Protein kinase A and mitogen-activated protein kinase pathways mediate cAMP induction of α-Epithelial Na + channels (α-ENaC

ORIGINAL ARTICLE 101J o u r n a l o fJ o u r n a l o f

CellularPhysiologyCellularPhysiology

Mitogen-Activated ProteinKinase Pathways Mediate cAMPInduction of a-Epithelial NaR

Channels (a-ENaC)

JEAN A. PETERSHACK, BARBARA M. HENSON, YVONNE M. MENDOZA,

AHSAN CHOUDARY, AND STEVEN R. SEIDNER

Department of Pediatrics/Division of Neonatology, University of Texas Health Science Center, San Antonio, Texas

A major mechanism for Naþ transport across epithelia occurs through epithelial Naþ channels (ENaC). ENaC is a multimeric channelconsisting of three subunits (a, b, and g). The a-subunit is critical for ENaC function. In specific culture conditions, the rat submandibulargland epithelial cell line (SMG-C6) demonstrates minimal Naþ transport properties and exposure to dibutyryl cAMP (DbcAMP) for up to48 h caused an elevation ofa-ENaCmRNA and protein expression and amiloride-sensitive short-circuit current (ISC). Here we examinedthe early signaling pathways evoked by DbcAMP which contribute to the eventual increase in Naþ transport. Treatment with either ofthe protein kinase A (PKA) inhibitors KT5720 or H-89 followed by exposure to 1 mM DbcAMP for 24 h markedly attenuatedDbcAMP-induced a-ENaC protein formation and ISC. Exposure of SMG-C6 cells to 1 mM DbcAMP induced a rapid, transientphosphorylation of the cAMP response element binding protein (CREB). This response was attenuated in the presence of either KT5720or H-89. Dominant-negative CREB decreased DbcAMP-induced a-ENaC expression. Suppression of the extracellular signal-regulatedprotein kinase (ERK 1,2) with PD98059 or the p38 mitogen-activated protein kinase (MAPK) pathway with SB203580 reducedDbcAMP-induced a-ENaC protein levels in SMG-C6 cells. DbcAMP-induced phosphorylation of CREB was markedly attenuated byPD98059 or SB203580. DbcAMP-induced activation of the either the p38 or the ERK 1,2 MAPK pathways was abolished by either of thePKA inhibitors, H-89 or KT5720. Cross talk between these signaling pathways induced by DbcAMP via the activation of CREB appears tocontribute to increased levels of a-ENaC observed after 24 h of treatment in SMG-C6 epithelial cells.J. Cell. Physiol. 215: 101–110, 2008. � 2007 Wiley-Liss, Inc.

Contract grant sponsor: American Heart Association-TexasAffiliate;Contract grant number: 0160116Y.

*Correspondence to: Shamimunisa B. Mustafa, Department ofPediatrics/Division of Neonatology, University of Texas HealthScience Center, 7703 Floyd Curl Drive, San Antonio, TX 78229.E-mail: [email protected]

Received 2 November 2006; Accepted 13 August 2007

DOI: 10.1002/jcp.21291

The major pathway for the movement of Naþ across polarizedepithelial cells occurs through apically located epithelial Naþ

channels (ENaC) (Canessa et al., 1993, 1994). ENaC has beenidentified in the epithelium lining of a variety of tissues includinglung, kidney collecting duct and tubules, colon, bladder, andsalivary ducts (Barbry and Hofman, 1997; Dagenais et al., 1997;Garty and Palmer, 1997). In these tissues net Naþ is absorbedfrom the lumen via ENaC across the apical epithelium and thenactively extruded through basolaterally located Naþ-Kþ-ATPase pumps (Benos et al., 1995; Barbry and Hofman, 1997).For example, fluid clearance at birth from the fetal lung and theresolution of pulmonary edema in mature lungs is coupledto Naþ transport across the alveolar epithelium (O’Brodovich,1991; Bland and Nielson, 1992; Otulakowski et al., 1999;Berthiaume et al., 2002). Also, Naþ transport via ENaC in therenal collecting duct plays a major role in blood pressurecontrol by regulating extracellular Naþ concentration and totalbody fluid (Barbry and Hofman, 1997; Garty and Palmer, 1997).The apicalmembrane expression of ENaC is tightly regulated bythe Nedd-4 (in particular Nedd4-2) family of ubiquitin proteinligases (Snyder et al., 2004; Snyder, 2005).Degradation of ENaCby these proteins provides a means by which Naþ transport isbalanced.

ENaC is a heteromultimer formed by three homologoussubunits, a, b, and g (Canessa et al., 1993, 1994). Whenindependently expressed in Xenopus laevis oocytes or ratthyroid epithelial cells only the a-subunit demonstratesNaþ-dependent current consistent with active Naþ transport(Canessa et al., 1994; Snyder, 2000). When b- or g-ENaC areexpressed either individually with or together witha-ENaC the

� 2 0 0 7 W I L E Y - L I S S , I N C .

resultant Naþ current is greatly elevated (Canessa et al., 1994;Snyder, 2000). The critical importance of a-ENaC requiredfor Naþ transport was further demonstrated by the generationof mice that were genetically deficient for a-ENaC (Hummleret al., 1996). These newborn mice were unable to clear airwayfluid and died within 40 h of birth from respiratory distressdespite having morphologically normal lungs (Hummler et al.,1996). In contrast, b- and g-ENaC knockout mice displayed adelay in lung water clearance after birth, but actually died fromelectrolyte imbalance resulting from abnormal renal function(Giallard et al., 2000).

Fluid clearance following lung injury can be enhanced byb-adrenergic agonists (Matthay et al., 1996; Saldias et al., 2000;McGraw et al., 2001; Berthiaume et al., 2002; Matthay et al.,2002; Sartori et al., 2002). Furthermore, studies in fetal animalsalso demonstrated the effectiveness of b-adrenergic agonists

102 M U S T A F A E T A L .

enhancing lung fluid clearance at the time of birth (Walters andOlver, 1978; O’Brodovich, 1991; Finley et al., 1998). Sincetreatment of newborn lambs with exogenous cAMP increaseslung fluid clearance (Finley et al., 1998), it is postulated that anincrease in maternal circulating catecholamines prior to and atthe time of birth accelerates fetal lung fluid clearance byb-adrenergic receptor stimulation (Walters and Olver, 1978;Tohda et al., 1994; Finley et al., 1998).

Exogenous cAMP regulates amiloride-sensitiveepithelial Naþ transport by increasing a-ENaC expression indistal lung epithelial cells (Dagenais et al., 2001; Thomas et al.,2004). Moreover, in rat thyroid, MDCK, and H441 epithelialcells, cAMP stimulates Naþ transport by increasing thetranslocation of transfected or endogenous ENaC subunits, inparticular a-ENaC, to the apical membrane surface (Snyder,2000; Morris and Schafer, 2002; Thomas et al., 2004). Usingpatch-clamp measurements, cAMP increased the openprobability and number of active Naþ channels in alveolar distallung epithelial cells (Chen et al., 2002). Furthermore, briefexposure of rat alveolar epithelial type II cells to terbutalineincreased apical expression of ENaC subunits without alteringENaC mRNA and protein expression (Planes et al., 2002). Inaddition, activation of cAMP-dependent protein kinaseA (PKA)has been shown to phosphorylate Nedd4-2, thus decreasingNedd4-2 attenuation of ENaC function (Snyder et al., 2004).Collectively, these independent studies indicate that theregulatory effects of exogenous cAMP on ENaC are exerted atdifferent stages; up regulation of each subunit expression,translocation to and increased expression at the apicalmembrane, and ultimately increased Naþ transport in differentepithelial cells. Given the wide range of regulatory modalitiesthat cAMP exerts on ENaC expression and consequently, Naþ

transport, we investigated the early signaling components thatare activated prior to these functional changes, followingexposure to cAMP in epithelial cells. In the present study weutilized the immortalized rat submandibular gland (SMG-C6)cell line (Quissell et al., 1997; Castro et al., 2000). WhenSMG-C6 cells are cultured in the absence of hydrocortisone,a-ENaCmRNA and protein are expressed at neglible levels andminimal amiloride-sensitive transepithelial Naþ transport ispresent. Treatment of SMG-C6 epithelial cells with exogenousdibutyryl cAMP (DbcAMP) for 48 h increased a-ENaC mRNAexpression, protein formation, and amiloride-sensitivetransepithelial Isc. We also found that following exposure toDbcAMP, PKA, and PKA-dependent activation of p38 and ERK1,2 MAPK played an important role in the onset of CREBphosphorylation. Furthermore, dominant-negative CREBdecreased a-ENaC mRNA and protein expression. Weconclude that early cross-talk between multiple signalingpathways all contribute toward DbcAMP-dependent upregulation of a-ENaC in SMG-C6 epithelial cells.

Materials and MethodsReagents

Dulbecco’s modified Eagle’s medium (DMEM), Ham F-12 media,fetal bovine serum (FBS), and trace element mix were obtainedfrom Gibco (Grand Island, NY). Transferrin and epidermal growthfactor (EGF) were obtained from JRH Biosciences (Lenexa, KS).Insulin was obtained from Novo Nordisk Pharmaceuticals(Princeton, NJ). Polycarbonate Snapwell1 filters were obtainedfrom Costar (Cambridge, MA). Amiloride, actinomycin D, andDbcAMP were purchased from Sigma Chemical Co. (St. Louis,MO). KT5720, H-89, PD98059, and SB203580 were all purchasedfrom Calbiochem (San Diego, CA). The anti-rat a-ENaC antibodywas obtained fromAffinity BioReagents (Golden, CO). The anti-ratb- and g-ENaC antibodies were obtained from Chemicon(Temecula, CA). All other antibodies used in this study werepurchased from Santa Cruz Biotechnology (Santa Cruz, CA).

JOURNAL OF CELLULAR PHYSIOLOGY

Dominant negative A-CREB was a generous gift fromDavid D. Ginty (Johns Hopkins University School of Medicine,Baltimore, MD).

Cell culture

SMG-C6 epithelial cells cultured in the absence ofhydrocortisone were seeded at a density of 2.5� 105 cells/cm2

on Snapwell1 tissue culture treated filters (diameter 12 mm, poresize 0.4 mm) coated with 1.0 mg/cm2 of human collagen type-I(Becton Dickinson, Franklin Lakes, NJ). The culture medium usedwas DMEM/F12 (1:1 mixture) with 2.5% FBS supplemented withtrace element mix, 5 mg/ml transferrin, 0.1 mM retinoic acid,2.0 nM T3, 5 mg/ml insulin, 80 ng/ml EGF, 100 U/ml penicillin, and100 mg/ml streptomycin (Castro et al., 2000).

Transepithelial transport studies

Transepithelial resistances (Rt) of the SMG-C6 monolayersgrown on permeable filter supports were measured daily followingseeding utilizing an EVOM-G, World Precision Instrument(Sarasota, FL). Filters exhibiting a Rt> 600V cm2 were placed in amodified Ussing chamber (Costar) equipped with Ag/AgClelectrodes and the apical and basolateral reservoir was bathedwithKreb’s-Ringer bicarbonate (KRB) media containing (in mM):120 NaCl, 2.5 K2HPO4, 0.6 KH2PO4, 1.2 CaCl2, 1.2 MgCl2, 20NaHCO3, and 10 glucose at pH 7.4. The media were air-lifted with5%CO2/21%O2 and temperature was maintained at 378C. Using aDVC-1000 voltage/current clamp with automatic fluid resistancecompensation, stable baseline potential difference (PD) wasmeasured. After equilibration (10–15 min), the monolayer’s PDwas clamped to 0 mV and short-circuit current (Isc) wascontinuously recorded. Every 30 sec, the monolayer was clampedto 1 mV for 0.5 sec so that measured changes in Isc enabledcalculation of Rt using Ohm’s law (Rt¼DPD/DIsc). Reported Iscvalues refer to the movement of positive charges from the apical tothe basolateral side of the membrane, and the PD is referenced tothe apical side (negative relative to the basolateral side). Afterreaching steady state Isc values, amiloride, an ENaC inhibitor(10 mM), was instilled in the apical reservoir and the change incurrent was measured as the amiloride-sensitive Isc. Amiloride wasdissolved in DMSO with the final dilution in the Ussing chamberfluid being 1:1,000. This amount of DMSO did not alter bioelectricproperties and previous studies established the concentrations ofamiloride where maximal effect was observed without altering Rt,indicating that changes in transepithelial Isc was a direct effect of thespecific agent studied. Neither basal bioelectric properties nor theresponse to amiloride varied following each cell passage.

Northern blot analysis

Following specific treatments total RNA was isolated fromcultured SMG-C6epithelial cells usingTRIzol reagent,GIBCO,BRL(Gaithersburg, MD) according to the manufacturer’sinstructions. RNA (20mg)was size-fractionated by electrophoresison a 0.8% agarose/2.2 M formaldehyde gel under denaturingconditions and transferred using a PosiBlot, Stratagene (La Jolla,CA) onto a Magna nylon membrane, Microns Separations, Inc.(Westborough, MA). RNA was immobilized by ultravioletcross-linking. The cDNA probe used in this study was a full-lengthrat a-ENaC cDNA (a kind gift from Dr. H. O’Brodovich, TheHospital for Sick Children, Toronto, Canada). Probes were labeledwith a multiprime DNA labeling system using [a-32P] dCTP(specific activity, 3,000 Ci/mmol). Northern blot hybridizationswere performed in 50% formamide, 1MNaCl, 10% dextran sulfate,50 mM Tris–HCl pH 7.5, 0.1% sodium pyrophosphate, and0.2% Denhardt’s solution at 428C for 16 h. The membranes werewashed once in 2� SSC, 0.1% SDS at room temperature for 10minand then in 2� SSC, 0.1% SDS at 658C for 5 min. Radioactivity wasvisualized using a PhosphorImager, Molecular Dynamics(Sunnyvale, CA). Control hybridizations were performed using a

U P - R E G U L A T I O N O F a - E N a C B Y D b c A M P 103

[a-32P] end-labeled oligonucleotide complimentary to rat 18SrRNA. Signal intensity was quantified by densitometry.

Western blot analysis

Following treatment of cultured SMG-C6epithelial cellswith eitherDbcAMP and/or specific inhibitors for different periods of time, themedium was removed and the cells were homogenized in lysisbuffer (Tris–HCl 50 mM, pH 7.4, TBS) containing 5 mM EDTA,5 mM EGTA, leupeptin (1 mM), pepstatin A (1 mM), aprotonin(1 mM), and phenylmethylsulphonyl fluoride (1 mM). Samplesunderwent a freeze-thaw procedure three times to ensure that thecells were lysed. Protein concentrations were determined usingthe BioRad DC protein assay, BioRad (Hercules, CA), with BSAused as the standard. About 50–60 mg of total protein wassubjected to SDS–PAGE. The separated proteins were transferredelectrophoretically to PVDF membranes, Millipore (Bedford, MA)using a semi-dry transfer blot system and blocked in TBS containing5% non-fat dried milk powder for 1 h and then incubated withprimary antibody in TBS containing 0.01% Tween-20 and5% non-fat dried milk powder for 24 h. The blots were thenincubated with secondary antibody conjugated to horseradishperoxidase appropriately diluted in the same buffer for 1 h.Peroxidase labeled proteins were visualized using an enhancedchemiluminescence (ECL) assay kit from Amersham PharmaciaBiotech. (Little Chalfont, England). The anti-rat-a-ENaC antibodywas obtained from Affinity BioReagents and was directed againstthe L20-C42 NH2 terminus portion of the rat a-ENaC subunit.This antibody detects an approx. 97 kDa protein representing theglycosylated form of rat a-ENaC. The anti-rat-b- and g-ENaCantibodies were purchased from Chemicon (Temecula, CA). Theanti-rat-b-ENaC antibody was directed against 20 residues nearthe NH2 terminus. The anti-rat g-ENaC antibody was directedagainst 14 residues near the NH2 terminus. Both polyclonal ratantibodies recognize bands between 83 and 90 kDa. The b-actinantibody was obtained from Sigma. All other antibodies used in thisstudywere purchased from SantaCruz Biotechnology. The relativeintensities of the bands were quantified by densitometry usingthe NIH imaging program. In certain experiments blots werere-probed for b-actin to confirm equal sample loading.

Lipid-mediated transfection of SMG-C6 cells

To study the role of the transcription factor CREB on a-ENaCexpression following treatment with DbcAMP, SMG-C6 cells weretransfected with dominant negative CREB (A-CREB) usingLipofectamine 2000, Invitrogen (Carlsbard, CA) according to themanufacturer’s instructions. Cells were transfected uponreaching 60–70% confluence. Lipofectamine 2000 was complexedwith 1 mg/ml of A-CREB cDNA in OptiMEM medium for 1 h andthen added to the cells. After 3 h, cells were treated with DbcAMPfor 24 h. The concentration of A-CREB (1 mg/ml) chosen fortransfection purposes in our experiments was based on a previousstudy by Ahn et al. (1998) indicating that this concentration ofA-CREB was sufficient to block cAMP-induced gene expression.

Statistical analysis

All data are presented as means� SE. All experiments wereperformed at least three times unless otherwise stated.Significance between groups was assessed by ANOVA followed byStudent’s t-test. A P-value of <0.05 was considered statisticallysignificant.

ResultsEffect of DbcAMP on a-ENaC expressionand ISC in SMG-C6 epithelial cells

Immunoblotting with the rabbit polyclonal anti-a-ENaCantibody detected a-ENaC protein at 94–97 kDa in SMG-C6whole cell lysates and rat lung homogenates, similar to that

JOURNAL OF CELLULAR PHYSIOLOGY

previously reported byDagenais et al., (2003) (Fig. 1A, left part).The specificity of the protein band was confirmed by repeatingthe experiment in the presence of the immunizing peptide andthe intensity of the resultant band was significantly diminished(data not shown). Western blotting showed that 1 mMDbcAMP increased a-ENaC protein levels 2.4- and 3.2-fold at24 and 48 h, respectively, compared to control cells (denotedby C in all figures, Fig. 1A, right part). Additionalimmunoblotting with the rabbit polyclonal anti-b-ENaCantibody and the anti-g-ENaC antibody detected the respectiveENaC subunit protein bands between 83 and 90 kDa inSMG-C6whole cell lysates and rat lung homogenates (Fig. 1B,C,left parts). Analysis of cells treated for 24 hwith 1mMDbcAMPdid not show any significant changes in the endogenousprotein levels of b- and g-ENaC (Fig. 1B,C, right parts). In theabsence of DbcAMP, control SMG-C6 monolayersexhibited minimal transepithelial amiloride-sensitive ISCconsistent with the limited a-ENaC protein levels. However, inthe presence of 1 mM DbcAMP an incremental elevation inamiloride-sensitive ISC over the same time frame wasobserved (Fig. 1D).

a-ENaC mRNA levels were minimally expressed innon-treated SMG-C6 epithelial cells in contrast to the higherdegree of expression in rat lung tissue (Fig. 2A). Exposure toincreasing concentrations of DbcAMP for 24 h caused anelevation in a-ENaC mRNA expression only apparent at thehighest concentration (1 mM, Fig. 2A). Exposure of culturedSMG-C6 epithelial cells to 1 mMDbcAMP over a period of 48 hresulted in a time-dependent increase in a-ENaC mRNAexpression when compared to control cells (Fig. 2B).Treatment of SMG-C6 epithelial cells with the transcriptionalinhibitor actinomycin D together with 1 mM DbcAMP for 24 hablated the DbcAMP-induced increase in a-ENaC mRNAexpression suggestive that the effects of DbcAMP on a-ENaCexpression may occur at the level of gene transcription(Fig. 2C). As shown above a-ENaC mRNA expression incontrol cells is barely detectable and increased expression byDbcAMP is detected at 24 h, thus it is possible thatDbcAMPnotonly induces gene transcription but may also affect the half-lifeof a-ENaC mRNA. To this end, cells were treated withDbcAMP for 24 h and then actinomycin Dwas added to controland DbcAMP treated-cells. RNA was then isolated after 0, 2, 4,6, and 8 h and assessed for a-ENaC mRNA content. As before,a-ENaC expression in control cells after 24 hwas negligible andremained so even after the addition of actinomycin D (data notshown). However, the calculated half-life of a-ENaC mRNA incells treated with DbcAMPwas found to be approximately 17 h(Fig. 2D).

Activation of PKA is a prerequisite for DbcAMP-induceda-ENaC protein expression and function

Elevation of intracellular cAMP levels results in the activation ofPKA which in turn phosphorylates other key proteins resultingin numerous cellular responses (Richards, 2001; Schwartz,2001). To confirm that activation of PKA is involved inDbcAMP-mediated increases in a-ENaC protein, SMG-C6epithelial cells were exposed to either of two PKA inhibitors,KT5720 (0.5 mg/ml) or H-89 (10 mM), for 1 h and thentreated with 1 mM DbcAMP for 24 h. As shown inFigure 3A, DbcAMP-mediated a-ENaC protein formationwas significantly reduced in the presence of either PKAinhibitor compared to cells exposed to DbcAMP alone.DbcAMP-induced amiloride-sensitive ISC was also markedlydecreased in the presence of either PKA inhibitor after 24 h(Fig. 3B). These findings strongly suggest that activation of thePKA signaling pathway is an important component necessaryfor the stimulation of a-ENaC protein formation and

Fig. 1. A (left part): A typical blot showing untreated SMG-C6 whole cell lysates and rat lung homogenates that were subjected to Western blotanalysis using a polyclonal anti-rat a-ENaC antibody. A (right part): A representative image and corresponding densitometric analysisshowing a-ENaC protein in whole cell lysates after exposure to DbcAMP for the times indicated. Results are expressed as meanWSE (nU 3–4determinationsateachtimepoint).MP< 0.05versuscontrolcells.B,C(leftparts):TypicalblotsshowinguntreatedSMG-C6wholecell lysatesandratlunghomogenates thatwere subjected toWesternblot analysisusing polyclonal anti-ratb- andg-ENaCantibodies, respectively.B,C (rightparts):Cultured SMG-C6 epithelial cells were treated with 1 mM DbcAMP for 24 h after which total protein was isolated. The levels of each of b- andg-ENaC subunits were detected by Western blot analysis using the relevant antibodies. A representative of blots from two similar experiments isshown. D: A representative time-course experiment showing transepithelial ISC in SMG-C6 epithelial cells that were treated with 1 mM DbcAMPover a period of 48 h. Amiloride (10 mM) was added to the apical bathing solution. Data are expressed as meanWSE. nU 4–9 independentobservations.

104 M U S T A F A E T A L .

consequent ENaC function in SMG-C6 epithelial cells followingexposure to DbcAMP for 24 h.

Activation of CREB by DbcAMP; phosphorylationand effect on a-ENaC expression

cAMP-mediated activation of PKA regulates gene expressionpartly through a conserved cAMP-response element (CRE) onthe promoter sequence of target genes (Gonzalez andMontminy, 1989; Kwok et al., 1994; Habener et al., 1995; DuandMontminy, 1998;Hansen et al., 1999). This regulatory effectis mediated via a 43 kDa CRE binding protein (CREB) whichfunctions as a transcription factor (Gonzalez and Montminy,1989; Kwok et al., 1994; Habener et al., 1995). Phosphorylationof CREB on the serine residue at 133 (Ser-133) is an importantstep prior to initiating transcription of target genes and alsoincreases its transcriptional activity (Gonzalez and Montminy,1989; Kwok et al., 1994). To confirm that phosphorylation ofCREB occurs in SMG-C6 epithelial cells following stimulationwith DbcAMP, Western blot analysis was performed using anantibody recognizing the CREB moieties phosphorylated atSer-133 (phospho-CREB antibody). Exposure of SMG-C6epithelial cells to 1 mM DbcAMP for 60 min resulted in thetransient phosphorylation of CREB which peaked after 10 minof stimulation (Fig. 4A). Pretreatment of SMG-C6 epithelial cells

JOURNAL OF CELLULAR PHYSIOLOGY

with either PKA inhibitorsKT5720 (0.5mg/ml) orH-89 (10mM)for 1 h followed by the addition of DbcAMP for 10 minsignificantly reduced DbcAMP-induced phosphorylation ofCREB protein indicating that phosphorylation of CREB atSer-133 occurs via a PKA-dependent pathway in SMG-C6epithelial cells (Fig. 4B) and is also an important component inthe signaling pathway involved in DbcAMP-induced upregulation of a-ENaC (Fig. 3A,B).

To further evaluate the contribution of CREB during theinduction of a-ENaC, SMG-C6 cells were transientlytransfected using the Lipofectamine 2000 method withA-CREB, a dominant negative inhibitor of CREB (Ahn et al.,1998). This particular construct has been reported to inhibitwild-type CREB from binding to DNA (Ahn et al., 1998).SMG-C6 cells treated with the transfection media(Lipofectamine, Lipo) alone for 24 h did not induce expressionof either a-ENaC mRNA or protein (Fig. 4C,D). In addition,earlier studies in our laboratory showed that neitherlipofectamine alone nor transfection with an empty vector hadany effect onDbcAMP-induced amiloride-sensitive ISC (data notshown). However, there was a 1.2-fold decrease in a-ENaCmRNA expression in DbcAMP-treated cells over expressingA-CREB compared to cells exposed toDbcAMP alone (Fig. 4C).Further studies showed that DbcAMP-induced a-ENaCprotein expression was significantly attenuated in cells over

Fig. 2. Effect of DbcAMP on ENaC expression and mRNA stability. In (A–D) total RNA was isolated as described in the Materials and Methodssection and subjected to Northern blot hybridization with a 32P-labeled a-ENaC cDNA probe (upper). Normalization for RNA loading isdemonstrated by hybridization with 18S rRNA (lower). A: Confluent SMG-C6 epithelial cell monolayers were treated for 24 h with increasingconcentrationsofDbcAMP.B:SMG-C6epithelialcellsweretreatedfor24and48hwith1mMDbcAMP.C:SMG-C6epithelialcellswerepretreatedwith actinomycin D (1 mM) for 1 h followed by exposure to 1 mM DbcAMP for 24 h. D: Confluent monolayers of SMG-C6 cells were exposed toDbcAMPfor24h.ThesecellswerethenfurthertreatedwithactinomycinD(1mM)andRNAwasharvestedafter0,2,4,6,and8h.Combinedfindingsfrom three separate experiments are depicted. Time points on graph represent means ofa-ENaC mRNA relative to 0 time point, that is,a-ENaCmRNA level after 24 h which was made to equal 100.

U P - R E G U L A T I O N O F a - E N a C B Y D b c A M P 105

expressing A-CREB after 24 h compared to cells stimulatedwith DbcAMP (Fig. 4D).

Involvement of MAPK pathways in DbcAMP-induceda-ENaC protein formation in SMG-C6 epithelial cells

Elevated levels of intracellular cAMP in addition toactivating PKA, can also exert its biological effects via themitogen-activated protein kinases (MAPKs) signaling pathwaysin a cell-specific manner (Vossler et al., 1997; Impey et al., 1998;Hansen et al., 1999; Schaeffer and Weber, 1999; Busca et al.,2000; Liebmann, 2001; Richards, 2001). The levels ofprotein expression of the extracellular-regulated kinases(ERK 1,2 kinase) and p38 MAPK in SMG-C6 epithelial cellstreated with 1 mM DbcAMP over a period of 240 min wereassessed by Western blot analysis, utilizing the correspondingphosphospecific and non-phosphospecific antibodies. InSMG-C6 epithelial cells elevated levels in the phosphorylationof ERK1,2 kinase (Fig. 5A) and p38MAPK (Fig. 5B) compared tocontrol cells were detectedwithin 5–10min of stimulationwithexogenous DbcAMP (1 mM). By 120 min following DbcAMPstimulation, phosphorylation of ERK 1,2 kinase and p38 MAPKreturned to basal levels (Fig. 5A,B).

To assess the role of ERK 1,2 kinase and p38 MAPK inDbcAMP-induced a-ENaC protein formation, SMG-C6epithelial cells were pretreated for 1 h with either PD98059(5mg/ml), an ERK 1,2 kinase inhibitor, or SB203580 (1mg/ml), ap38MAPK inhibitor followed by treatmentwith 1mMDbcAMPfor 24 h. As depicted in Figure 5C, DbcAMP-induced a-ENaCprotein expressionwas significantly suppressed in the presenceof either PD98059 or SB203580.

MAPK pathways mediate DbcAMP-induced CREBphosphorylation in SMG-C6 epithelial cells

We observed earlier that phosphorylation of CREB atSer-133 by PKA is an important component of the signalingpathway up regulatinga-ENaC levels in SMG-C6 epithelial cells.Since different protein kinases also phosphorylate CREB (Du

JOURNAL OF CELLULAR PHYSIOLOGY

and Montminy, 1998; Maizels et al., 1998; Liebmann, 2001), weinvestigated whether the MAPK pathways were involved inDbcAMP-induced CREB phosphorylation. Cells werepretreated with the MAPK inhibitors for 1 h and thenstimulated with DbcAMP for 10 min. DbcAMP-dependentCREB phosphorylation was significantly suppressed by the ERK1,2 kinase inhibitor, PD98059 (Fig. 6A) and also by the p38MAPK inhibitor, SB203580 (Fig. 6B). When SMG-C6 epithelialcells were treated simultaneously with both inhibitors followedby exposure to DbcAMP for 10 min, there was a greaterreduction (3.6-fold) of CREB phosphorylation compared tothat observed with the individual inhibitors (Fig. 6C). Theseobservations suggest that the ERK1,2 kinase and the p38MAPKpathways are both important participants in the regulatorysignaling pathways of DbcAMP-induced a-ENaC proteinexpression via the phosphorylation of CREB in SMG-C6epithelial cells.

Inhibition of PKA also suppresses activation of MAPK

Following activation of PKA by exogenous cAMP, in addition tophosphorylating CREB, PKA can either inhibit or activate othersignaling pathways such as the MAPK pathways (Vossler et al.,1997; Impey et al., 1998; Richards, 2001). To further addressthis phenomenon, cultured SMG-C6 epithelial cells wereseparately pretreated with the PKA inhibitors H-89 (10mM) orKT5720 (0.5 mg/ml) for 1 h followed by exposure to 1 mMDbcAMP for 10 min. As shown in Figure 7A, DbcAMP-inducedactivation of p38 MAPK pathway was significantly attenuated inthe presence of either H-89 or KT5720 as compared toDbcAMP-treated cells. Similarly, DbcAMP-dependent upregulation of P-ERK 1,2 was markedly decreased in thepresence of either H-89 or KT5720 as compared to DbcAMP-treated cells (Fig. 7B).

Discussion

In the present study we report that treatment of SMG-C6epithelial cells with exogenous DbcAMP for up to 48 h

Fig. 3. Effect of PKA inhibitors on DbcAMP-mediated a-ENaCprotein expression and ENaC function. A: Cultured SMG-C6epithelial cells were treated with either KT5720 (0.5 mg/ml) orH-89 (10 mM) followed by 1 mM DbcAMP for 24 h. Whole cell lysateswere subjected to Western blot analysis using a polyclonal a-ENaCantibody. Densitometric analysis of the corresponding blots andmean valuesWSE of three independent experiments are shown.MP< 0.05 versus control cells and MMP< 0.05 versus DbcAMP-treatedcells. B: A representative plot of transepithelial ISC in SMG-C6epithelial cells subjected to KT5720 (0.5 mg/ml) or H-89 (10 mM)followed by 1 mM DbcAMP for 24 h. Amiloride (10 mM) wasadded to the apical bathing solution. Data are expressed asmeanWSE. nU 3–6 independent experiments.

106 M U S T A F A E T A L .

increased a-ENaC mRNA expression, protein formation, andamiloride-sensitive transepithelial Isc similar to previouslydescribed studies which used other tissue-derived epithelialcells (Dagenais et al., 2001; Chen et al., 2002; Thomaset al., 2004). We also showed that DbcAMP-dependentincrease in a-ENaC mRNA was ablated in the presence ofactinomycin D. Furthermore, we calculated that the half-life ofa-ENaC mRNA induced by DbcAMP in SMG-C6 cells wasapproximately 17 h. An earlier study by Dagenais et al. (2001)reported that DbcAMP-induced a-ENaC mRNA has a half-lifeof 13.8 h in rat alveolar epithelial cells versus 15.1 h in controlcells. In contrast to this study we were unable to measure thehalf-life of endogenousa-ENaCmRNA in control SMG-C6 cellsas it was extremely low or in some cases undetectable. At thispoint, we can suggest that due to a comparatively long half life inSMG-C6 cells, DbcAMP-induced increases in a-ENaC mRNAexpression probably occurs at the level of gene transcription.

In contrast to the rapid increase in amiloride-sensitivetransepithelial Isc following the addition of DbcAMP to H441cells, we measured increases in ENaC expression and functionafter 24 h (Thomas et al., 2004). The very rapid rise in Isc after5 min of stimulation with cAMP with H441 cells may be due tothe higher basal levels of a-ENaC protein present compared to

JOURNAL OF CELLULAR PHYSIOLOGY

the SMG-C6 cell line (Thomas et al., 2004). Thus, we speculatethat in SMG-C6 epithelial cells the increase in ENaC functionafter 24 h as measured by amiloride-sensitive transepithelial Iscis critically dependent upon newly formed a-ENaC subunits.Decreased degradation by NEDD 4-2 might further potentiatethe levels and function of these newly formed subunits to alesser degree (especially during the further increase in a-ENaCexpression and function from 24 to 48 h). With higher basalexpression of ENaC, cAMP-dependent inhibition of Nedd4-2degradation was shown to occur due to increasedphosphorylation of Nedd4-2 by PKA, thus prolonging ENaCexpression and function in epithelial cells transfected with eachof the ENaC subunits (Snyder et al., 2004).

In untreated SMG-C6 epithelial cells endogenous b- andg-ENaC protein levels were detectable by Western blotanalysis. In contrast to effects on primary cultures of alveolarepithelial cells, DbcAMP did not stimulateb-org-ENaC proteinlevels in SMG-C6 epithelial cells (Dagenais et al., 2001).However, expression of b- and g-ENaC mRNA were alsoreported to be unchanged after forskolin exposure in H441epithelial cells (Thomas et al., 2004). These differences couldeither be due to cell specific responses or to a relative excess ofb- and g-ENaC subunits when there is lower basal expressionofa-ENaC. It appeared that the endogenous levels of theb- andg-ENaC subunitswere sufficient in our studies to associatewiththe newly generated a-ENaC subunits, forming functionalchannels, and displaying an increase in amiloride-sensitivetransepithelial Isc following treatment with DbcAMP (Fig. 1D).As suggested by an earlier study (Jain et al., 1999), it is alsoworthy to consider here that following treatment withDbcAMP there maybe additional ENaC in SMG-C6 cellscomposed of just the a-subunits.

Since SMG-C6 epithelial cells have little to noendogenous Naþ transport, several of the early intermediarysignaling pathways are revealed with exogenous cAMP-stimulated ENaC expression. Our present findings indicate thatDbcAMP-dependent upregulation of a-ENaC proteinformation involves the activation of at least three differentintracellular signaling pathways: (1) PKA-dependent pathwaythat can be suppressed by the PKA-specific inhibitors H-89 andKT5720; (2) p38 MAPK pathway that can be attenuated by theinhibitor SB203580; and (3) ERK 1,2 kinase pathway that can bedepressed in the presence of PD98059.

Activation of PKA byDbcAMP in SMG-C6 epithelial cells wasseen to be an important step for the increase ina-ENaCproteinand amiloride-sensitive transepithelial Isc as the specific PKAinhibitors KT5720 and H-89 clearly attenuated both theseresponses. PKA, activated by cAMP mediate specificintracellular signaling events including the transcription ofspecific genes via the phosphorylation of CREB (Gonzalez andMontminy, 1989; Kwok et al., 1994; Habener et al., 1995;Maizels et al., 1998; Hansen et al., 1999; Richards, 2001).Phosphorylation of CREB at Ser-133 allows it to complex withother proteins such as the CREB-binding protein (CBP), atranscriptional co-activator, thus resulting in the transcriptionof several genes having putative CRE site in their promoterregions, for example, mouse a-ENaC promoter region(Dagenais et al., 2001). In the present study we observed thatthe transient phosphorylation of CREB in SMG-C6 epithelialcells in the presence of DbcAMP peaked after 10 minof stimulation and then declined by 60 min (Fig. 4A). We havedemonstrated that the PKA inhibitors, KT5720 and H-89 alsomarkedly reduced the phosphorylation of CREB in thepresence of DbcAMP. In addition, we observed that in thepresence of a dominant-negative construct of CREB, A-CREB,a-ENaCmRNAexpression, and protein levels were decreased,but not entirely abolished. An earlier study by Otulakowskiet al. (1999), suggested that no consensus cAMP responsiveelement was detected in the rat a-ENaC promoter. Since we

Fig. 4. DbcAMP-dependent activation of CREB; phosphorylation and effect on a-ENaC expression. A: A representative Western blot analysisshowing the phosphorylation of CREB in SMG-C6 epithelial cells that were treated with 1 mM DbcAMP for the indicated times. Data arerepresentative of three independent experiments. B (upper part): SMG-C6 epithelial cells were either treated with KT5720 (0.5 mg/ml) or H-89(10 mM) respectively, followed by the addition of 1 mM DbcAMP for 10 min. Phosphorylation of CREB was determined by Western blot analysisusing phosphospecific CREB antibody. Lower part, densitometric analysis of experiments performed as described above. Results are expressed asmeanvaluesofWSEof fourseparateexperiments.MP< 0.05versusDbcAMP-treatedcells.C:ConfluentSMG-C6 cellsweretransiently transfectedwith dominant negative A-CREB as described in the Materials and Methods section of this paper and then stimulated with 1 mM DbcAMP for 24 h.Total RNA was isolated and subjected to Northern blot hybridization with a 32P-labeled a-ENaC cDNA probe (upper). Normalization for RNAloading is demonstrated by hybridization with 18S rRNA (lower). A typical image from three independent experiments is depicted. D: Arepresentative Western blot prepared from cell lysates from four independent experiments of SMG-C6 cells treated as described above forpart C was probed with anti-rat a-ENaC antibody. Results are expressed as mean values ofWSE of four separate experiments. MP< 0.05 versusDbcAMP-treated cells.

Fig. 5. Activation of the MAPK signaling pathways by DbcAMP and the effect of MAPK inhibitors on DbcAMP-mediated a-ENaC formation incultured SMG-C6 epithelial cells. Whole cell lysates previously treated with exogenous DbcAMP (1 mM) for the indicated times were subjected toWestern blot analysis using phospho-specific antibodies (A,B, upper) to extracellular signal-regulated kinase (ERK 1,2, A) and p38 MAPK (B). Ascontrols, thesamecell lysatesweresubjectedtoWesternblotanalysisusingcorrespondingnon-phospho-specificantibodies (A,B, lower)todetecttotal ERK 1,2 (A) and total p38 (B). Data are representative of three similar experiments. C: SMG-C6 epithelial cells were pretreated with orwithout PD98059 (5 mg/ml) or SB203580 (1 mg/ml) for 1 h and then incubated with 1 mM DbcAMP for 24 h. Whole cell lysates were subjected toWestern blot analysis using anti-a-ENaC antibody. Also shown is the densitometric analysis of corresponding blot. Mean values of 4–5determinationsWSE are shown. MP< 0.05 versus DbcAMP-treated cells.

JOURNAL OF CELLULAR PHYSIOLOGY

Fig. 6. EffectsofMAPKinhibitorsonDbcAMP-mediatedCREBphosphorylation.CulturedSMG-C6epithelialcellswerepreincubatedwitheitherPD98059(5mg/ml,A)orSB203580(1mg/ml,B)orPD980595mg/ml, togetherwithSB2035801mg/ml, (C) for1handthenstimulatedwithDbcAMP(1 mM) for 10 min. Phosphorylation of CREB was determined by Western blot analysis using phospho-specific CREB antibody. The blots werere-probed with anti-b-actin to confirm equal loading of the samples, lower part. Results are expressed as meanWSE. nU 4–5 independentexperiments. MP< 0.05 versus DbcAMP-treated cells.

108 M U S T A F A E T A L .

did not detect a significant increase in a-ENaC expression until24 h after exposure to DbcAMP, it is extremely probable thatother intermediary proteins/transcription factors areupregulated byCREB and that these proteins in turn, initiate thetranscription of a-ENaC mRNA in SMG-C6 epithelial cells.

Increased intracellular cAMP levels can regulate othersignaling pathways such as the MAPKs in a cell- andtissue-specific manner (Vossler et al., 1997; Hansen et al., 1999;Busca et al., 2000; Liebmann, 2001; Richards, 2001).Intermediary cAMP-binding proteins such as the cAMP-guanineexchange factors (cAMP-GEFs) or exchange proteins activatedby cAMP (Epac) may provide an alternative pathway for cAMPto stimulate other signaling cascade pathways in addition toactivating PKA (Richards, 2001). The MAPK cascade pathwaysregulate gene expression, cell proliferation, and survival inmammalian cells (Davis, 1994; Schaeffer andWeber, 1999). TheMAPK family consists of three subclasses, each with multiplemembers: the extracellular-regulated kinases (ERK 1,2), the Junamino-terminal-kinases/stress-activated kinases (JNKs/SAPKs), and the p38 MAPKs. Each MAPK is a member of athree-protein kinase cascade and are initially activated by aMAPK kinase kinase (MKKK) followed by a MAPK kinase(MKK) (Schaeffer and Weber, 1999). Stimulation of the small,monomeric Ras and Raf proteins by a variety of agonists,including cAMP initiate the signaling cascades of the ERK1,2 andp-38 MAPK pathways (Vossler et al., 1997; Busca et al., 2000;Liebmann, 2001).

Our present findings demonstrate that DbcAMP rapidlystimulated the transient phosphorylation of ERK 1,2 in atime-dependent manner. Also, in the presence of PD98059, aspecific inhibitor of ERK 1,2, DbcAMP-dependent increases in

JOURNAL OF CELLULAR PHYSIOLOGY

a-ENaC protein were noticeably suppressed. Activation ofERK 1,2 by 12-0-tetradecanoyl-1-phorbol-13-acetate (TPA) inthe rat parotid epithelial cell line (PAR-C5) down-regulatedendogenously expressed a-ENaC mRNA; a phenomenonrepressed in the presence of PD98059 (Zentner et al., 1998).SMG-C6 epithelial cells and PAR-C5 epithelial cells are bothderived from the acinar regions of the rat salivary gland(Quissell et al., 1997). In agreement with Zentner andco-workers we have also demonstrated that cultured PAR-C5epithelial cells abundantly express endogenousa-ENaCproteinbut did not demonstrate amiloride-sensitive Isc in the UssingChamber (S.B. Mustafa and R. Castro, unpublishedobservations). The noticeable lack of ENaC function in PAR-C5epithelial cells in contrast to that observed in SMG-C6 epithelialcells remains to be elucidated; however, these findings may bedue to the apparent differences in the regulatory effects ofactivated ERK1,2MAPKona-ENaCexpression between theseparticular two salivary epithelial cell lines.

In addition to PKA other protein kinases such as proteinkinase B and the MAPKs have also been reported tophosphorylate CREB (Du and Montminy, 1998; Hansenet al., 1999). We observed that in SMG-C6 epithelial cellsDbcAMP-induced phosphorylation of CREB was significantlyattenuated either in the presence of PD98059 or in thepresence of SB203580 (Fig. 6A,B). Although CREBphosphorylation was reduced in cells treated individually witheitherMAPK inhibitor, the levels detectedwere still higher thanthat observed in untreated or control cells. However, whencells were exposed to both inhibitors together the levels ofphosphorylatedCREBweremarkedly reduced and comparableto untreated cells (Fig. 6C). These observations suggest that

Fig. 7. Effect of PKA inhibitors on DbcAMP-induced p38 and ERK1,2 MAPK phosphorylation. Cultured SMG-C6 cells were pretreatedwith either H-89 (10 mM) or KT5720 (0.5 mg/ml) for 1 h thenstimulated with 1 mM DbcAMP for 10 min. Western blot analysisusing antibodies against phosphorylated form of p38 MAPK(p-p38, upper, A) and total p38 MAPK (p38, lower, A) andphosphorylated formsofERK1,2(p-ERK1,2,upper,B)or totalERK1,2(ERK 1,2, lower, B) was performed as described earlier. Arepresentative Western blot image is shown. Correspondingdensitometricanalysis isdepicted for thephosphorylated forms.Meanvalues of 3–4 independent experiments are shownWSE. MP< 0.05versus DbcAMP-treated cells.

Fig. 8. A potential working model of the early cAMP signalingpathways in SMG-C6 epithelial cells. We speculate that inSMG-C6 epithelial cells exposure to exogenous DbcAMP results inthe phosphorylation of the CREB transcription factor via thestimulation of PKA. In addition, DbcAMP leads to thephosphorylation of p38 and ERK 1,2 MAPKs. PKA may phosphorylatep38 MAPK and ERK 1,2 MAPK via additional intermediary proteins inthis pathway. Activation of these particular MAPKs by either orboth pathways phosphorylates CREB. The block lines indicate thespecific inhibitors used in this study to attenuate that particularpathway. 1, H-89; 2, KT5720; 3, PD98059; 4, SB203580.

U P - R E G U L A T I O N O F a - E N a C B Y D b c A M P 109

both the p38 and ERK 1,2 MAPK pathways activated byDbcAMP also contribute to the phosphorylation of CREB, andbesides PKA appear to play a key role in the early signalingcascade of events which lead to the eventual up regulation ofa-ENaC in SMG-C6 epithelial cells.

In addition to phosphorylating CREB, activation orinactivation of other kinases such as the Ras-Raf-MAPKpathways has also been attributed to PKA (Vossler et al., 1997;Maizels et al., 1998; Liebmann, 2001; Richards, 2001). InSMG-C6 epithelial cells, inhibition of PKAwith KT5720 orH-89decreased DbcAMP-induced activation of p38 MAPK (Fig. 7A)and phosphorylation of P-ERK 1,2 (Fig. 7B). These findingssuggest that PKA-induced phosphorylation of CREB could alsobe mediated through activation of p38 and ERK 1,2 MAPK. Infact, inhibiting both p38 and ERK 1,2 MAPK simultaneously,decreased CREB phosphorylation more than the individualMAPK inhibitors (Fig. 6C). Furthermore, the time-course ofDbcAMP-induced ERK 1,2 and p38 phosphorylation at 5 and10 min, respectively, and CREB phosphorylation after 10 minsuggest that p38 and ERK 1,2 MAPKmay also contribute to thephosphorylation of CREB in addition to PKA. Figure 8 providesa brief, schematic outline of our experimental observationspresented in this study.

JOURNAL OF CELLULAR PHYSIOLOGY

In summary, our present findings indicate that in the SMG-C6epithelial cell line, long-term treatment (up to 48 h) withDbcAMP increased transepithelial Naþ transport by upregulating low endogenous levels of a-ENaC mRNA andprotein expression without altering the basal levels ofb- and g-ENaC. Also, we have shown that followingexposure to DbcAMP, PKA-dependent activation of p38 andERK 1,2 MAPK plays an important role in the onset of CREBphosphorylation within 10 min of stimulation and in turna-ENaC mRNA expression is increased after 24 h. Given thefact that in our cell system CREB was transientlyphosphorylatedwithin 1 hof stimulationwithDbcAMP (Fig. 4A)but a-ENaC mRNA was not detected until 24 h (Fig. 2A) anddominant-negativeCREBdid not completely attenuatea-ENaCmRNA expression (Fig. 4C), we speculate that CREB very likelyregulates the expression of other transcription factors/adaptorproteins and these intermediary proteins are involved in the upregulation of a-ENaC. It is important to recognize that p38 andERK 1,2 MAPK in addition to activating CREB, as we havedemonstrated here, can also influence transcription of genes byactivating other effector molecules. Clearly, cross-talkbetween multiple signaling pathways all contribute towardDbcAMP-dependent up regulation of a-ENaC in SMG-C6epithelial cells.

Literature Cited

Ahn S, Olive M, Aggarwal S, Krylov D, Ginty DD, Vinson C. 1998. A dominant-negativeinhibitor of CREB reveals that it is a general mediator of stimulus-dependent transcriptionof c-fos. Mol Cell Biol 18:967–977.

Barbry P, Hofman P. 1997. Molecular biology of Naþ absorption. Am J Physiol273:G571–G585.

Benos DJ, Awayda MS, Ismailov II, Johnson JP. 1995. Structure and function of amiloride-sensitive Naþ channels. J Membr Biol 143:1–18.

Berthiaume Y, Folkesson HG, Matthay MA. 2002. Lung edema clearance: 20 years ofprogress: Invited review: Alveolar edema fluid clearance in the injured lung. J Appl Physiol93:2207–2213.

Bland RD, Nielson DW. 1992. Developmental changes in lung epithelial ion transport andliquid movement. Annu Rev Physiol 54:373–394.

Busca R, Abbe P, Mantoux F, Aberdam E, Peyssonnaux C, Eychene A, Ortonne JP, Ballotti R.2000. Ras mediates the cAMP-dependent activation of extracellular signal-regulatedkinases (ERKs) in melanocytes. EMBO J 19:2900–2910.

CanessaCM,Horisberger JD, Rossier BC. 1993. Epithelial sodium channel related to proteinsinvolved in neurodegeneration. Nature 361:467–470.

110 M U S T A F A E T A L .

Canessa CM, Schild L, Buell G, Thorens B, Gautschi I, Horisberger JD, Rossier BC. 1994.Amiloride-sensitive epithelial Naþ channel is made of three homologous subunits. Nature367:463–467.

Castro R, Barlow-Walden L, Woodson T, Kerecman JD, Zhang GH, Martinez JR. 2000. Iontransport in an immortalized rat submandibular cell line SMG- C6. Proc Soc Exp Biol Med225:39–48.

Chen XJ, Eaton DC, Jain L. 2002. Beta-adrenergic regulation of amiloride-sensitive lungsodium channels. Am J Physiol 282:L609–L620.

Dagenais A, Kothary R, Berthiaume Y. 1997. The alpha subunit of the epithelial sodiumchannel in themouse:Developmental regulation of its expression. Pediatr Res 42:327–334.

Dagenais A, Denis C, Vives MF, Girouard S, Masse C, Nguyen T, YamagataT, Grygorczyk C, Kothary R, Berthiaume Y. 2001. Modulation of alpha-ENaC andalpha1-Naþ-Kþ-ATPase by cAMP and dexamethasone in alveolar epithelial cells. Am JPhysiol 281:L217–L230.

Dagenais A, Frechette R, Yamagata Y, Yamagata T, Carmel J-F, Clermont M-E, Brochiero E,Masse C, Berthiaume Y. 2003. Downregulation of ENaC activity and expression by TNF-ain alveolar epithelial cells. Am J Physiol 286:L301–L311.

Davis RJ. 1994. MAPKs: New JNK expands the group. Trends Biochem Sci 19:470–473.Du K, Montminy M. 1998. CREB is a regulatory target for the protein kinase Akt/PKB. J BiolChem 273:32377–32379.

Finley N, Norlin A, Baines DL, Folkesson HG. 1998. Alveolar epithelial fluid clearance ismediated by endogenous catecholamines at birth in guinea pigs. J Clin Invest 101:972–981.

Garty H, Palmer LG. 1997. Epithelial sodium channels: Function, structure, and regulation.Physiol Rev 77:359–396.

Giallard DJ, Hinnrasky S, Coscoy P, Hofman P, Matthay MA, Puchelle E, Barbry P. 2000. Earlyexpression of b- and g-subunits of epithelial sodium channel during human airwaydevelopment. Am J Physiol 278:L177–L484.

Gonzalez GA, Montminy MR. 1989. Cyclic AMP stimulates somatostatin gene transcriptionby phosphorylation of CREB at serine 133. Cell 59:675–680.

Habener JF, Miller CP, Vallejo M. 1995. cAMP-dependent regulation of gene transcription bycAMP response element-binding protein and cAMP response element modulator. VitamHorm 51:1–57.

Hansen TO, Rehfeld JF, Nielsen FC. 1999. Mitogen-activated protein kinase and proteinkinase a signaling pathways stimulate cholecystokinin transcription via activation of cyclicadenosine 30 ,50-monophosphate response element-binding protein. Mol Endocrinol13:466–475.

Hummler E, Barker P, Gatzy J, Beermann F, Verdumo C, Schmidt A, Boucher R, Rossier BC.1996. Early death due to defective neonatal lung liquid clearance inalpha-ENaC-deficient mice. Nat Genet 12:325–328.

Impey S, Obrietan K, Wong ST, Poser S, Yano S, Wayman G, Deloulme JC, Chan G,Storm DR. 1998. Cross talk between ERK and PKA is required for Ca2þ stimulation ofCREB-dependent transcription and ERK nuclear translocation. Neuron 21:869–883.

Jain L, Chen X, Malik B, Al-Khalili O, Eaton DC. 1999. Antisense oligonucleotides against thea-subunit of ENaCdecrease lung epithelial cation-channel activity. Am J Physiol 276:L1046–1051.

Kwok RP, Lundblad JR, Chrivia JC, Richards JP, Bachinger HP, Brennan RG, Roberts SG,GreenMR, GoodmanRH. 1994. Nuclear proteinCBP is a coactivator for the transcriptionfactor CREB. Nature 370:223–226.

Liebmann C. 2001. Regulation of MAP kinase activity by peptide receptor signaling pathway:Paradigms of multiplicity. Cell Signal 13:777–785.

Maizels ET, Cottom J, Jones JC, Hunzicker-DunnM. 1998. Follicle stimulating hormone (FSH)activates the p38 mitogen-activated protein kinase pathway, inducing small heat shockprotein phosphorylation and cell rounding in immature rat ovarian granulosa cells.Endocrinology 139:3353–3356.

JOURNAL OF CELLULAR PHYSIOLOGY

Matthay MA, Folkesson HG, Verkman AS. 1996. Salt and water transport across alveolarand distal airway epithelia in the adult lung. Am J Physiol 270:L487–L503.

MatthayMA, Clerici C, SaumonG. 2002. Invited review: Active fluid clearance from the distalair spaces of the lung. J App Physiol 93:1533–1541.

McGraw DW, Fukuda N, James PF, Forbes SL, Woo AL, Lingrel JB, Witte DP, Matthay MA,Liggett SB. 2001. Targeted transgenic expression of beta(2)-adrenergic receptors to type IIcells increases alveolar fluid clearance. Am J Physiol 281:L895–L903.

Morris RG, Schafer JA. 2002. cAMP increases density of ENaC subunits in the apicalmembrane of MDCK cells in direct proportion to amiloride-sensitive Na(þ) transport.J Gen Physiol 120:71–85.

O’Brodovich H. 1991. Epithelial ion transport in the fetal and perinatal lung. Am J Physiol261:C555–C564.

Otulakowski G, Rafii B, Bremner HR, O’Brodovich H. 1999. Structure and hormoneresponsiveness of the gene encoding the alpha-subunit of the rat amiloride-sensitiveepithelial sodium channel. Am J Respir Cell Mol Biol 20:1028–1040.

Planes C, Blot-Chabaud M, Matthay MA, Couette S, Uchida T, Clerici C. 2002. Hypoxia andb2-agonists regulate cell surface expression of the epithelial sodium channel in nativealveolar epithelial cells. J Biol Chem 277:47318–47324.

Quissell DO, Barzen KA, Gruenert DC, Redman RS, Camden JM, Turner JT. 1997.Development and characterization of SV40 immortalized rat submandibular acinar celllines. In Vitro Cell Dev Biol 33:164–173.

Richards JS. 2001. New signaling pathways for hormones and cyclic adenosine3’,5’-monophosphate action in endocrine cells. Mol Endocrinol 15:209–218.

Saldias FJ, Lecuona E, Comellas AP, Ridge KM, Rutschman DH, Sznajder JI. 2000.Beta-adrenergic stimulation restores rat lung ability to clear edema in ventilator-associatedlung injury. Am J Resp Crit Care Med 162:282–287.

Sartori C, Fang X, McGraw DW, Koch P, Snider ME, Folkesson HG, Matthay MA, Sartori C,Fang X, McGraw DW, Koch P, Snider ME, Folkesson HG, Matthay MA. 2002. Selectedcontribution: Long-term effects of beta(2)-adrenergic receptor stimulation on alveolarfluid clearance in mice. J Appl Physiol 93:1875–1880.

Schaeffer HJ, Weber MJ. 1999. Mitogen-activated protein kinases: Specific messages fromubiquitous messengers. Mol Cell Biol 19:2435–2444.

Schwartz JH. 2001. The many dimensions of cAMP signaling. Proc Natl Acad Sci USA98:13482–13484.

Snyder PM. 2000. Liddle’s syndrome mutations disrupt cAMP-mediated translocation of theepithelial Na(þ) channel to the cell surface. J Clin Invest 105:45–53.

Snyder PM. 2005. Minireview: Regulation of epithelial Naþ channel trafficking. Endocrinology146:5079–5085.

Snyder PM, Olson DR, Kabra R, Zhou R, Steines JC. 2004. cAMP and serum andglucocorticoid-inducible kinase (SGK) regulate the epithelial Na(þ) channel throughconvergent phosphorylation of Nedd4-2. J Biol Chem 279:45753–45758.

Thomas CP, Campbell JR, Wright PJ, Husted RF. 2004. cAMP-stimulated Naþ transport inH441 distal lung epithelial cells: Role of PKA, phosphatidylinositol 3-kinase, and sgk1. Am JPhysiol 287:L843–L851.

Tohda H, Foskett JK, O’Brodovich H, Marunaka Y. 1994. Cl- regulation of aCa(2þ)-activated nonselective cation channel in beta-agonist-treated fetal distal lungepithelium. Am J Physiol 266:C104–C109.

Vossler MR, Yao H, York RD, Pan MG, Rim CS, Stork PJ. 1997. cAMP activates MAP kinaseand Elk-1 through a B-Raf- and Rap1-dependent pathway. Cell 89:73–82.

Walters DV, Olver RE. 1978. The role of catecholamines in lung liquid absorption at birth.Pediatr Res 12:239–242.

Zentner MD, Lin HH, Wen X, Kim KJ, Ann DK. 1998. The amiloride-sensitive epithelialsodium channel alpha-subunit is transcriptionally down-regulated in rat parotid cells by theextracellular signal-regulated protein kinase pathway. J Biol Chem 273:30770–30776.