/ {+ {( ^* l
Fig. 9. Whole cell expression of selected dominant negative constructs in X.laevis oocytes. Oocytes were coinjected with -ENaCcRNAs and either -M1H1-ECL, -GT-ECL, or -Ct cRNA. Thedominant negative constructs were epitope tagged at the COOH terminus witheither myc ( -M1H1-ECL and -GT-ECL) or V5 ( -Ct).Twentyfour hours after injection, -M1H1-ECL, -GT-ECL, or -Ct was readily detected by immunoprecipitation, followed byimmunoblotting with appropriate antibodies directed against myc or V5.
, |9 |1 f- s) B% o. {4 N
DISCUSSION. \) X: U; K; s( V0 ^. h2 v
  T2 J; a. {( o8 Y: F# f: w
The interactions of newly synthesized ENaC subunits within the ER must beof sufficient affinity to promote Na channel assembly. Asthousands of polypeptides are being synthesized on the ribosomes of a cell atany given time ( 68 ), it islikely that there are multiple sites of interaction between channel subunits that help to drive the assembly of this multimeric protein by increasing theaffinity of intersubunit associations. Previous studies have shown thatpolyvalent interactions may increase binding affinities by orders of magnitudecompared with monovalent interactions. For example, polyvalent immunoglobulinscan bind polyvalent antigens with "apparent" affinities 1,000-foldhigher than that of a monovalent immunoglobulin species( 24 ). This increase inapparent affinity was attributed to a greatly reduced rate of dissociation andnot to a change in the intrinsic affinity of the antibody.6 c) q2 f6 Y, r, ^; P. Y

Our work indicates that multiple -subunit domains, when coexpressed with full-length -ENaC, confer a dominant negativephenotype. These domains include 1 ) the cytoplasmic NH 2 terminus ( 3 ); 2 ) -M1H1; 3 ) -M1H1-ECL; 4 ) the ECL itself whenfused to a type 2 signal-anchor sequence to facilitate ER translocation; 5 ) -H2M2, when properly oriented in the ER membrane by fusion to a type 1 start transfer sequence; and 6 ) -Ct. Furthermore, the cellular localization of an ENaC domain (cytoplasm or ER lumen) was animportant factor in determining whether it conferred a dominant negativephenotype. For example, the ECL only conferred a dominant negative phenotypewhen translocated into the ER lumen by fusion to a signal-anchor sequence( -M1H1 or -GT). Similarly, the COOH terminus conferred adominant negative phenotype when expressed in the cytoplasm but did notinhibit functional expression when translocated into the ER lumen by fusion toa signal-anchor sequence ( -M1H1). The mislocalization of the COOHterminus within the -M1H1-Ct construct might hinder interactionsbetween -M1H1 with full-length ENaC subunits, which could prevent M1H1from inhibiting functional ENaC expression. Alternatively, exposure of thenormally cytoplasmic COOH terminus to the lumen of the ER may have induced therapid degradation of the -M1H1-Ct before -M1H1 could interactwith full-length subunits.1 e& @, L" R, T, [. e' x) t# |

Interactions of any of the dominant negative constructs with full-length -, -, or -ENaC could inhibit functional ENaC expression bycausing the sequestration of full-length subunits in the ER, by interferingwith the function of channels at the plasma membrane, or by inducing theER-associated degradation of full-length channel subunits associated with thedominant negative constructs in nonfunctional channel complexes. Dominantnegative-induced degradation of full-length channel subunits was observed in the suppression of -ENaC currents mediated by the overexpression oftruncated -subunits ( 2 )and also in the suppression of Kir 4.1 currents by the coexpression of membersof the Kir 3.0 family ( 61 ). Weobserved that the -M1H1-ECL dominant negative construct was associatedwith a marked reduction in surface expression of the full-length -subunit, when coexpressed with full-length - and -subunits, suggesting that overexpression of the -M1H1-ECLconstruct led to the sequestration and/or degradation of full-length channelsubunits in the ER; however, other possible explanations such as enhanceddegradation in post-ER compartments cannot be excluded.

These results, together with previous reports( 2, 3, 9 ), suggest the presence ofmultiple sites of intersubunit association within the heterooligomericNa channel complex and are in agreement with studies demonstratingthat other oligomeric ion channels have multiple sites of intersubunitinteractions. For example, Tu et al.( 60 ) used a dominant negativeapproach and demonstrated that multiple sites within the central core of Kv1.3were involved in intersubunit association. Similarly, multiple sites ofinteraction have been demonstrated in the homooligomerization of Kir 1.1achannels ( 35 ) and in theheterooligomerization of Kir3.1 and Kir3.2( 66 ). These domains includedthe NH 2 -terminal, core (i.e., M1 through M2), and COOH-terminaldomains.

The ECL of each ENaC subunit contains 16 conserved cysteine residues,suggesting a complex organization of the loop structures. Mutations of thefirst, sixth, eleventh, and twelfth cysteine residues have been shown to leadto reduced surface expression and channel activity of rat -ENaC in oocytes( 17 ). In addition, whole orpartial deletion of the -ECL resulted in no functional expression of -mENaC in oocytes (Sheng S and Kleyman TR, unpublished observations), and it was reported that even a six-residue deletion within the -ECL led to no detectable current or surface protein expression( 31 ). These observations areconsistent with an expectation that the ECL has a role in ENaC assembly.Similarly, it is expected that -M1H1 and -H2M2 would have a rolein subunit assembly, as the assembly of multiple transmembrane domains requires helix-helix interactions to ensure proper packing ( 1, 15, 62 ). However, the role of theCOOH termini in ENaC assembly is less clear, as truncation of the entire COOHterminus of the - or -subunit in fact increases channel activityand cell surface density ( 52 ).While the -Ct confers a dominant negative phenotype on functional ENaCexpression, it should be noted that a domain that confers a dominant negativephenotype may be involved in, but not required for, subunit assembly. It ispossible that the mechanism of inhibition of functional ENaC expression by -Ct may differ from that observed with the other constructs. Forexample, -Ct may confer a dominant negative phenotype by preventinginteractions with accessory proteins that facilitate functional channelexpression.) e4 V5 e; d$ T* X3 k1 ~+ M7 z

Truncations of ENaC subunits have been described in patients with autosomalrecessive pseudohypoaldosteronism type I( 6, 10 ). Bonny et al.( 6 ) reported that coexpressionof an -subunit truncated just before the pore region with full-length and allowed for functional ENaC expression in X.laevis oocytes. However, the onset of expression was delayed 48 h and themagnitude of the expressed current was markedly reduced with respect to wild-type -ENaC. The truncated -subunit was notexpressed with full-length -ENaC to test for a dominantnegative effect. It is tempting to speculate that heterozygotes expressingboth truncated and full-length -ENaC subunits may have reduced levels of Na channel expression due to a dominant negative effect onchannel assembly.
8 D1 e7 _/ V7 L. U! _
Transgenic overexpression of dominant negatives has become a useful toolfor the in vivo study of ion channel function. For example, London andcolleagues ( 38 ) have used theoverexpression of a Kv1.1 dominant negative construct to disrupt Shaker-like K channel function in the hearts of transgenic mice. Beyond providing insight into -ENaC domains that may be important in Na channel assembly, these data may prove useful for a dominant negative knockout of ENaC function in cell lines and animals.# T  i4 e0 j0 p- h+ O
) M: ]+ O! ?2 F3 R5 `
In summary, our results indicate that multiple domains throughout -ENaC, when coexpressed with -ENaC, confer adominant negative phenotype. We propose that these domains may play animportant role in facilitating and stabilizing subunit-subunit interactions that allow for the proper assembly and functional expression of theheterooligomeric ENaC.  {" V6 F8 {& L" a& ~4 i4 q
4 P; m, t  u$ D7 n
DISCLOSURES
0 M  a9 v$ f1 f3 I" G6 K0 y2 Y! u' c
This work was supported by National Institute of Diabetes and Digestive andKidney Diseases Grants DK-54354 (to T. R. Kleyman) and DK-54787 (to R. P.Hughey). B. Hu, Y. J. Ahn, and S. Sheng were recipients of postdoctoralfellowship awards from the Cystic Fibrosis Foundation.
      【参考文献】  R; p1 |2 B$ O% d) ]
Adamian L andLiang J. Helix-helix packing and interfacial pairwise interactions ofresidues in membrane proteins. J Mol Biol 311: 891-907,2001.
2 I+ E8 v9 F1 H$ H  k0 F
0 Z: A; v7 K2 V2 k

Adams CM,Snyder PM, and Welsh MJ. Interactions between subunits of the humanepithelial sodium channel. J Biol Chem 272: 27295-27300,1997.: P# c" U6 H  Z9 E9 H  v

, a; C# p# c( Z4 G

Ahn YJ, BrookerDR, Kosari F, Harte BJ, Li J, Mackler SA, and Kleyman TR. Cloning andfunctional expression of the mouse epithelial sodium channel. Am JPhysiol Renal Physiol 277:F121-F129, 1999.# n# q0 M2 ~9 y" o; D
' R4 C5 a' c% H3 I

  f2 y8 u7 d. `5 M' ^' @8 R$ b
Altschuler Y,Kinlough CL, Poland PA, Bruns JB, Apodaca G, Weisz OA, and Hughey RP. Clathrin-mediated endocytosis of MUC1 is modulated by its glycosylation state. Mol Biol Cell 11:819-831, 2000.
  h6 y$ q' R9 j# a# D7 b
( ]2 C8 h4 c5 R  ]. [* D+ f1 j

Benos DJ andStanton BA. Functional domains within the degenerin/epithelial sodiumchannel (Deg/ENaC) superfamily of ion channels. JPhysiol 520:631-644, 1999.0 O9 W* M$ e, [. k4 R$ `0 l  K$ L) A

, ~5 V) n# V, v% A# @! y  O- k
) E, A3 i1 A! z0 B8 ]* O& L
Bonny O, KnoersN, Monnens L, and Rossier BC. A novel mutation of the epithelialNa channel causes type 1 pseudohypoaldosteronism. Pediatr Nephrol 17:804-808, 2002.
6 X$ D$ C, ^* \1 z: ]

0 R! L; |' ]9 e
Canessa CM,Merillat AM, and Rossier BC. Membrane topology of the epithelial sodiumchannel in intact cells. Am J Physiol Cell Physiol 267: C1682-C1690,1994.

9 F, m- z3 F5 }! f5 q; o
Canessa CM,Schild L, Buell G, Thorens B, Gautschi I, Horisberger JD, and Rossier BC. Amiloride-sensitive epithelial Na channel is made of threehomologous subunits. Nature 367: 463-467,1994.1 I- O+ m$ a% r9 t2 [

Chalfant ML,Denton JS, Langloh AL, Karlson KH, Loffing J, Benos DJ, and Stanton BA. The NH 2 terminus of the epithelial sodium channel contains anendocytic motif. J Biol Chem 274: 32889-32896,1999.6 u- [' y: H' ~

( r. i1 X3 U/ ~- O# `% V
Chang SS,Grunder S, Hanukoglu A, Rosler A, Mathew PM, Hanukoglu I, Schild L, Lu Y,Shimkets RA, Nelson-Williams C, Rossier BC, and Lifton RP. Mutations insubunits of the epithelial sodium channel cause salt wasting withhyperkalaemic acidosis, pseudohypoaldosteronism type 1. NatGenet 12:248-253, 1996.

0 r) W# j- {+ i: u( e& D
' m( V8 p; {9 r& N
Cheng C, PrinceLS, Snyder PM, and Welsh MJ. Assembly of the epithelial Na channel evaluated using sucrose gradient sedimentation analysis. JBiol Chem 273:22693-22700, 1998.7 }* }; H" ^8 D- C" C9 s
: g$ g* k1 u3 t& ]2 L4 J
- O6 M- q" T7 B! H% b# H$ u4 E, ]

Copeland SJ,Berdiev BK, Ji HL, Lockhart J, Parker S, Fuller CM, and Benos DJ. Regionsin the carboxy terminus of -ENaC involved in gating andfunctional effects of actin. Am J Physiol Cell Physiol 281: C231-C240,2001.

De La Rosa DA,Li H, and Canessa CM. Effects of aldosterone on biosynthesis, traffic, andfunctional expression of epithelial sodium channels in A6 cells. JGen Physiol 119:427-442, 2002.& J1 R1 Q% W! F- U1 @, E) n6 U8 ^% O

/ Q" s3 r$ h" `2 m
$ {2 b! m$ E5 W* W3 r3 ?! ]
Eaton DC, MalikB, Saxena NC, Al-Khalili OK, and Yue G. Mechanisms of aldosterone's actionon epithelial Na transport. J Membr Biol 184: 313-319,2001.

+ A, T# `3 S( `) J" y  E1 \
5 q# V/ U) |$ G( J& l& X
Eilers M,Shekar SC, Shieh T, Smith SO, and Fleming PJ. Internal packing of helicalmembrane proteins. Proc Natl Acad Sci USA 97: 5796-5801,2000.( J; l: s* f: h4 A& @3 _- b  M  A4 s

  {" C2 }) ^5 o2 P7 b1 e
( h3 T) C2 O9 E
Firsov D,Gautschi I, Merillat AM, Rossier BC, and Schild L. The heterotetramericarchitecture of the epithelial sodium channel (ENaC). EMBOJ 17:344-352, 1998.
  S, F9 R2 b4 w( e) T
2 v7 g# C$ g2 w" O. d7 H. `! O" N
' a! S, j; x& O
Firsov D,Robert-Nicoud M, Gruender S, Schild L, and Rossier BC. Mutational analysisof cysteine-rich domains of the epithelium sodium channel (ENaC).Identification of cysteines essential for channel expression at the cellsurface. J Biol Chem 274:2743-2749, 1999.
, I4 L8 b- i3 X" d- j' ~% O

Fuller CM,Berdiev BK, Shlyonsky VG, Ismailov II, and Benos DJ. Point mutations in ENaC regulate channel gating, ion selectivity, and sensitivity toamiloride. Biophys J 72:1622-1632, 1997.

! r. u( ]8 A- B! t# w: H8 t- M
& Y/ @: I2 d( L# U6 Z! @
Fyfe GK, ZhangP, and Canessa CM. The second hydrophobic domain contributes to thekinetic properties of epithelial sodium channels. J BiolChem 274:36415-36421, 1999.

Grunder S,Firsov D, Chang SS, Jaeger NF, Gautschi I, Schild L, Lifton RP, and RossierBC. A mutation causing pseudohypoaldosteronism type 1 identifies aconserved glycine that is involved in the gating of the epithelial sodiumchannel. EMBO J 16:899-907, 1997.3 D+ c% ]+ f' F6 r3 G6 I; X% o+ [

  w+ \/ e( A  F  z- h. ~* z: \
9 c0 c) ]0 p1 w$ A2 o+ Q: P; Z
Grunder S,Jaeger NF, Gautschi I, Schild L, and Rossier BC. Identification of ahighly conserved sequence at the N-terminus of the epithelial Na channel alpha subunit involved in gating. PflügersArch 438:709-715, 1999., F2 J! I5 X! X1 n
& R) N1 J  @' u2 ?9 G
. @0 ]8 L& N7 u' t
% }" @! I$ R, z, ]
Hansson JH,Nelson-Williams C, Suzuki H, Schild L, Shimkets R, Lu Y, Canessa C, Iwasaki T,Rossier B, and Lifton RP. Hypertension caused by a truncated epithelialsodium channel subunit: genetic heterogeneity of Liddle syndrome. Nat Genet 11:76-82, 1995.7 C; v% ~# v( K1 Y4 G; T

+ M. C6 n6 v% Z% V7 n8 ]
Hanwell D,Ishikawa T, Saleki R, and Rotin D. Trafficking and cell surface stabilityof the epithelial Na channel expressed in epithelial Madin-Darbycanine kidney cells. J Biol Chem 277: 9772-9779,2002.
) v* `5 ]( G, J8 u/ k. t
  w) G; A2 }9 h/ H7 h
. m9 V% o1 E& {, X, x6 M9 y
Hornick CL andKaruch F. Antibody affinity 3. The role of multivalance. Immunochemistry 9:325-340, 1972.$ {+ c% g( u  g. L& ], A
/ ~7 M! ]! z+ b/ C% z5 m
# l5 D, V6 P8 y# t! a& Z5 a3 S; m$ R
+ d7 O% p+ s, T" i
Hummler E andHorisberger JD. Genetic disorders of membrane transport. V. The epithelialsodium channel and its implication in human diseases. Am J PhysiolGastrointest Liver Physiol 276:G567-G571, 1999.

" n- H8 E0 h& b5 a' H8 P
Inoue J, IwaokaT, Tokunaga H, Takamune K, Naomi S, Araki M, Takahama K, Yamaguchi K, andTomita K. A family with Liddle's syndrome caused by a new missensemutation in the beta subunit of the epithelial sodium channel. JClin Endocrinol Metab 83:2210-2213, 1998.
) ^6 w2 z5 r# s2 u% _2 u- Q  d

9 E% b1 `& ~/ d. e9 N
Ismailov II,Kieber-Emmons T, Lin C, Berdiev BK, Shlyonsky VG, Patton HK, Fuller CM,Worrell R, Zuckerman JB, Sun W, Eaton DC, Benos DJ, and Kleyman TR. Identification of an amiloride binding domain within the alpha-subunit of theepithelial Na channel. J Biol Chem 272: 21075-21083,1997.4 R2 w% T3 }. X; k9 ]* a$ D
6 t6 K  r7 q1 F; S0 r( E
% C; Z3 {8 x- U& F: t4 q, N$ m
! A6 Y! h$ P; d( r
Joyce-Brady M,Jean JC, and Hughey RP. -Glutamyltransferase and its isoformmediate an endoplasmic reticulum stress response. J BiolChem 276:9468-9477, 2001.: ?3 W" h3 Y% o( w! X8 ^( A

/ n7 W) E4 Y: n- l: S

Kellenberger S,Auberson M, Gautschi I, Schneeberger E, and Schild L. Permeabilityproperties of ENaC selectivity filter mutants. J GenPhysiol 118:679-692, 2001.
+ A+ H; c2 f' ~4 P- R7 \0 n
6 K5 E+ C  d; o4 J' K: t5 H, F* r# r) e
9 c  `8 o! Z/ X: h
Kellenberger S,Gautschi I, and Schild L. A single point mutation in the pore region ofthe epithelial Na channel changes ion selectivity by modifyingmolecular sieving. Proc Natl Acad Sci USA 96: 4170-4175,1999.
3 k0 l0 s+ \! w

Kellenberger S,Hoffmann-Pochon N, Gautschi I, Schneeberger E, and Schild L. On themolecular basis of ion permeation in the epithelial Na channel. J Gen Physiol 114:13-30, 1999.3 {$ [+ I/ ~8 C# y1 b* c! V

6 d1 X; J) d9 J( K
Kellenberger S and Schild L. Epithelial sodium channel/degenerin family of ion channels:a variety of functions for a shared structure. PhysiolRev 82:735-767, 2002.
" d/ h; H% `, N) z$ T2 S/ w

Kleyman TR,Sheng S, Kosari F, and Kieber-Emmons T. Mechanism of action of amiloride:a molecular prospective. Semin Nephrol 19: 524-532,1999.1 [; A- \  s" t1 A

Kosari F, ShengS, Li J, Mak DO, Foskett JK, and Kleyman TR. Subunit stoichiometry of theepithelial sodium channel. J Biol Chem 273: 13469-13474,1998.7 R& W7 A9 d# k4 c/ b, r
, r; v2 t8 T& i3 u" ^. u0 f4 p
9 C0 C. |3 b" t5 D9 B

Koster JC,Bentle KA, Nichols CG, and Ho K. Assembly of ROMK1 (Kir 1.1a) inwardrectifier K channel subunits involves multiple interaction sites. Biophys J 74:1821-1829, 1998.

Langloh AL,Berdiev B, Ji HL, Keyser K, Stanton BA, and Benos DJ. Charged residues inthe M2 region of -hENaC play a role in channel conductance. Am J Physiol Cell Physiol 278:C277-C291, 2000.. X" F9 l) e; x' M( V/ M% C  f

Li J, Sheng S,Perry CJ, and Kleyman TR. Asymmetric organization of the pore region ofthe epithelial sodium channel. J Biol Chem 278: 13867-13874,2003./ X& g) z$ D; c+ e4 p

9 u! v- d# D8 v8 r
London B, JeronA, Zhou J, Buckett P, Han X, Mitchell GF, and Koren G. Long QT andventricular arrhythmias in transgenic mice expressing the N terminus and firsttransmembrane segment of a voltage-gated potassium channel. ProcNatl Acad Sci USA 95:2926-2931, 1998.) T: y5 _/ W) X2 c, p7 v( Y* r
1 W# \2 S* w+ Q
, p- j: b9 Q( B  ~/ g5 _

Mallet WG andMaxfield FR. Chimeric forms of furin and TGN38 are transported with theplasma membrane in the transGolgi network via distinct endosomal pathways. J Cell Biol 146:345-359, 1999.: C# u/ z% _0 c& {: S  @& J

& A$ Y8 Q# y( A& u" a
& E5 T/ t# j% \0 P' `) f
May A, Puoti A,Gaeggeler HP, Horisberger JD, and Rossier BC. Early effect of aldosteroneon the rate of synthesis of the epithelial sodium channel alpha subunit in A6renal cells. J Am Soc Nephrol 8: 1813-1822,1997.

. e& V8 k, J  z  A* I2 K; U

Prince LS andWelsh MJ. Cell surface expression and biosynthesis of epithelialNa channels. Biochem J 336: 705-710,1998.
$ X3 I! W- a3 v9 H: D

$ L8 b2 h$ A& |) a& \
Ramsey-Ewing A and Moss B. Recombinant protein synthesis in Chinese hamster ovary cellsusing a vaccinia virus/bacteriophage T7 hybrid expression system. JBiol Chem 271:16962-16966, 1996.

, K4 }$ i3 A9 K3 A
Renard S,Lingueglia E, Voilley N, Lazdunski M, and Barbry P. Biochemical analysisof the membrane topology of the amiloride-sensitive Na channel. J Biol Chem 269:12981-12986, 1994.1 B5 v9 y% z2 V

Rossier BC,Pradervand S, Schild L, and Hummler E. Epithelial sodium channel and thecontrol of sodium balance: interaction between genetic and environmentalfactors. Annu Rev Physiol 64:877-897, 2002.; A& a2 H+ h5 h' h7 I2 P

: X6 ]  c/ e- j) V# Z
  }3 }% t0 H2 T; ~! s
Rotin D,Kanelis V, and Schild L. Trafficking and cell surface stability of ENaC. Am J Physiol Renal Physiol 281:F391-F399, 2001.

; J+ ~) R" Z* Y" V% I" d
Schild L,Schneeberger E, Gautschi I, and Firsov D. Identification of amino acidresidues in the alpha, beta, and gamma subunits of the epithelial sodiumchannel (ENaC) involved in amiloride block and ion permeation. JGen Physiol 109:15-26, 1997.

7 z& C# r2 N* C
6 m% M) t$ m: F. Q8 d' z2 n7 \
Sheng S, Li J,McNulty KA, Avery D, and Kleyman TR. Characterization of the selectivityfilter of the epithelial sodium channel. J Biol Chem 275: 8572-8581,2000." Y; o. i/ H1 r) _3 j6 y
& [- q9 q! G$ v

Sheng S, Li J,McNulty KA, Kieber-Emmons T, and Kleyman TR. Epithelial sodium channelpore region structure and role in gating. J Biol Chem 276: 1326-1334,2001.6 w6 q% u; b) Q; W6 ?
( x/ q6 p) Q: E4 ~" C* |! _2 w

; `  z& |$ p2 z
Sheng S,McNulty KA, Harvey JM, and Kleyman TR. Second transmembrane domains ofENaC subunits contribute to ion permeation and selectivity. J BiolChem 276:44091-44098, 2001.

. O" k: Q" R- b' f" E% z$ w
6 R2 ^; h5 c3 w- H; i) c" e  x
Sheng S, PerryCJ, and Kleyman TR. External nickel inhibits epithelial sodium channel bybinding to histidine residues within the extracellular domains of alpha andgamma subunits and reducing channel open probability. J BiolChem 277:50098-50111, 2002.

% K+ {" X7 U  }1 y
+ n- b- D4 D, @- Z1 [+ I* f' L
Shimkets RA,Warnock DG, Bositis CM, Nelson-Williams C, Hansson JH, Schambelan M, Gill JRJr, Ulick S, Milora RV, Findling JW, Canessa CM, Rossier BC, and LiftonRP. Liddle's syndrome: heritable human hypertension caused by mutations inthe beta subunit of the epithelial sodium channel. Cell 79:407-414, 1994./ O; |/ `4 p% K# g1 K
! }! f# y* h" V+ j- o+ Q% \

! ?2 w& g) D8 o
Snyder PM. The epithelial Na channel: cell surface insertion and retrieval inNa homeostasis and hypertension. EndocrRev 23:258-275, 2002.; P9 e# ]( O8 }) g4 S* _+ Q

- @- ]$ T; J/ l0 _4 M0 n

Snyder PM,Bucher DB, and Olson DR. Gating induces a conformational change in theouter vestibule of ENaC. J Gen Physiol 116: 781-790,2000.) e* U' p% N, J, P2 ]9 E; v0 q
# j$ j! Q! o! }2 e2 s. E9 j

Snyder PM,Cheng C, Prince LS, Rogers JC, and Welsh MJ. Electrophysiological andbiochemical evidence that DEG/ENaC cation channels are composed of ninesubunits. J Biol Chem 273:681-684, 1998." }) J5 i) Z' L5 t4 e! J

  W& q, y) C7 A  b( u
1 l/ [- |# l: Z/ f$ ^% I" ]
Snyder PM,McDonald FJ, Stokes JB, and Welsh MJ. Membrane topology of theamiloride-sensitive epithelial sodium channel. J BiolChem 269:24379-24383, 1994.
6 n  f8 `, W7 M7 K
8 v' `. X, }# H+ W; d

Snyder PM,Olson DR, and Bucher DB. A pore segment in DEG/ENaC Na channels. J Biol Chem 274:28484-28490, 1999." U# h! g, N; z+ D
2 R0 T3 d+ F4 X' a! [* q. n' v0 k( t
/ i$ J( ^, d* B$ ?/ n3 m
  G. V9 j/ Z, P# [; b8 H: ]+ k
Stockand JD. New ideas about aldosterone signaling inepithelia. Am J Physiol Renal Physiol 282: F559-F576,2002.- V! E& e' I5 z$ ~1 Q1 P9 U  U8 Q  x, Y
( P9 B; s1 x  \; J# b! M

3 L3 R! V- L! J; h# J
Strautnieks SS,Thompson RJ, Gardiner RM, and Chung E. A novel splice-site mutation in thegamma subunit of the epithelial sodium channel gene in threepseudohypoaldosteronism type 1 families. Nat Genet 13: 248-250,1996.
. [' ~. K5 @; c) w+ X

Tamura H,Schild L, Enomoto N, Matsui N, Marumo F, and Rossier BC. Liddle diseasecaused by a missense mutation of beta subunit of the epithelial sodium channelgene. J Clin Invest 97:1780-1784, 1996.
& c) z3 J* q! F$ m  z. w

Tu L,Santarelli V, Sheng Z, Skach W, Pain D, and Deutsch C. Voltage-gatedK channels contain multiple intersubunit association sites. J Biol Chem 271:18904-18911, 1996.# i" d" w3 O6 D, Q, y
8 B/ H* t' N- Y9 ]! C; @

Tucker SJ, BondCT, Herson P, Pessia M, and Adelman JP. Inhibitory interactions betweentwo inward rectifier K channel subunits mediated by thetransmembrane domains. J Biol Chem 271: 5866-5870,1996.
- X" Q9 w, v& C
3 ^2 p1 D* l& E& U7 l+ \
0 w! |  }6 i* C4 M
Ubarretxena-Belandia I and Engelman DM. Helical membrane proteins:diversity of functions in the context of simple architecture. CurrOpin Struct Biol 11:370-376, 2001.2 a1 m0 B$ _/ V7 i8 @1 M7 I

% w9 i! i7 {1 i8 |, E0 L8 e+ ]
Verrey F. Early aldosterone action: toward filling the gap between transcription andtransport. Am J Physiol Renal Physiol 277: F319-F327,1999.
8 Y, b, C! H9 Q- b3 f& z) S: j
- f2 @) {& k) Y1 `4 K
4 ]' W) ^- Y% q7 k1 s
Weisz OA andMachamer CE. Use of recombinant vaccinia virus vectors for cell biology. Methods Cell Biol 43:137-159, 1994.
+ r+ P  c% \' {" P

9 a$ d* |/ A! K5 t! ~7 R" `* N5 e0 F
Weisz OA, WangJM, Edinger RS, and Johnson JP. Noncoordinate regulation of endogenousepithelial sodium channel (ENaC) subunit expression at the apical membrane ofA6 cells in response to various transporting conditions. J BiolChem 275:39886-39893, 2000.- I$ ^+ F! E3 p4 J) \- @" V

Woodward R,Stevens EB, and Murrell-Lagnado RD. Molecular determinants for assembly ofG-protein-activated inwardly rectifying K channels. JBiol Chem 272:10823-10830, 1997.$ x/ n2 ^! g8 L

9 v! @! |! Q: L9 N- l) ?/ ]9 U
Yamashita Y,Koga M, Takeda Y, Enomoto N, Uchida S, Hashimoto K, Yamano S, Dohi K, MarumoF, and Sasaki S. Two sporadic cases of Liddle's syndrome caused by de novoENaC mutations. Am J Kidney Dis 37: 499-504,2001.5 n! K9 p* f; W6 ?' X
# h" w" Z( P2 C

0 _- l5 ?% }# j
Yewdell JW. Not such a dismal science: the economics of protein synthesis, folding,degradation and antigen processing. Trends Cell Biol 11: 294-297,2001.

收藏0 分享0 顶0 踩0