Hypoxia-Induced Inhibition of Epithelial Na1 Channels in the Lung

ORIGINAL RESEARCH

Hypoxia-Induced Inhibition of Epithelial Na1 Channels in the LungRole of Nedd4-2 and the Ubiquitin-Proteasome PathwayThomas Gille1,6, Nadia Randrianarison-Pellan1, Arnaud Goolaerts2, Nicolas Dard1, Yurdagul Uzunhan1,6,Evelyne Ferrary3,4, Edith Hummler5, Christine Clerici2,4,6, and Carole Plane`s1,6

1Universite Paris 13, Sorbonne Paris Cite, Laboratoire Reponses Cellulaires et Fonctionnelles a` lHypoxie (EA 2363), Bobigny,France; 2INSERM U773, CRB3, Paris, France; 3INSERM UMR-S 867, Paris, France; 4Universite Denis Diderot, Sorbonne Paris Cite,Paris, France; 5Departement de Pharmacologie et de Toxicologie, Universite de Lausanne, Lausanne, Switzerland; and 6AssistancePublique-Hopitaux de Paris, Paris, France

Abstract

Transepithelial sodium transport via alveolar epithelialNa1 channels(ENaC) and Na1,K1ATPase constitutes the driving force forremoval of alveolar edema uid. Alveolar hypoxia associated withpulmonary edema may impair ENaC activity and alveolar Na1

absorption through a decrease of ENaC subunit expression at theapical membrane of alveolar epithelial cells (AECs). Here, weinvestigated the mechanism(s) involved in this process in vivo in theb-Liddle mouse strain mice carrying a truncation of b-ENaC C-terminus abolishing the interaction between b-ENaC and theubiquitin protein-ligase Nedd42 that targets the channel forendocytosis and degradation and in vitro in rat AECs. Hypoxia (8%O2 for 24 h) reduced amiloride-sensitive alveolar uid clearance by69% in wild-type mice but had no effect in homozygous mutatedb-Liddle littermates. In vitro, acute exposure of AECs to hypoxia(0.53% O2 for 16 h) rapidly decreased transepithelial Na

1

transport as assessed by equivalent short-circuit current Ieq and theamiloride-sensitive component of Na1 current across the apicalmembrane, reecting ENaC activity. Hypoxia induced a decrease ofENaC subunit expression in the apical membrane of AECs with nochange in intracellular expression and induced a 2-fold increase in

a-ENaC polyubiquitination. Hypoxic inhibition of amiloride-sensitive Ieq was fully prevented by preincubation with theproteasome inhibitorsMG132 and lactacystin orwith the antioxidantN-acetyl-cysteine. Our data strongly suggest that Nedd42-mediatedubiquitination of ENaC leading to endocytosis and degradation ofapical Na1 channels is a key feature of hypoxia-induced inhibition oftransepithelial alveolar Na1 transport.

Keywords: ubiquitination; alveolar uid clearance; membranetrafcking; Nedd4-2; antioxidant

Clinical Relevance

Alveolar uid clearance secondary to sodium ion absorption byalveolar epithelial cells plays a major role in the resolution ofpulmonary edema. However, it is impaired by alveolar hypoxia,which is usually present in lung injury. This paper investigatesthe mechanism of hypoxia-induced decrease in epithelialsodium channel activity and alveolar uid clearance in vitro incultured alveolar epithelial cells and in vivo in the mouse.

Active transepithelial sodium (Na1)transport by alveolar epithelial cells (AECs)is a driving force for the reabsorption ofuid from the alveolar space, a functionthat accounts for the ability of the lung toremove alveolar uid at the time of birthand represents the main mechanism for

alveolar edema resolution (13). The twomajor proteins involved in this process arethe basolateral sodium potassiumadenosine triphosphatase (Na1,K1ATPase) and the amiloride-sensitiveepithelial Na1 channel (ENaC) located inthe apical membrane of type I and type II

AECs (3, 4). ENaC, composed of threehomologous subunits (a, b, and g),represents in vivo a limiting factor for lunguid clearance (2, 510). The regulation ofENaC-mediated lung Na1 absorption istherefore critical for normal lung uidbalance and function (1, 3, 4).

(Received in original form December 15, 2012; accepted in final form August 11, 2013 )

This work was supported by Universite Paris 13, PRES Sorbonne Paris Cite, by Chancellerie des Universites de Paris (Fondation du Legs Poix), and byAssistance Publique Hopitaux de Paris.

Correspondence and requests for reprints should be addressed to Carole Plane`s, M.D., Ph.D., EA 2363, UFR SMBH, 74 rue Marcel Cachin, F-93017 Bobigny,France. E-mail: [email protected]

This article has an online supplement, which is accessible from this issues table of contents at www.atsjournals.org

Am J Respir Cell Mol Biol Vol 50, Iss 3, pp 526537, Mar 2014

Copyright 2014 by the American Thoracic SocietyOriginally Published in Press as DOI: 10.1165/rcmb.2012-0518OC on October 4, 2013

Internet address: www.atsjournals.org

526 American Journal of Respiratory Cell and Molecular Biology Volume 50 Number 3 | March 2014

Alveolar hypoxia, a situation usuallyassociated with pulmonary edema, has beenshown by our group and other investigatorsto impair in vivo or ex vivo Na1drivenalveolar uid clearance (AFC) in rat lung(11, 12). In vitro, exposure of rat AECs tohypoxia decreases Na1,K1ATPase andENaC activity (11, 1316). The mechanismwhereby hypoxia inhibits ENaC is notunivocal and may vary according to theduration and/or severity of oxygendeprivation (17). Whereas prolonged andsevere hypoxia (01% O2 for 12 h or more)was shown to inhibit ENaC subunitsynthesis at transcriptional andtranslational levels in cultured rat AECs,short exposure (01% O2 for 3 h) decreasedENaC activity without affecting ENaCmRNA transcript expression levels,suggesting a posttranslational regulation(16). Indeed, we have previously reportedthat mild hypoxia (3% O2) inhibited ENaCactivity in primary rat AECs, mostlythrough a decrease in ENaC subunit cellsurface expression, with no change in ENaCmRNA and protein expression (18). Themechanisms involved in the decrease ofENaC cell surface expression in hypoxicAECs have not been investigated. Thedensity of ENaC channels expressed at thecell surface depends on the balance betweenthe rate of insertion of new channels fromintracellular pools and the rate ofendocytosis. One major pathway for ENaCretrieval from the plasma membrane is itsubiquitination by the ubiquitin-proteinligase Nedd42 (neural precursor cellexpressed developmentally down-regulatedprotein), which leads to internalization ofthe channel and degradation by theproteasome or the lysosome, depending onthe cell type (1924). Because the Nedd42-mediated pathway is the target of numeroushormonal and biological regulations, wehypothesized that it could be involved inhypoxia-induced decrease in ENaC cellsurface expression.

The objective of the present study wasto investigate the cellular mechanism(s)responsible for the decrease of ENaC activityand cell surface expression in AECs exposedto hypoxia, with a particular focus on theNedd42-mediated endocytosis pathway.To do so, we combined in vitro and in vivoapproaches. We studied in vivo the effect ofhypoxia on AFC in wild-type and Liddlemice, a previously established knock-inmouse strain carrying a premature Stopcodon within the Scnn1b (b-ENaC) gene (7,

9). This mutation leads to the deletion ofthe b-ENaC C-terminus PY motif, thusabolishing the interaction between ENaCand the ubiquitin protein ligase Nedd42that targets the channel for endocytosis anddegradation (2529). We also establishedand characterized an in vitro model of acuteexposure of primary rat AECs to hypoxia,allowing us to perform electrophysiological,biochemical, and pharmacological studies.Our data show that ENaC-mediated AFCwas signicantly impaired by hypoxicexposure (8% O2 for 24 h) in wild-typemice but not in homozygous mutantb-Liddle mice. In vitro, acute exposure ofprimary rat AECs to various levels ofhypoxia (0.53% O2 for 16 h) induceda rapid decrease of amiloride-sensitivetransepithelial Na1 transport and ENaCactivity related to a decrease of ENaCsubunit protein expression in the apicalplasma membrane. This functional decreasewas associated with increasedpolyubiquitination of a-ENaC subunit andwas fully prevented by incubation withproteasome inhibitors and antioxidants.Taken together, our results strongly suggestthe involvement of Nedd42-mediatedubiquitination and subsequent endocytosisof ENaC channels in hypoxia-inducedimpairment in transepithelial alveolar Na1

transport.

Materials and Methods

Transgenic MiceThe generation of b-Liddle mice harboringthe R566 stop mutation on the Scnn1b genehas been previously described (7, 9).Experiments were performed on litter-matched mice aged 2 to 6 months (n = 56transgenic mice). Male C57B/6J mice(Elevage Janvier, Le Genest St. Isle, France)aged 2 to 3 months were also used inpreliminary experiments to determine thetime course of hypoxic exposure. Theexperiments were approved by theinstitutional review board on animalexperimentation and accorded with animalwelfare guidelines (Ministe`re Franais de laPeche et de lAgriculture, #751045).

Measurement of AFCMice were exposed to 21% O2 (controlnormoxic mice) or to 8% O2 (hypoxic mice;estimated O2 tension in alveolar gas: 40 mmHg) from 8 to 24 hours in speciallydesigned environmental Plexiglas chambers

(Minerva, Esternay, France) (12). At theend of exposure, Na1driven AFC wasimmediately measured in vivo at 378C overa 15-minute period using an in situnonventilated model in which the airspacewas instilled with an isoosmolar Ringerslactate solution containing 125I-albumin aspreviously described (8, 10, 30). AFC(percentage uid absorption at 15 min) wascalculated from the increase in alveolaruid albumin as follows: AFC (%) = (Cf Ci)/Cf 3 100, where Ci and Cf represent theinitial and nal concentrations of 125I-albumin in the aspirate at 1 and 15 minutes,respectively, as assessed by radioactivitymeasurements. In some experiments,amiloride (nal concentration in thealveolar instillate, 1 mM) was added to theinstillate to inhibit ENaC activity.

Quantication of ENaC Subunit andNedd42 Protein Expression in DistalLung Homogenates from Liddle MiceMice were killed with intraperitonealpentobarbital before a thoracotomy wasperformed. The lungs were removed andimmediately homogenized for 3 minutes inice-cold lysis buffer (pH 8) containing 20mM Tris, 150 mM NaCl, 1% Triton X-100,0.1% SDS, 0.5% deoxycholate, and proteaseinhibitors. The lysate was centrifuged(15,000 rpm, 10 min) at 48C, andsupernatants were aliquoted andimmediately frozen before use. Samples ofprotein extracts (100200 mg) wereresolved through 10% acrylamide gels,electroblotted, electrically transferred tonitrocellulose paper, and probed for a- org-ENaC subunits using rabbit polyclonalantia-rat ENaC (rENaC) and antig-rENaC antibodies (31) as previouslydescribed (7, 10, 30) or with rabbitpolyclonal anti-Nedd42 antibody(dilution, 1:2,000) (ab46521; Abcam,Cambridge, MA) (32). Quantication ofb-ENaC protein could not be done inLiddle mice because antib-ENaCantibodies recognize epitopes located in theC-terminus of b-ENaC protein, which isdeleted in Liddle mice (10).

Isolation and Culture of Rat AECsThe procedure of rat AEC isolationaccorded with legislation currently in forcein France and with animal welfareguidelines. AECs were isolated from adultmale, pathogen-free Sprague-Dawley rats byelastase digestion (Worthington, Lakewood,NJ) as previously described (16, 18, 33).

ORIGINAL RESEARCH

Gille, Randrianarison-Pellan, Goolaerts, et al.: Hypoxic Inhibition of ENaC in the Lung 527

Cells (purity . 92%; viability . 95%) wereseeded onto polycarbonate membranelters with a pore size of 0.4 mm(Transwell/Snapwell; Corning Inc.,Corning, NY) and cultured in DMEMcontaining 25 mM D-glucose, 10 mMHepes, 23.8 mM NaHCO3, 2 mM L-glutamine, 10% FBS, 50 U/ml penicillin, 50mg/ml streptomycin, and 10 mg/mlgentamycin. On Day 3 after isolation, theFBS concentration was reduced to 5%, anddexamethasone (nal concentration: 1 mM)was added to the culture medium exceptwhen indicated.

Exposure of AECs to Hypoxiaand DrugsFour or ve days after plating, the cells wereplaced in a humidied airtight incubatorwith inow and outow valves, and thehypoxic gas mixture containing 0.5% O2, 5%CO2, and 94.5% N2 was ushed at 5 l/minfor 20 minutes before the airtight incubatorwas sealed and kept at 378C for 1, 2, 3, 4, or6 hours. Oxygen tensions assayed in culturemedia at the end of 1-, 2-, 4-, and 6-hourhypoxic exposures were 65, 50, 30, and 30mm Hg, respectively. Control normoxiccells were maintained in a humidiedincubator (21% O2, 5% CO2, 74% N2) forthe same periods of time (oxygen tension inculture media, 140 mm Hg). To test varioushypoxia levels, cells were exposed for 4hours to moderate hypoxia (1.5% O2;oxygen tension in culture media, 45 mmHg) or mild hypoxia (3% O2; oxygentension in culture media, 60 mm Hg).

In some experiments, cells werepreincubated for 30 minutes witha lysosome inhibitor (chloroquine; nalconcentration, 75 mM) (Sigma-Aldrich, St.Louis, MO) or proteasome inhibitors,lactacystin (nal concentration, 20 mM)(Sigma-Aldrich) or MG132 (nalconcentration, 10 mM) (Sigma-Aldrich), orvehicle before being exposed to 0.5 or 21%O2 for 2 hours and processed forelectrophysiological measurements. Inother experiments, cells were preincubatedfor 30 minutes with the glutathioneprecursor N-acetyl-cysteine (NAC) (nalconcentration, 5 mM) (Sigma-Aldrich) orvehicle before normoxic/hypoxic exposure.

Electrophysiological Studies inPrimary Cultures of Rat AECs

Measurements under open-circuitconditions. Transepithelial resistance (Rte)

and voltage were measured in rat AECsgrown on semipermeable transwell ltersfor 5 days under open-circuit conditionsusing dual silver/silver chloride electrodesconnected to the Millicell Electricalresistance Clamp apparatus (Millipore,Billerica, MA) as previously described (8,34). Equivalent short-circuit current (Ieq)was calculated with Ohms law from Rteand PD. Measurements of Rte and PD wereperformed at 378C before and after apicaladdition of amiloride (Sigma) (nalconcentration, 10 mM) for 10 minutes toblock ENaC activity to calculate theamiloride-sensitive Ieq dened as thedifference between Ieq value in the absenceand Ieq value in the presence of amiloride.Short-circuit current measurements.Measurements of short-circuit current (Isc),transepithelial potential difference (PD),and Rte were performed in rat AECs grownon Snapwell lters mounted in a Ussingchamber system (Physiologic InstrumentsInc., San Diego, CA) as previouslydescribed (35). All experiments wereperformed with basolateral membranepermeabilization (achieved by the additionof nystatin in the basolateral chamber atnal concentration of 60 mg/ml) andasymetric bath solutions (NaClconcentration, 140 mM in apical bath vs. 10mM in basolateral bath). Short-circuitcurrent was measured before and afterexposure to amiloride (nal concentrationin the apical bath, 10 mM), and thedifference current, representing theamiloride-sensitive component of the Na1

current across the apical membrane(amiloride-sensitive Isc nyst), was calculated.

Detection of ENaC SubunitsExpressed at the Apical Surface ofCultured Rat AECsBiotinylation and recovery of apicalmembrane proteins from cultured rat AECswere performed as previously described (18,34). Samples of biotinylated andnonbiotinylated proteins were resolvedthrough 10% acrylamide gels,electroblotted, electrically transferred tonitrocellulose paper, and probed for rENaCsubunits (18, 31, 34) and for b-actin.

Detection of Polyubiquitinateda-ENaC Subunits in Rat AECsPolyubiquitinated proteins from rat AECextracts were separated fromnonubiquitinated and monoubiquitinated

proteins using agarose-TUBE2 resin(Tandem Ubiquitin-Binding Entities type 2;LifeSensors, Malvern, PA), and controlagarose beads (LifeSensors) were used toassess nonspecic binding according to themanufacturers instructions (details areprovided in the online supplement).Tandem Ubiquitin-Binding Entities type 2recognize tetra-ubiquitin with a very highafnity and are valuable tools for purifyingendogenous polyubiquitin-conjugatedproteins (36). The samples(nonubiquitinated/monoubiquitinatedproteins and polyubiquitinated proteins)were processed for Western blotting andprobed for a-rENaC as described above orwith mouse monoclonal antiubiquitinantibody (P4D1; Santa Cruz BiotechnologyInc., Santa Cruz) (dilution, 1:1,000) (36).

Statistical AnalysisResults are presented as means 6 SE. Forfunctional data, one-way or two-wayANOVA were performed, and, whenallowed by the F value, results werecompared by the modied protected leastsignicant difference (Fishers PLSD),adjusted by the Bonferroni correction whenneeded. For Western blot experiments,differences between groups were evaluatedwith paired or unpaired t test performed onraw densitometric data. A P value , 0.05was considered signicant. StatView (SASInstitute Inc., Cary, NC) and Prism(GraphPad Software Inc., La Jolla, CA)software was used.

Results

Effect of Hypoxia on AFC and ENaCProtein Expression in Wild-Type andLiddle MiceTo examine the role of the ubiquitinprotein-ligase Nedd42 in hypoxia-induceddecrease of alveolar Na1 transport, westudied in vivo the effect of hypoxicexposure on AFC in Liddle mice harboringthe R566 stop mutation on the Scnn1b gene.The R566 stop mutation leads to the deletionof the b-ENaC C-terminus PY motif, thusabolishing the interaction between ENaCand Nedd42 that targets the channel forendocytosis and degradation (9). In a rstset of experiments, we established the timecourse of hypoxic exposure on Na1drivenAFC in C57Bl6J mice (Figure 1A). Hypoxicexposure (8% O2 for 824 h) induceda time-dependent decrease of total AFC,

ORIGINAL RESEARCH


being signicant at 24 hours (P , 0.05,ANOVA and Fishers PLSD). In subsequentexperiments, littermate 1/1 andhomozygous mutated L/L mice weresubmitted to normoxic/hypoxic exposurefor 24 hours (Figures 1B and 1C). TotalAFC measured in the absence of amilorideunder normoxic conditions was

signicantly higher in L/L mice than in1/1 mice (16.5 6 1.35% vs. 10.9 6 0.94%uid cleared in 15 min; P, 0.001, ANOVAand Bonferroni) (Figure 1B) as previouslyreported (10). Hypoxia induced a 39%decrease in total AFC in 1/1 mice (P ,0.01) but had no signicant effect in L/Llittermates (Figure 1B). The amiloride-

insensitive component of AFC was similarin 1/1 and L/L mice and was not affectedby hypoxic exposure (Figure 1B).Therefore, the inhibition of total AFCobserved in hypoxic wild-type mice wassolely due to a decrease of the amiloride-sensitive component of AFC reectingENaC-mediated Na1 absorption (269% ascompared with normoxic condition; P ,0.01, ANOVA and Bonferroni). Bycontrast, amiloride-sensitive AFC was notsignicantly modied by hypoxic exposurein L/L mice (Figure 1C).

To determine whether the Liddlemutation and/or exposure to hypoxiamodied the expression of ENaC and of theubiquitin protein-ligase Nedd42 in lungepithelial cells, a- and g-ENaC subunit andNedd42 protein levels were quantied indistal lung homogenates from littermate wild-type and homozygous mutated L/L miceexposed for 24 hours to normoxia or hypoxia.Protein levels of a- and g-ENaC subunits andof Nedd42 were similar under normoxiccondition in 1/1 and L/L mice (Figure 2).Exposure to hypoxia has no signicant effecton ENaC or Nedd42 protein expression.

Effect of Acute Hypoxic Exposure onTransepithelial Sodium Transportacross Rat AEC MonolayersTo further investigate the cellularmechanisms involved in hypoxia-inducedinhibition of ENaC activity, we establishedan in vitro model of short-term exposure ofAECs to hypoxia, allowing us to performfunctional, pharmacological, andbiochemical experiments. Rat AECmonolayers were rst exposed to normoxiaor severe hypoxia (0.5% O2) for 1 to 6hours, and transepithelial Na1 transportwas assessed under open-circuit conditions(Figure 3). The total equivalent short-circuit current Ieq was stable in allnormoxic experimental conditions(Figure 3A). Hypoxia induced a progressivedecrease in total Ieq, which was signicantfor exposures of at least 2 hours. Thisdecrease of total Ieq was solely due toa decrease of the amiloride-sensitive part ofthe current because the amiloride-insensitive part was not signicantlymodied by hypoxia (Figure 3A). Theamiloride-sensitive component of Ieq,reecting the activity of apical ENaCchannels and basolateral Na,K-ATPase, wasreduced to 60% of normoxic control after 2hours of hypoxic exposure (P , 0.01) andcontinued to decrease to reach a plateau at

Figure 1. Effect of hypoxia on alveolar fluid clearance (AFC) in wild-type and Liddle mice. (A) MaleC57Bl6J mice aged 2 to 3 months were exposed to hypoxic atmosphere (8% O2, dark gray) forincreasing times (024 h) as indicated. AFC was measured at the end of exposure over a 15-minuteperiod at 378C using an in situ nonventilated model in which the airspace was instilled with anisoosmolar Ringers lactate solution containing 125I-albumin as a volume marker, as described inMATERIALS AND METHODS. Results are expressed as percentage fluid absorption at 15 minutes andrepresent means 6 SE (n = 59 mice per condition). *Significantly different from value at time 0 (P ,0.05) (ANOVA and Fishers protected least significant difference test). (B) Sodium-driven AFC wasmeasured at baseline (total AFC, open bars) and in the presence of amiloride (final concentration inthe instillate, 1 mM) (amiloride-insensitive AFC, hatched bars) in wild-type (1/1) and homozygousmutated (L/L) littermates aged 2 to 5 months exposed to normoxic (21% O2, light gray) or hypoxic (8%O2, dark gray) atmosphere for 24 hours. (C) Calculated values of amiloride-sensitive AFC. Resultsrepresent means 6 SE (n = 49 mice per condition). Significantly different from value in (1/1) miceexposed to normoxia at *P , 0.05 and **P , 0.01. NS = nonsignificant (ANOVA and Bonferroni).

ORIGINAL RESEARCH


40% of normoxic value after 4 hours ofhypoxic exposure (P , 0.001) (Figure 3B)(two-way ANOVA and Bonferroni). Thisinhibition was observed whether the cells werecultured or not with dexamethasone (seeFigure E1 in the online supplement). Exposureof AECs to moderate hypoxia (1.5% O2) for 4hours induced a marked decrease of theamiloride-sensitive Ieq (P , 0.001) similar tothat observed with 0.5% O2, whereas exposureto mild hypoxia (3% O2) had a more moderateinhibitory effect (P , 0.001) (Figure 3C).

To specically assess the effect ofhypoxia on ENaC activity, AEC monolayersexposed to hypoxia/normoxia for 4 hourswere mounted in an Ussing chamber, and Iscnyst was measured after permeabilization ofbasolateral membrane with nystatin. Theamiloride-sensitive Isc nyst, representing thecomponent of the apical Na1 currentmediated by ENaC, was reduced by almost70% in hypoxic cells as compared withnormoxic control cells (P , 0.01, unpairedt test) (Figure 3D).

Effect of Acute Hypoxic Exposure onENaC Subunit Protein Expression inthe Apical Membrane of Rat AECsTo further examine whether the inhibitoryeffect of hypoxia on transepithelial Na1

transport could be due to a decrease of theabundance rENaC subunits expressed at theapical membrane, rat AEC monolayerswere exposed to normoxia or hypoxia for 2hours before apical cell surface biotinylationand subsequent separation of intracellular(nonbiotinylated) and surface (biotinylated)proteins were performed. Immunoblottingwith rabbit polyclonal anti a-rENaCantibody revealed a major band withmolecular mass of 80 to 85 kD inintracellular extracts of AECs, whereas theprominent band in biotinylated extractsmigrated at 65 kD, as previously described(Figure 4A, left panel) (18, 34).Immunoblotting with rabbit polyclonalantib-rENaC (Figure 4A, middle panel)showed one single band migrating at 95 to100 kD in intracellular and surface extracts.Immunoblotting with rabbit polyclonalantig-rENaC (Figure 4A, right panel)detected a major band around 90 kD anda minor band at 75 kD in intracellularextracts but detected only a band at 90 kD inbiotinylated extracts. A 2-hour hypoxicexposure did not modify the abundance ofintracellular rENaC subunit proteins(Figure 4B). By contrast, levels ofbiotinylated a-, b-, and g-rENaC werereduced by 37% (P , 0.01), 68% (P , 0.05),

and 70% (P , 0.05), respectively, in hypoxiccells (Figure 4C) (paired t test).

Effect of Hypoxic Exposure on theUbiquitination of a-ENaC SubunitsUbiquitination is a common mechanism totarget endocytosis and degradation of cellsurface proteins. In epithelial cellsendogenously expressing ENaC, thismechanism mainly involves the binding ofseveral ubiquitin molecules on ENaCsubunits by the ubiquitin protein-ligaseNedd42 (19, 20, 22, 24, 37). To testwhether hypoxia affected ENaCubiquitination in our in vitro model, ratAEC monolayers were exposed tonormoxia or hypoxia for 4 hours beforepolyubiquitinated proteins were puriedusing agarose beads coupled with TUBEs,and polyubiquitinated a-ENaC protein wasquantied (Figure 5). An antiubiquitinantibody detected a large amount ofubiquitinated proteins of various molecularmasses in cell extracts incubated withagarose beads coupled with TUBEs(Figure 5A, diffuse bands in lanes 1 and 2)as expected (36), whereas no signal wasdetected in cell extracts incubated withcontrol agarose beads (Figure 5A, lanes 3and 4), excluding a nonspecic binding

Figure 2. Effect of hypoxia on a- and g-epithelial Na1 channel (ENaC) subunit protein expression in distal lung homogenates from wild-type and Liddlemice. Wild-type (1/1) and homozygous mutated (L/L) littermates aged 2 to 5 months were exposed to normoxic (21% O2, light gray) or hypoxic (8% O2,dark gray) atmosphere for 24 hours. Western blotting was performed on whole distal lung homogenates as described in MATERIALS AND METHODS.Representative immunoblots show the expression of the a-ENaC subunit (A), the g-ENaC subunit (B), and the Nedd42 protein (C). The intracellularprotein b-actin was used as a loading control. Arrowheads indicate molecular mass markers. Different parts of the gel delineated in C come from the sameoriginal gel and not from different experiments. Only the order of indicated lanes has been changed to fit with the text and to facilitate the comprehensionof the figure. Quantification of a-ENaC, g-ENaC, and Nedd42 signals was obtained using NIH image software, and data were normalized for thecorresponding b-actin signal. Results are expressed as the ratio of the protein of interest to b-actin and represent means 6 SE (n = 3 mice per group).There was no significant difference between groups (ANOVA and Fishers protected least significant difference).

ORIGINAL RESEARCH


between agarose and polyubiquitinatedproteins. a-ENaC was detected as a majorband around 80 to 85 kD in the inputsamples (total cell extracts, not incubatedwith beads) (Figure 5B, lanes 1 and 2) andin the unbound fraction of samplesincubated with TUBEs (representingnonubiquitinated or monoubiquitinatedproteins) (Figure 5B, lanes 5 and 6). In thesamples bound to TUBEs (representingpolyubiquitinated proteins) (Figure 5B,lanes 3 and 4), a shift in a-ENaC proteinsignal was observed, with a major band at90 to 95 kD and a minor band at 110 kD(data not shown), as previously describedby Malik and colleagues (37) (Figure 5B).The amount of a-ENaC subunits in totalcell extracts was not signicantly differentafter normoxic or hypoxic exposure(Figure 5B, lanes 1 and 2; Figure 5C, leftpanel). Also, there was no signicantdifference in the amount of a-ENaCsubunits unbound to TUBEs (Figure 5B,lanes 5 and 6; Figure 5C, right panel). Onthe contrary, hypoxia induced a 2-foldincrease of polyubiquitinated a-ENaCsubunits (P , 0.01; ANOVA and FishersPLSD) (Figure 5B, lanes 3 and 4; Figure 5C,right panel).

Exposure to hypoxia did not modifyNedd42 protein expression in rat AECextracts (details are provided in the onlinesupplement and in Figure E3).

Effect of Proteasome and LysosomeInhibitors on Hypoxia-InducedDecrease in TransepithelialSodium TransportIn most native epithelial cellsendogenously expressing ENaC,polyubiquitinated ENaC proteins arepreferentially degraded in the proteasome(22, 37). However, in transfected cell lines,monoubiquitinated ENaC can also bedegraded via the lysosomal pathway. Wetherefore tested whether pharmacologicalinhibition of these pathways couldprevent the hypoxia-induced decrease ofamiloride-sensitive alveolar Na1

transport in AECs. Rat AECs werepreincubated for 30 minutes withproteasome inhibitors (lactacystin orMG132), with the lysosome inhibitorchloroquine, or with vehicle before beingexposed for 2 hours to a normoxic/hypoxic atmosphere and processed forelectrophysiological measurements.Preincubation with lactacystin (20 mM)(Figure 6A) or MG132 (10 mM)

Figure 3. Effect of acute hypoxia on transepithelial sodium transport across rat alveolar epithelial cells(AECs). Rat airway epithelial cells (AECs) grown for 5 days on Transwell filters were exposed tonormoxic (21% O2, light gray) or hypoxic (0.5% O2, dark gray) atmosphere for increasing times (06h). (A) Equivalent short-circuit current (Ieq) was measured under open-circuit conditions immediately atthe end of exposure (plain bars), and after apical treatment with amiloride (10 mM) (hatched bars).Results are expressed in mA/cm2 and represent means 6 SE (422 filters per condition from sixindependent experiments). Significantly different from corresponding normoxic value at *P , 0.05,**P , 0.01, and ***P , 0.001; significantly different from hypoxic value at time 0 at #P , 0.05 and###P , 0.001. NS = nonsignificant (two-way ANOVA and Bonferroni). (B) Amiloride-sensitive Ieq wascalculated as the difference between Ieq values without and with amiloride. Significantly different fromcorresponding normoxic value at *P , 0.05, **P , 0.01, and ***P , 0.001 (two-way ANOVA andBonferroni). Dashed line: nonlinear regression analysis of amiloride-sensitive Ieq values in hypoxicAECs. (C) Rat AECs were exposed for 4 hours to normoxia or to various levels of hypoxia (0.5, 1.5,and 3% O2) before amiloride-sensitive Ieq was measured (2030 filters per condition from sixindependent experiments). ***Significantly different from value in normoxic cells at P , 0.001;###significantly different from value in cells exposed to 3% O2 at P , 0.001 (ANOVA and Fishersprotected least significant difference). (D) Rat AECs on Snapwell filters were exposed for 4 hours tonormoxic or hypoxic (0.5% O2) atmosphere and rapidly mounted in Ussing chambers. Basolateralmembrane permeabilization was achieved by adding nystatin in the basolateral compartment. Short-circuit current was measured before and after apical treatment with amiloride (10 mM), and thedifference in current, representing the amiloride-sensitive component of the Na1 current across theapical membrane (amiloride-sensitive Isc nyst), was calculated. Results represent means 6 SE (eightfilters per condition from three independent experiments). **Significantly different from correspondingnormoxic value at P , 0.01) (unpaired t test).

ORIGINAL RESEARCH


(Figure 6B) fully prevented the inhibitoryeffect of hypoxia on amiloride-sensitiveIeq. By contrast, preincubation withchloroquine (75 mM) had no protectiveeffect in hypoxic AECs (Figure 6C).

Effect of NAC on TransepithelialSodium Transport and Ubiquitinationof a-ENaC Subunits underHypoxic ConditionHypoxia has been shown to increase theproduction of mitochondrial reactive

oxygen species (ROS) in AECs (38). Toexamine whether increased ROSproduction could be involved inhypoxia-induced inhibition oftransepithelial Na1 transport, the effectof NAC, a glutathione precursor withantioxidant properties, was tested.Preincubation of AECs with NAC (5mM) completely prevented the decreaseof amiloride-sensitive Ieq induced bya 4-hour exposure to 0.5 or 1.5% O2(Figure 7).

Discussion

The present study was designed toinvestigate the cellular mechanism(s)responsible for hypoxia-induced decrease ofENaC cell surface expression andtransepithelial alveolar Na1 transport bycombining in vitro and in vivo approachesin rodents. The number of channels on theapical surface of epithelial cells is a majordeterminant of the magnitude of ENaC-mediated Na1 transport (19, 20, 23).Channel density in the plasma membranedepends on the balance between the rate ofENaC insertion and the rate of removalfrom the apical membrane. The mechanismof channel insertion in the plasmamembrane is not clearly understood,although it is well recognized that ENaCdelivery at the apical membrane isincreased by cAMP/PKA stimulation (18,23, 39). Retrieval of ENaC from the plasmamembrane has been widely studied and isconsidered as a key process in theregulation of apical channel density.Accumulating evidence indicate thatubiquitin conjugation is a prerequisite forENaC endocytosis and subsequentdegradation (1921, 23, 24, 26). In mostnative epithelial cells expressingendogenous ENaC, ENaC subunits areusually polyubiquitinated and theninternalized and directed to the proteasomefor degradation (22, 24, 37). The binding ofconsensus PY motifs located in b- org-ENaC C-terminus to the WW domainsof Nedd42, a specic ubiquitin-ligaseisoform, is required for ubiquitination ofENaC subunits (24, 27, 28). Two recentstudies using Nedd42-decient miceclearly demonstrated that Nedd42, whichis coexpressed with ENaC in lung epithelialcells transporting Na1, plays a crucial rolein the regulation of ENaC activity in thelung (40, 41). The deletion of Nedd42,whether it was restricted or not to the lung,induced the premature death of the mice(before the age of 3 wk), preventing the useof these animal models to study Nedd42in the adult lung.

To evaluate the role of Nedd42 inhypoxia-induced ENaC inhibition in theadult lung, we compared in vivo the effectof hypoxia on Na1driven AFC in wild-type and b-Liddle mice carryinga premature Stop codon in the Scnn1b(b-ENaC) gene corresponding to the R566 stopgain-of-function mutation found in human

Figure 4. Effect of acute hypoxia on ENaC subunit cell surface expression in rat AECs. RatAECs grown on transwell filters were exposed to normoxic (21% O2) or hypoxic (0.5% O2)atmosphere for 2 hours before apical cell surface biotinylation experiments were performed asdescribed in MATERIALS AND METHODS to assess cell surface expression of rat ENaC (rENaC)subunits. (A) Representative immunoblots showing the expression of a-rENaC subunit (left),b-rENaC (middle), and g-rENaC (right) subunits in nonbiotinylated (intracell. = intracellular) andbiotinylated extracts (surface). b-actin, an abundant intracellular protein, was not detected inbiotinylated (surface) extracts. Quantification of intracellular (nonbiotinylated) (B) and cellsurface (biotinylated) (C) a-, b-, and g-rENaC protein signals in normoxic (light gray) andhypoxic (dark gray) cells was obtained using NIH image software. Data were normalized forthe b-actin signal in corresponding intracellular extracts. Results represent means 6 SE (n =3five separate experiments). Statistical significance was calculated from the raw data bypaired t test. Significantly different from normoxic value at *P , 0.05 and **P , 0.01 (pairedt test).

ORIGINAL RESEARCH


patients with Liddles syndrome (9, 10, 25,27). Liddles syndrome is an autosomaldominant form of salt-sensitive hypertensiondue to increased Na1 absorption in the distalnephron (25, 27, 29). The R566 stopmutation leads to the deletion of theb-ENaC C-terminus containing consensusPY motifs that bind to the WW domainsof Nedd42 (9), thus abolishing theinteraction between b-ENaC andNedd42. In wild-type mice, hypoxia (8%O2 for 24 h) decreased the amiloride-sensitive component of AFC reectingENaC-mediated Na1 transport by almost70%. The hypoxic inhibition of ENaC-mediated AFC was not associated witha decrease of the ENaC subunit proteinlevels in distal lung homogenates,suggesting a posttranslational mechanism.In homozygous mutated Liddle mice, totaland amiloride-sensitive AFC weresignicantly higher than in wild-typelittermates under normoxic condition, aspreviously reported (10). The stimulationof basal AFC in Liddle mice can beattributed to increased density of ENaCchannels on the apical membrane of AECsdue to reduced endocytosis anddegradation of ENaC subunits (26, 27). InLiddle mice, exposure to hypoxia did notimpair total or amiloride-sensitive AFC asit did in wild-type mice. This indicatesthat the interaction between b-ENaC andNedd42 is required for hypoxicinhibition of alveolar Na1 transport, atleast in mouse lung, and suggests thatubiquitination of ENaC by Nedd42 isinstrumental in this process.

To further investigate the effect ofhypoxia on ENaC ubiquitination anddegradation, we established an in vitromodel of acute hypoxic exposure of ratAECs, allowing us to perform functional,biochemical, and pharmacological studies.Our data show that exposure to severehypoxia (0.5% O2) for as little as 2 hoursdecreased amiloride-sensitive Na1

transport and ENaC activity in rat AECmonolayers by approximately 40%.Exposure to moderate hypoxia (1.5% O2)induced a functional inhibition ofequivalent magnitude, whereas mildhypoxia (3% O2) had a weaker but stillsignicant inhibitory effect. Steady-stateoxygen tensions in cell culture media(z 30, 45, and 60 mm Hg for 0.5, 1.5, and3% O2, respectively) were in the range ofwhat can be found in the alveolar milieu incase of hydrostatic pulmonary edema or acute

Figure 5. Effect of acute hypoxia on polyubiquitination of a-rENaC subunits. Rat AECs grown onTranswell filters were exposed to normoxic (21% O2) or hypoxic (0.5% O2) atmosphere for 4 hoursbefore cell extracts were incubated with Tandem Ubiquitin-Binding Entities (TUBEs) beads asdescribed in MATERIALS AND METHODS to separate nonubiquitinated and monoubiquitinated proteins(unbound fraction) from polyubiquitinated proteins (bound to TUBEs) or to control agarose beads. (A)Representative immunoblot showing polyubiquitinated proteins purified with TUBEs as detected withan antiubiquitin antibody (lanes 1and 2), as indicated. No signal was detected in protein extractsincubated with control agarose beads (lanes 3 and 4), indicating the absence of nonspecific bindingof ubiquitinated proteins to agarose beads. (B) Representative immunoblots showing the expressionof a-rENaC subunits in total cell extracts (input samples, 1 and 2), in protein extracts bound to TUBEs(lanes 3 and 4) and in the unbound fraction (lanes 56), or in protein extracts incubated with controlagarose beads (lanes 7 and 8). Different parts of the gel delineated in B come from the same originalgel and not from different experiments. Only the order of indicated lanes has been changed to fit withthe text of the manuscript and to facilitate the comprehension of the figure. Note the shift of a-rENaCsignal detected in ubiquitinated protein extracts (lanes 3 and 4) (9095 kD as compared with 8085kD in unbound extracts). b-actin, an abundant intracellular protein, was not detected in samplesbound to TUBEs or incubated with control agarose beads. (C) Quantification of a-ENaC subunits afternormoxic (light gray) or hypoxic (drak gray) exposure. Signal quantification was obtained using NIHImageJ software, and data were normalized for the b-actin signal in corresponding input samples,used as a loading control. Results represent means 6 SE of four separate experiments. Statistical

ORIGINAL RESEARCH


lung injury/acute respiratory distresssyndrome. Inhibition of transepithelial Na1

transport was observed whether the cells

were cultured or not with dexamethasone(further details are provided in the onlinesupplement and in Figure E1), in line with

previous work showing no protective effectof corticosteroids on hypoxia-induceddecrease of Na1 transport in renalepithelial cells (42). The time course ofhypoxic inhibition of transepithelial Na1

transport that we observed in vitro in ratAECs was somewhat shorter than thatobserved in mouse lung in vivo. This couldbe due to differences between species (ratvs. mouse) and to the presence in vivo ofadaptive responses to hypoxic stress (suchas release of endogenous catecholamines)that may affect ENaC and Na1,K1ATPaseactivities (3). The very short time course ofENaC hypoxic inhibition in vitro could notbe explained by a transcriptionalmechanism. Indeed, we previously showedthat severe hypoxic exposure (0% O2) for 3hours did not modify a- and b-ENaCmRNA transcript levels in rat AECs andonly slightly decreased g-EnaC mRNAlevels (18). In the present study,intracellular pools of a-, b-, and g-ENaCsubunit proteins in rat AECs were notaffected by the 2-hour hypoxic exposure,whereas cell surface expression was reducedby 37, 68, and 74%, respectively. Also,treatment with terbutaline, which promotesthe insertion of new ENaC channels in theapical membrane (18, 39), completely andrapidly reversed the functional decreaseinduced by hypoxia (details are provided inthe online supplement and in Figure E2).Together, these data indicate that exposureof AECs to hypoxia rapidly inhibitedENaC-mediated Na1 transport througha posttranslational mechanism inducinga decrease in ENaC cell surface expression.It was also previously reported that short-term exposure of rat AECs or A549 cells tohypoxia induced a rapid decrease of Na1,K1ATPase activity due to increasedinternalization of the protein and decreasedcell surface expression (13, 14, 43).

Consistent with in vivo ndings inmice, our in vitro data strongly suggest therole of ENaC ubiquitination andsubsequent internalization in response tohypoxia. Indeed, short-term hypoxicexposure induced a 2-fold increase ina-ENaC polyubiquitination in cultured rat.To our best knowledge, this constitutes therst evidence that ubiquitination of ENaCin native epithelial cells is up-regulated byhypoxia. Also, our data show that thehypoxia-induced decrease of ENaC-mediated Na1 transport as assessed byamiloride-sensitive Ieq was fully preventedby preincubation with lactacystin or

Figure 5. (Continued). significance was calculated from the raw data. There was no significantdifference between groups in input samples (paired t test). **Significantly different from correspondingnormoxic value at P , 0.01 (ANOVA and Fishers protected least significant difference test).

Figure 6. Effect of proteasome and lysosome inhibitors on hypoxia-induced decrease intransepithelial sodium transport. Rat AECs grown for 5 days on Transwell filters were preincubated for30 minutes with lactacystin (20 mM) (A), MG132 (10 mM) (B), chloroquine (75 mM) (C), or vehicle beforebeing exposed to normoxic (21% O2, light gray) or hypoxic (0.5% O2, dark gray) atmosphere for 2hours. Amiloride-sensitive Ieq was measured under open-circuit condition immediately at the end ofexposure. Results are expressed in mA/cm2 and represent means 6 SE (610 filters per conditionfrom six independent experiments). Significantly different from value in normoxic cells treated withvehicle at **P , 0.01 and ***P , 0.001. Significantly different from value in hypoxic cells treated withvehicle at #P , 0.05 and ###P , 0.001. NS = nonsignificant (ANOVA and Fishers protected leastsignificant difference test).

ORIGINAL RESEARCH


MG132, which block the chymotrypticactivity of the proteasome, whereaspreincubation with chloroquine, aninhibitor of the lysosome, had no protectiveeffect. This suggests that theubiquitinproteasome proteolytic pathway(rather than the lysosomal pathway) isinvolved in the degradation ofpolyubiquitinated ENaC channels inhypoxic AECs. This is in accordance withprevious studies showing that, inuntransfected epithelial cells endogenouslyexpressing ENaC, polyubiquination ofENaC by Nedd42 leads to endocytosis ofthe channel and degradation via theproteasomal pathway (37). However,because we did not directly measure theproteasomal degradation of ENaCchannels, we cannot exclude the possibilitythat the preventive effect of proteasome

inhibitors was simply the consequence ofdecreased endocytosis due to a depletion ofthe cellular pool of free ubiquitin, aspreviously reported for MG132 (44).

Another interesting nding of thepresent study is the fact that the antioxidantNAC completely prevented the hypoxia-induced decrease in transepithelial Na1

transport in rat AECs. Decreased O2availability induces the production of ROSby complex III in the mitochondrialelectron transport chain (38),a phenomenon that could explain many ofthe biological effects of hypoxia,particularly in the eld of ion transport.Lazrak and colleagues recently reportedthat increased steady-state concentrationsof ROS (induced in their study by inuenzavirus M2 protein) in human respiratory celllines (H441 and A549) decreased the

activity of wild-type ENaC (but not ofLiddle mutants) by enhancing itsendocytosis and degradation in theproteasome (45). The inhibitory effect ofROS on ENaC could have severalexplanations. Hypoxia-inducedmitochondrial production of ROS canactivate adenosine monophosphate-activated kinase (43), which has beenshown to activate Nedd42 and inhibitENaC in various epithelial cell types(4648). Also, ROS can activate PKC,which promotes targeted degradation ofENaC channels in some cell types (45, 49).Finally, ROS were recently shown to inducea reversible and global inactivation ofdeubiquitinases, which remove ubiquitinconjugates from diverse substrates (50),a phenomenon that could potentiallyincrease ubiquinated forms of ENaC andtheir subsequent internalization.

In conclusion, the present studyprovides strong evidence that the interactionbetween Nedd42 and b-ENaC as well aspolyubiquitination and subsequentendocytosis of Na1 channels plays a crucialrole in the pathophysiology of hypoxia-induced inhibition of transepithelialalveolar Na1 transport. Inhibiting Nedd42and the proteosomal degradation pathwayor decreasing ROS concentrations in AECscould represent future therapeutic targets toenhance ENaC activity and AFC inpathological situations associated withalveolar hypoxia. n

Author disclosures are available with the textof this article at www.atsjournals.org.

Acknowledgments: The authors thankFrancoise Cluzeaud, Nicole Fowler-Jaeger, andAnissa Bouhalfaia for helpful technicalassistance and Dr Alain Vandewalle for fruitfuldiscussions.

References

1. Eaton DC, Helms MN, Koval M, Bao HF, Jain L. The contribution ofepithelial sodium channels to alveolar function in health and disease.Annu Rev Physiol 2009;71:403423.

2. Hummler E, Barker P, Gatzy J, Beermann F, Verdumo C, Schmidt A,Boucher R, Rossier BC. Early death due to defective neonatal lungliquid clearance in alpha-ENaC-decient mice. Nat Genet 1996;12:325328.

3. Matthay MA, Folkesson HG, Clerici C. Lung epithelial uid transportand the resolution of pulmonary edema. Physiol Rev 2002;82:569600.

4. Matalon S, OBrodovich H. Sodium channels in alveolar epithelial cells:molecular characterization, biophysical properties, and physiologicalsignicance. Annu Rev Physiol 1999;61:627661.

5. Canessa CM, Horisberger JD, Rossier BC. Epithelial sodium channelrelated to proteins involved in neurodegeneration. Nature 1993;361:467470.

6. Canessa CM, Schild L, Buell G, Thorens B, Gautschi I,Horisberger JD, Rossier BC. Amiloride-sensitive epithelial Na1

channel is made of three homologous subunits. Nature 1994;367:463467.

7. Hummler E, Plane`s C. Importance of ENaC-mediated sodium transportin alveolar uid clearance using genetically-engineered mice. CellPhysiol Biochem 2010;25:6370.

8. Plane`s C, Randrianarison NH, Charles R-P, Frateschi S, Cluzeaud F,Vuagniaux G, Soler P, Clerici C, Rossier BC, Hummler E. ENaC-mediated alveolar uid clearance and lung uid balance dependon the channel-activating protease 1. EMBO Mol Med 2010;2:2637.

Figure 7. Effect of N-acetyl-cysteine (NAC) on hypoxia-induced decrease in transepithelial sodiumtransport. Rat AECs grown for 5 days on Transwell filters were preincubated for 30 minutes with NAC(5 mM, dark gray) or vehicle (light gray) before being exposed to normoxia (21% O2) or tomoderate (1.5% O2) or severe (0.5% O2) hypoxia for 2 hours. Amiloride-sensitive Ieq was measuredunder open-circuit condition immediately at the end of exposure. Results are expressed in mA/cm2

and represent means 6 SE (1018 filters per condition from four independent experiments).***Significantly different from value in normoxic cells treated with vehicle at P , 0.001 (ANOVA andBonferroni).

ORIGINAL RESEARCH


9. Pradervand S, Wang Q, Burnier M, Beermann F, Horisberger JD,Hummler E, Rossier BC. A mouse model for Liddles syndrome. J AmSoc Nephrol 1999;10:25272533.

10. Randrianarison N, Escoubet B, Ferreira C, Fontayne A, Fowler-JaegerN, Clerici C, Hummler E, Rossier BC, Plane`s C. Beta-Liddle mutationof the epithelial sodium channel increases alveolar uid clearanceand reduces the severity of hydrostatic pulmonary oedema in mice. JPhysiol 2007;582:777788.

11. Guney S, Schuler A, Ott A, Hoschele S, Zugel S, Baloglu E, Bartsch P,Mairbaurl H. Dexamethasone prevents transport inhibition byhypoxia in rat lung and alveolar epithelial cells by stimulatingactivity and expression of Na1-K1-ATPase and epithelial Na1

channels. Am J Physiol Lung Cell Mol Physiol 2007;293:L1332L1338.

12. Vivona ML, Matthay M, Chabaud MB, Friedlander G, Clerici C. Hypoxiareduces alveolar epithelial sodium and uid transport in rats: reversalby beta-adrenergic agonist treatment. Am J Respir Cell Mol Biol2001;25:554561.

13. Comellas AP, Dada LA, Lecuona E, Pesce LM, Chandel NS, QuesadaN, Budinger GRS, Strous GJ, Ciechanover A, Sznajder JI. Hypoxia-mediated degradation of Na,K-ATPase via mitochondrial reactiveoxygen species and the ubiquitin-conjugating system. Circ Res2006;98:13141322.

14. Dada LA, Chandel NS, Ridge KM, Pedemonte C, Bertorello AM,Sznajder JI. Hypoxia-induced endocytosis of Na,K-ATPase inalveolar epithelial cells is mediated by mitochondrial reactiveoxygen species and PKC-zeta. J Clin Invest 2003;111:10571064.

15. Mairbaurl H, Mayer K, Kim K-J, Borok Z, Bartsch P, Crandall ED.Hypoxia decreases active Na transport across primary rat alveolarepithelial cell monolayers. Am J Physiol Lung Cell Mol Physiol 2002;282:L659L665.

16. Plane`s C, Escoubet B, Blot-Chabaud M, Friedlander G, Farman N,Clerici C. Hypoxia downregulates expression and activity of epithelialsodium channels in rat alveolar epithelial cells. Am J Respir Cell MolBiol 1997;17:508518.

17. Clerici C, Plane`s C. Gene regulation in the adaptive process to hypoxiain lung epithelial cells. Am J Physiol Lung Cell Mol Physiol 2009;296:L267L274.

18. Plane`s C, Blot-Chabaud M, Matthay MA, Couette S, Uchida T, ClericiC. Hypoxia and beta 2-agonists regulate cell surface expression ofthe epithelial sodium channel in native alveolar epithelial cells. J BiolChem 2002;277:4731847324.

19. Butterworth MB, Edinger RS, Frizzell RA, Johnson JP. Regulation of theepithelial sodium channel by membrane trafcking. Am J PhysiolRenal Physiol 2009;296:F10F24.

20. Eaton DC, Malik B, Bao H-F, Yu L, Jain L. Regulation of epithelialsodium channel trafcking by ubiquitination. Proc Am Thorac Soc2010;7:5464.

21. Li T, Koshy S, Folkesson HG. Involvement of {alpha}ENaC and Nedd42in the conversion from lung uid secretion to uid absorption at birth inthe rat as assayed by RNA interference analysis. Am J Physiol LungCell Mol Physiol 2007;293:L1069L1078.

22. Malik B, Price SR, Mitch WE, Yue Q, Eaton DC. Regulation of epithelialsodium channels by the ubiquitin-proteasome proteolytic pathway.Am J Physiol Renal Physiol 2006;290:F1285F1294.

23. Snyder PM. Minireview: regulation of epithelial Na1 channel trafcking.Endocrinology 2005;146:50795085.

24. Staub O, Rotin D. Role of ubiquitylation in cellular membrane transport.Physiol Rev 2006;86:669707.

25. Abriel H, Lofng J, Rebhun JF, Pratt JH, Schild L, Horisberger JD, RotinD, Staub O. Defective regulation of the epithelial Na1 channel byNedd4 in Liddles syndrome. J Clin Invest 1999;103:667673.

26. Firsov D, Schild L, Gautschi I, Merillat AM, Schneeberger E, RossierBC. Cell surface expression of the epithelial Na channel and a mutantcausing Liddle syndrome: a quantitative approach. Proc Natl AcadSci USA 1996;93:1537015375.

27. Schild L, Canessa CM, Shimkets RA, Gautschi I, Lifton RP, Rossier BC.A mutation in the epithelial sodium channel causing Liddle diseaseincreases channel activity in the Xenopus laevis oocyte expressionsystem. Proc Natl Acad Sci USA 1995;92:56995703.

28. Schild L, Lu Y, Gautschi I, Schneeberger E, Lifton RP, Rossier BC.Identication of a PY motif in the epithelial Na channel subunits asa target sequence for mutations causing channel activation found inLiddle syndrome. EMBO J 1996;15:23812387.

29. Shimkets RA, Warnock DG, Bositis CM, Nelson-Williams C, HanssonJH, Schambelan M, Gill JR Jr, Ulick S, Milora RV, Findling JW.Liddles syndrome: heritable human hypertension caused bymutations in the beta subunit of the epithelial sodium channel. Cell1994;79:407414.

30. Randrianarison N, Clerici C, Ferreira C, Fontayne A, Pradervand S,Fowler-Jaeger N, Hummler E, Rossier BC, Plane`s C. Lowexpression of the beta-ENaC subunit impairs lung uid clearance inthe mouse. Am J Physiol Lung Cell Mol Physiol 2008;294:L409L416.

31. Duc C, Farman N, Canessa CM, Bonvalet JP, Rossier BC. Cell-specicexpression of epithelial sodium channel alpha, beta, and gammasubunits in aldosterone-responsive epithelia from the rat: localizationby in situ hybridization and immunocytochemistry. J Cell Biol 1994;127:19071921.

32. Ronzaud C, Lofng-Cueni D, Hausel P, Debonneville A, Malsure SR,Fowler-Jaeger N, Boase NA, Perrier R, Maillard M, Yang B, et al.Renal tubular NEDD42 deciency causes NCC-mediated salt-dependent hypertension. J Clin Invest 2013;123:657665.

33. Ouiddir A, Plane`s C, Fernandes I, VanHesse A, Clerici C. Hypoxiaupregulates activity and expression of the glucose transporterGLUT1 in alveolar epithelial cells. Am J Respir Cell Mol Biol 1999;21:710718.

34. Plane`s C, Leyvraz C, Uchida T, Angelova MA, Vuagniaux G, HummlerE, Matthay MA, Clerici C, Rossier B. In vitro and in vivo regulation oftransepithelial lung alveolar sodium transport by serine proteases.Am J Physiol Lung Cell Mol Physiol 2005;288:L1099L1109.

35. Goolaerts A, Roux J, Ganter MT, Shlyonsky V, Chraibi A, Stephane R,Mies F, Matthay MA, Naeije R, Sariban-Sohraby S, et al. Serotonindecreases alveolar epithelial uid transport via a direct inhibition ofthe epithelial sodium channel. Am J Respir Cell Mol Biol 2010;43:99108.

36. Hjerpe R, Aillet F, Lopitz-Otsoa F, Lang V, England P, Rodriguez MS.Efcient protection and isolation of ubiquitylated proteins usingtandem ubiquitin-binding entities. EMBO Rep 2009;10:12501258.

37. Malik B, Yue Q, Yue G, Chen XJ, Price SR, Mitch WE, Eaton DC. Roleof Nedd42 and polyubiquitination in epithelial sodium channeldegradation in untransfected renal A6 cells expressingendogenous ENaC subunits. Am J Physiol Renal Physiol 2005;289:F107F116.

38. Chandel NS, McClintock DS, Feliciano CE, Wood TM, Melendez JA,Rodriguez AM, Schumacker PT. Reactive oxygen species generatedat mitochondrial complex III stabilize hypoxia-inducible factor-1aduring hypoxia. J Biol Chem 2000;275:2513025138.

39. Chen X-J, Eaton DC, Jain L. Beta-adrenergic regulation of amiloride-sensitive lung sodium channels. Am J Physiol Lung Cell Mol Physiol2002;282:L609L620.

40. Kimura T, Kawabe H, Jiang C, Zhang W, Xiang Y-Y, Lu C, Salter MW,Brose N, Lu W-Y, Rotin D. Deletion of the ubiquitin ligase Nedd4L inlung epithelia causes cystic brosis-like disease. Proc Natl Acad SciUSA 2011;108:32163221.

41. Boase NA, Rychkov GY, Townley SL, Dinudom A, Candi E, Voss AK,Tsoutsman T, Semsarian C, Melino G, Koentgen F, et al. Respiratorydistress and perinatal lethality in Nedd42-decient mice. NatCommun 2011;2:287.

42. Husted RF, Lu H, Sigmund RD, Stokes JB. Oxygen regulation of theepithelial Na channel in the collecting duct. Am J Physiol RenalPhysiol 2011;300:F412F424.

43. Gusarova GA, Dada LA, Kelly AM, Brodie C, Witters LA, Chandel NS,Sznajder JI. Alpha1-AMP-activated protein kinase regulateshypoxia-induced Na,K-ATPase endocytosis via directphosphorylation of protein kinase C zeta. Mol Cell Biol 2009;29:34553464.

44. Krishnan MN, Ng A, Sukumaran B, Gilfoy FD, Uchil PD, Sultana H,Brass AL, Adametz R, Tsui M, Qian F, et al. RNA interference screenfor human genes associated with West Nile virus infection. Nature2008;455:242245.

ORIGINAL RESEARCH


45. Lazrak A, Iles KE, Liu G, Noah DL, Noah JW, Matalon S. Inuenza virusM2 protein inhibits epithelial sodium channels by increasing reactiveoxygen species. FASEB J 2009;23:38293842.

46. Bhalla V, Oyster NM, Fitch AC, Wijngaarden MA, Neumann D,Schlattner U, Pearce D, Hallows KR. AMP-activated kinase inhibitsthe epithelial Na1 channel through functional regulation of theubiquitin ligase Nedd42. J Biol Chem 2006;281:2615926169.

47. Woollhead AM, Scott JW, Hardie DG, Baines DL. Phenformin and 5-aminoimidazole-4-carboxamide-1-beta-D-ribofuranoside (AICAR)activation of AMP-activated protein kinase inhibits transepithelialNa1 transport across H441 lung cells. J Physiol 2005;566:781792.

48. Almaa J, Kongsuphol P, Hieke B, Ousingsawat J, Viollet B,Schreiber R, Amaral MD, Kunzelmann K. AMPK controlsepithelial Na1 channels through Nedd42 and causes anepithelial phenotype when mutated. Pugers Arch 2009;458:713721.

49. Booth RE, Stockhand JD. Targeted degradation of ENaC in response toPKC activation of the ERK1/2 cascade. Am J Physiol Renal Physiol2003;284:F938F947.

50. Lee J-G, Baek K, Soetandyo N, Ye Y. Reversible inactivation ofdeubiquitinases by reactive oxygen species in vitro and in cells. NatCommun 2013;4:1568.

ORIGINAL RESEARCH


link2externallink2externallink2external

Hypoxia-Induced Inhibition of Epithelial Na1 Channels in the Lung

Documents

Transcript of Hypoxia-Induced Inhibition of Epithelial Na1 Channels in the Lung