HUMANMUTATION 0,1^12,2008

RESEARCH ARTICLE

N-Terminal CFTR Missense Variants SeverelyAffect the Behavior of the CFTR Chloride Channel

G.G. Gene,1 A. Llobet,2 S. Larriba,1 D. de Semir,1 I. Martınez,1 A. Escalada,2 C. Solsona,2 T. Casals,1

and J.M. Aran1�

1Medical and Molecular Genetics Center, Institut d’Investigacio Biomedica de Bellvitge (IDIBELL), Hospital Duran i Reynals, Barcelona, Spain;2Laboratory of Molecular and Cellular Neuroscience, Department of Pathology and Experimental Therapeutics, Faculty of Medicine, Universitatde Barcelona (UB)–Institut d’Investigacio Biomedica de Bellvitge (IDIBELL), Barcelona, Spain

Communicated by Mireille Claustres

Over 1,500 cystic fibrosis transmembrane conductance regulator (CFTR) gene sequence variations have beenidentified in patients with cystic fibrosis (CF) and related disorders involving an impaired function of the CFTRchloride channel. However, detailed structure–function analyses have only been established for a few of them.This study aimed evaluating the impact of eight N-terminus CFTR natural missense changes on channelbehavior. By site-directed mutagenesis, we generated four CFTR variants in the N-terminal cytoplasmic tail(p.P5L, p.S50P, p.E60K, and p.R75Q) and four in the first transmembrane segment of membrane-spanningdomain 1 (p.G85E/V, p.Y89C, and p.E92K). Immunoblot analysis revealed that p.S50P, p.E60K, p.G85E/V, andp.E92K produced only core-glycosylated proteins. Immunofluorescence and whole cell patch-clamp confirmedintracellular retention, thus reflecting a defect of CFTR folding and/or trafficking. In contrast, both p.R75Q andp.Y89C had a glycosylation pattern and a subcellular distribution comparable to the wild-type CFTR, while thepercentage of mature p.P5L was considerably reduced, suggesting a major biogenesis flaw on this channel.Nevertheless, whole-cell chloride currents were recorded for all three variants. Single-channel patch-clampanalyses revealed that the channel activity of p.R75Q appeared similar to that of the wild-type CFTR, whileboth p.P5L and p.Y89C channels displayed abnormal gating. Overall, our results predict a major impact of theCFTR missense variants analyzed, except p.R75Q, on the CF phenotype and highlight the importance of theCFTR N-terminus on channel physiology. Hum Mutat 0,1–12, 2008. rr 2008 Wiley-Liss, Inc.

KEY WORDS: cystic fibrosis; CF; CFTR; maturation; channel gating

INTRODUCTION

Cystic fibrosis (CF; MIM] 219700) is a severe disordercharacterized by progressive lung disease, pancreatic dysfunction,male infertility, and high concentration of sweat electrolytes. Thisautosomal recessive disease is caused by mutations in the CFtransmembrane conductance regulator gene (CFTR; also symbo-lized ABCC7; MIM] 602421) [Welsh et al., 2001]. Since thecloning of the CFTR gene [Rommens et al., 1989; Riordan et al.,1989; Kerem et al., 1989], more than 1,500 sequence variationshave been reported within its sequence. Missense changesrepresent near 50% of the typical CF phenotypes (CF Interna-tional Consortium Mutation Data Base (CFMDB): www.genet.sickkids.on.ca/cftr). Furthermore, CFTR analysis carried out inatypical CF (CF with pancreatic sufficiency [CF-PS], congenitalbilateral absence of vas deferens [CBAVD]) and CFTR-relatedphenotypes (chronic pancreatitis, bronchiectasis, and other lungdiseases), strongly suggest the involvement of the CFTR gene intheir clinical manifestation. However, these phenotypes present adifferent mutational spectrum than that found in classic CFpatients, with a higher proportion, approaching 70%, of missensechanges [Casals et al., 2000, 2004a, 2004b; Zielenski, 2000; Welshet al., 2001]. Determining whether missense variants are or notdeleterious is extremely relevant in the diagnosis, treatment andprognosis of patients.

The CFTR protein (ABCC7) is a member of the structurallyrelated ATP-binding cassette (ABC) transporter family [Hydeet al., 1990], comprising two membrane-spanning domains(MSD1, MSD2), two nucleotide-binding domains (NBD1,

Published online inWiley InterScience (www.interscience.wiley.com).DOI10.1002/humu.20721

The Supplementary Material referred to in this article can be ac-cessed at http://www.interscience.wiley.com/jpages/1059-7794/suppmat.

Received 17 April 2007; accepted revised manuscript 9 December2007.

Grant sponsor: Fundacio¤ n Sira Carrasco para ayuda a la ¢brosisqu|¤ stica; Grant sponsor: Fondo de Investigaciones Sanitarias (FIS)/Fondo Europeo de Desarrollo Regional (FEDER); Grant number:PI020099; Grant sponsor: Spanish Ministry of Science and Educa-tion; Grant number: SAF 2005-736.

A. Llobet and S. Larriba contributed equally to this work.Present address for D. de Semir: Auerback Melanoma Research

Laboratory, Cutaneous Oncology Program, Comprehensive CancerCenter, Department of Dermatology, University of California, SanFrancisco, San Francisco, CA 94115.

�Correspondence to: Josep M. Aran, Centre de GeneØ tica MeØ dica iMolecular, Institut d’Investigacio¤ BiomeØ dica de Bellvitge (IDIBELL),Hospital Duran i Reynals, GranV|¤ a s/n km 2,7,08907 L’Hospitalet deLlobregat, Barcelona, Spain. E-mail: jaran@idibell.org

rr 2008 WILEY-LISS, INC.

NBD2), and a highly charged regulatory domain (RD) withconsensus sequences for phosphorylation by protein kinase A(PKA) and protein kinase C (PKC) [Riordan et al., 1989]. Thisplasma membrane (PM)-localized cyclic adenosine monopho-sphate (cAMP)-activated chloride channel regulates the fluid andelectrolyte balance in the epithelial cells [Riordan et al., 1989;Vankeerberghen et al., 2002]. Thus, defining the biochemical andphysiological features of the different mutated channels, e.g., theefficiency of their synthesis, folding, and/or trafficking, and thestatus of their ion transport activities, is attracting major efforts tocorrelate genotype and phenotype at the molecular level. Missensemutations, independently of their location within the CFTR gene,have been related to reduced chloride currents [Fanen et al.,1997], trafficking defects [Mendes et al., 2003], and abnormalsplicing processes [Aznarez et al., 2003; Pagani et al., 2003].

It has been reported that mild MSD mutants can alter theconduction and gating behavior of CFTR [Sheppard et al., 1993].Conversely, little experimental data are available regarding theimportance of the N-terminal cytoplasmic tail to CFTR function.It has been postulated that CFTR channel activity can beregulated through an intramolecular interaction between the N-terminus and the RD [Fu et al., 2001; Naren et al., 1999].Moreover, the CFTR N-terminus contains residues suggested tobind soluble N-ethylmaleimide-sensitive factor (NSF) attachmentreceptor (SNARE) proteins and regulate channel activity at thecell surface [Naren et al., 2000]. In addition, arginine-framedtripeptide sequences, such as that spanning residues 29–31 ofCFTR, might constitute a checkpoint for endoplasmic reticulumquality control [Chang et al., 1999; Roxo-Rosa et al., 2006;Hegedus et al., 2006].

We therefore sought to investigate the impact of naturallyoccurring 50 end CFTR subtle gene alterations on the behavior oftheir corresponding gene products. To address this aim, we haveperformed a comparative analysis of the posttranslational proces-sing, intracellular trafficking, and channel function of eight humanCFTR missense variants that have been overexpressed in culturedhuman epithelial cells. All but one of these variants have beenidentified in the Spanish population.

MATERIALSANDMETHODSSite-Directed Mutagenesis

The pCMVCFTRNot6.2 expression vectors carrying 6.2-kbhuman wild type or p.F508del CFTR cDNA (GenBankNM_000492.2) were kindly provided by Dr. Johanna Rommens(Hospital for Sick Children, Toronto, Ontario, Canada). The eightCFTR variants included in this study: p.P5L, p.S50P, p.E60 K,p.R75Q, p.G85E, p.G85V, p.Y89C, and p.E92K (Fig. 1A)were generated by oligonucleotide-directed mutagenesis inpCMVCFTRNot6.2wt using the QuickChangeTM XL-Site-Direc-ted Mutagenesis Kit (Stratagene, La Jolla, CA) according to themanufacturer’s instructions (see Supplementary Table S1 for adetailed description of the mutagenesis primers employed; availableonline at http://www.interscience.wiley.com/jpages/1059-7794/suppmat). Automatic DNA sequencing of the complete CFTR cDNAvariants verified the exclusive incorporation of each specificmutation. Transfection-quality plasmid DNA was prepared usingreagents supplied by Qiagen (Hilden, Germany).

Cell Culture andTransfections

Human Embryonic Kidney 293 (HEK 293) cells were grown at371C with 5% CO2 in Dulbecco’s modified Eagle’s medium(DMEM; Gibco, Life Technologies Ltd, Paisley, UK) supplemented

with 10% fetal bovine serum (FBS), 100 U/ml penicillin, 100 mg/mlstreptomycin, and 2 mM glutamine.

To optimize the transfection procedure, a preliminary studyusing the pEGFP-N1 reporter plasmid (BD Biosciences Clontech,Palo Alto, CA) and the nonviral vector Polyfect (Qiagen) wasperformed on HEK 293 cells. Flow cytometry analysis of thetransfected cells revealed efficiencies ranging from 30 to 40%, withan overall viability over 95%. Thus, the day before transfection6� 105 cells per well of a six-well plate were seeded andsubsequently these cells were transiently cotransfected with 2mg/each of the appropriate CFTR expression plasmid plus 200 ng ofthe pEGFP-N1 plasmid according to the Polyfect protocol providedby the manufacturer.

Cell Lysis and Immunoblotting

At 48 hr posttransfection, cotransfected HEK 293 cells werewashed with ice-cold PBS and solubilized in 150 ml of radio-immunoprecipitation assay (RIPA) buffer for 20 min on ice.Lysates were clarified by centrifugation at 14,000 rpm in a rotorFA-45-24-11 (Eppendorf, Hamburg, Germany) for 10 min at 41Cand used for immunoblotting.

Following protein quantification (BCA assay; Pierce Biotech-nology, Rockford, IL), 50 mg of protein extract per sample wereresuspended in SDS-PAGE solubilizing buffer, fractionated by7.3% SDS-PAGE, and transferred to Hybond-C extra membranes(Amersham Biosciences, Buckinghamshire, UK). CFTR proteinwas detected with monoclonal antibody MAB1660 (1:500; R&DSystems, Minneapolis, MN). The simultaneous detection of thecontrol Golgin 97 was performed with anti-Golgin 97 monoclonalantibody (1:2,000; Molecular Probes, Eugene, OR). Both primaryantibodies were recognized with a polyclonal horseradish perox-idase-conjugated goat anti-mouse immunoglobulin G (IgG)(1:2,000; DakoCytomation, Glostrup, Denmark). The blots weredeveloped by an enhanced chemiluminescence method using theECL Western Blotting detection Kit (Amersham Biosciences),and visualized and quantified with a Versadoc Imaging System(Bio-Rad, Hercules, CA).

Immuno£uorescenceAssay

HEK 293 cells were seeded on coverslips and transfected asdescribed above with wild-type and the different CFTR cDNAvariants.

To detect the CFTR protein at the cell membrane, at 48 hrposttransfection intact cells were rinsed twice in ice-cold PBSand submerged in preservation solution (1% [wt/vol] bovineserum albumin [BSA], 0.1% [wt/vol] sodium azide, in PBS) for1 hr at 41C. The PM was labeled with Alexa Fluor 488-conjugated Concanavalin A (at 5 mg/ml; Molecular Probes) for20 min at 41C. Cells were then washed three times withpreservation solution, fixed in 4% paraformaldehyde for 10 minat 41C, blocked with 20% FBS plus 0.2% Triton X-100 in PBS for1 hr, and incubated with monoclonal antibody MAB1660(1:1,000 dilution) in incubation solution (1% FBS plus 0.2%Triton X-100 in PBS) overnight at 41C. For primary antibodydetection, we added Alexa Fluor 568-conjugated goat anti-mouseIgG (1:500 dilution; Molecular Probes) in incubation solution for1 hr at room temperature. Nuclei were counterstained using TO-PRO 3 (1:3,000 dilution; Molecular Probes) in PBS for 1 hr atroom temperature. After successive washes with PBS thecoverslips were mounted in Vectashield (Vector Laboratories,Inc., Burlingame, CA).

2 HUMANMUTATION 0,1^12,2008

To assess the presence of CFTR protein at the endoplasmicreticulum (ER), cells were first fixed with paraformaldehyde andsubsequently blocked, stained with Alexa Fluor 488-conjugatedConcanavalin A, and incubated with monoclonal antibodyMAB1660, Alexa Fluor 568-conjugated goat anti-mouse IgG,and TO-PRO 3 as described above.

Controls for background fluorescence/autofluorescence wereperformed omitting the primary MAB1660 antibody, the second-ary Alexa Fluor 568-conjugated secondary antibody, and/or theAlexa Fluor 488-conjugated Concanavalin A marker.

Image acquisition at three fluorescence emission wavelengths(488, 543, and 633 nm), was performed in a Leica TCS SLconfocal scanning laser microscope (Leica Lasertechnik, Heidel-berg, Germany) adapted to an inverted Leitz DMIRBE microscope(Zeiss, Jena, Germany). Images were taken using a 63�Leitz Plan-Apochromatic objective and processed through the LCS LeicaConfocal Software.

Whole Cell Patch-Clamp Recording

HEK 293 cells were cotransfected with each of the CFTRexpression plasmids and the pEGFP-N1 reporter plasmid, asdescribed above, immediately seeded on glass circular coverslipsand recorded 8–36 hr after transfection.

Bath solution contained 140 mM N-methyl-D-glucamine chlor-ide (NMDG-Cl), 1 mM MgCl2, 10 mM glucose, and 10 mMHEPES (pH 7.4 with Tris). Pipette solution contained 140 mMNMDG-Cl, 5 mM EGTA, 5 mM Mg-ATP, 5 mM MgCl2, and10 mM HEPES (pH 7.2 with Tris). Chloride currents werestimulated with 5mM forskolin for 2 min. Patch electrodes of2–4 megaohm were produced from borosilicate glass (outerdiameter, 1.2 mm; inner diameter, 0.94 mm; Harvard Apparatus,Edenbridge, UK) with a P-97 puller (Sutter Instruments, Novato,CA). Seals of 4–5 gigaohm were obtained prior to establishingwhole-cell configuration. An EPC-9 patch clamp amplifier(HEKA, Heidelberg, Germany) controlled by Pulse software wasused for data recording and analysis. Currents were acquired at3 kHz and filtered at 0.5 kHz. Membrane potential was held at–40 mV. The CFTR inhibitor of cAMP-mediated chloride trans-port (CFTRinh-172; Sigma-Aldrich, St. Louis, MO) [Ma et al.,2002], was applied at 1mM to the extracellular solution.Experiments were conducted at room temperature. The averagecapacitances of HEK 293 cells transfected with wild-type CFTRand each of the three active channel variants were: wild-typeCFTR (1671 pF; n 5 24); p.R75Q (2473 pF; n 5 18); p.Y89C(2072 pF; n 5 6); and p.P5L (2173 pF; n 5 20). The values ofcurrent density were measured at the peak of the CFTR-mediatedcurrent.

Single-Channel Patch-Clamp Recording

Recordings were performed at room temperature using HEK 293cells 9–36 hr posttransfection. Excised inside-out patch-clampexperiments were carried out in identical bath (intracellular) andpipette (extracellular) solutions containing 140 mM NMDG,3 mM MgCl2, 1 mM Cs EGTA, and 10 mM TES (pH 7.3 withHCl), providing a chloride concentration of 146 mM. Electrodeswere pulled from borosilicate glass (Harvard Apparatus), coatedwith Sylgards (Dow Corning Corp., Midland, MI), and hadresistances of 8–12 megaohm when filled with the solution.

After excision of membrane patches, the catalytic subunit ofPKA (75 nM) and Mg-ATP (1 mM) (all from Sigma-Aldrich) wereadded to the bath solution within 5 min of patch excision toactivate CFTR channels. An EPC-9 patch clamp amplifier

(HEKA) was used for voltage clamping and data recording.Analysis was performed using EDR software v. 3.6.8 (Dr. JohnDempster, Strathclyde University, Glasgow, Scotland) and Igor Pro5 (Wavemetrics, Lake Oswego, OR). Currents were recorded at10 kHz, and digitally low-pass filtered at frequencies ranging from200 to 500 Hz.

To measure single-channel current amplitude all-point histo-grams were constructed. Chord conductance was calculated bydividing unitary current by the difference between the appliedvoltage and the reversal potential obtained through a rampprotocol in which the membrane potential was shifted from–90 mV to 180 mV during a 2-s period. For kinetic analyses, opentimes were obtained using an 50% crossing criterion for eventdetection. Transitions shorter than 1 ms in duration wereexcluded.

Because it was difficult to estimate the exact number ofchannels in a patch, open probability (Po) was expressed as theproduct of the number of channels in a patch (N Po). NPo wascalculated using the following equation:

NPo ¼ �ðt1 þ t2 þ . . .þ tnÞ=ttotal;

where N is the number of channels and tn is the time that one ormore channels are open during a recording lasting ttotal. Thosepatches showing more than five channels were excluded fromanalysis. Stable recordings, lasting from 1 to 6 min were used tocalculate the NPo.

Statistical analysis

Results are expressed as means7standard error of the mean(SEM) of n observations. To compare sets of data we used one-wayanalysis of variance (ANOVA). Differences were consideredstatistically significant when Po0.05. All tests were performedusing GraphPad Instat software v. 3.0 (GraphPad Software, SanDiego, CA).

RESULTSDescription and Cross-Species Analysis of NaturalN-Terminus CFTR Variants

We chose eight naturally occurring sequence variants, fourlocated across the N-terminal CFTR tail (p.P5L, p.S50P, p.E60K,and p.R75Q), and four within the first segment of MSD1 (p.G85E,p.G85V, p.Y89C, and p.E92 K) (Fig. 1A; Table 1).

Alignment of the first 100 N-terminal amino acids of CFTRfrom different species shows striking absolute sequence conserva-tion in each of the mutated sites of our study across all speciesexamined (Fig. 1B), which suggests that these amino acid moietiesare important for CFTR structure and function.

E¡ect of N-TerminusVariants on CFTR Expression andProcessing

We examined the steady-state production of the N-terminalCFTR variants after transient transfection into HEK 293 cells.Immunoblot analysis of the overexpressed CFTR proteins wasperformed using the CFTR-specific MAB1660 monoclonal anti-body, which recognizes the R domain of the channel. Blotnormalization by detection of endogenous Golgin allowed areliable comparison and relative quantification of the correspond-ing mutant CFTR maturation patterns (Fig. 2). Monitoring theglycosylation state of CFTR, from the ER core-glycosylated form(band B; 145 kDa) to the Golgi-processed, complex oligosacchar-ide-containing form (band C; 180 kDa) provides a useful approachto study CFTR maturation and trafficking [Cheng et al., 1990]. As

HUMANMUTATION 0,1^12,2008 3

expected, wild-type CFTR produced substantial amounts ofmature, fully glycosylated protein (band C), whereas the p.F508delmutant produced no mature protein, being the immature protein(band B) the sole translation product (Fig. 2A). Some of thevariants such as p.R75Q and p.Y89C, presented a glycosylationstatus comparable to the wild-type CFTR, suggesting that theymight reach the PM. In contrast, variants p.S50P, p.E60K, p.G85E,p.G85V, and p.E92K, produced only higher mobility band Bproteins suggesting that, like the p.F508del mutant, the resultingmisfolded channels are retained and degraded in the cytoplasm(Fig. 2A). Thus, these variants could be regarded as Class IImutants, eliminating the expression of CFTR protein at the cellsurface. Finally, variant p.P5L presented also an abnormalmaturation pattern, although some protein (near 13%, relativeto wild-type CFTR) was still able to proceed to the 180-kDa lower

mobility form compatible with membrane-associated channelactivity (Fig. 2B).

Further investigation into the behavior of the different CFTRvariants at the cellular level came after assessing their subcellulardistribution by immunolocalization, when overexpressed in HEK293 epithelial cells. To distinguish between PM and cytoplasm (C)localized CFTR, we performed colocalization assays using thefluorescently labeled lectin Concanavalin A, which binds to theinternal and nonreducing terminal alpha-mannosyl groups ofglycoproteins and other sugar-containing entities, and visualizedthe resulting patterns by confocal microscopy. As expected(Supplementary Fig. S1; left panel) wild-type CFTR (red) wasexpressed in the PM correlating with prior Concanavalin A(green) staining in unfixed, nonpermeabilized cells. Moreover,some nascent/immature CFTR protein present in the ER and

A

B 501

E92KY89C

G85E/V

P5LS50PE60KR75Q

NBD1 NBD2

R

FIGURE 1. A: Schematic representationof theN-terminalCFTRmissensevariants analyzed in this study.Variants (depicted ascircles)are distributed along theN-terminal cytoplasmic tail and the ¢rst transmembrane domain of CFTR. B: Alignment of the N-terminus(amino acids 1 to 100) of the CFTR protein derived from di¡erent species.The sequences derived from human (Homo sapiens, Gen-BankNM_000492), mouse (Musmusculus,GenBankNM_021050), Norway rat (Rattusnovergicus,GenBankNM_031506), Europeanrabbit (Oryctolagus cuniculus, GenBank NM_001082716), cow (Bos taurus, GenBank NM_174018), sheep (Ovis aries, GenBankNM_001009781), African clawed frog (Xenopus laevis, GenBank X65256), and spiny dog¢sh (Squalus acanthias, GenBankM83785) were aligned using the ClustalWmultiple sequence alignment program.The amino acid residue a¡ected by each of the var-iants analyzed (p.P5L, p.S50P, p.E60K, p.R75Q, p.G85E, p.G85V, p.Y89C, and p.E92K) is indicated in bold.

4 HUMANMUTATION 0,1^12,2008

Golgi compartments (C) was also visible after colocalization withConcanavalin A in paraformaldehyde fixed and permeabilizedcells. Likewise, variants p.R75Q and p.Y89C (Fig. 3) presented alocalization pattern similar to that of wild-type CFTR, with most ofthe protein visualized in the PM (yellow colocalization of themerge images). In contrast to wild-type CFTR, the p.F508delmutant was retained in the intracellular ER/Golgi compartments(Supplementary Fig. S1; right panel) and excluded from the PM.Similarly, variants p.S50P, p.E60K, p.G85V, p.G85E, and p.E92Kdisplayed a yellow colocalization pattern clearly compatible withretention of the anomalous CFTR proteins within the intracellularcompartments and no detection of PM staining (Fig. 3). Regardingthe colocalization pattern of the p.P5L variant, most cellspresented a predominant intracellular CFTR staining, with somedegree of PM localization (Fig. 3). Thus, these immunocytochem-ical findings fully correlate with our previous biochemical resultsobtained by immunoblot analysis.

Channel activity of the N-Terminus CFTR Variants

To delineate the physiological features of the N-terminal CFTRmutant channels overexpressed in HEK 293 cells, we measuredcAMP-stimulated currents in whole-cell voltage clamp configura-tion during forskolin application. In cells coexpressing wild-typeCFTR and EGFP, the application of forskolin produced largecurrents (ranging from 0.3 to 5.0 nA) that developed as early as6–8 hr after transfection (Fig. 4A). These currents were generatedby CFTR, because cells that did not express EGFP did not exhibitforskolin-induced currents. Moreover, they were sensitive to theapplication of the thiazolidinone CFTR inhibitor CFTRinh-172(Supplementary Fig. S2) [Ma et al., 2002]. Conversely, cellstransfected with p.F508del mutant, which is not able to traffic tothe PM, did not show any current during forskolin exposure(Fig. 4A). Likewise, none of the five variants (p.S50P, p.E60K,p.G85E, p.G85V, and p.E92K) in which severe misprocessing waspreviously demonstrated (Figs. 2 and 3), was able to generatecAMP-activated currents (Fig. 4B). From the variants expressed inthe PM (Fig. 3), only p.R75Q and p.Y89C generated cAMP-stimulated currents comparable to that of wild-type CFTR channel(Fig. 4C). To correct for cell size, peak currents were normalized to

TABLE 1. Genotype^PhenotypeCorrelation in theN-Terminal CFTRMissenseVariants Under Study�

Missensevarianta Phenotype Second allele (number of patients)b

p.P5L CF p.F508del (1), p.P205S (1)p.S50P CBAVD p.F508del (1), p.E115del (1)p.E60K CF p.G542X (1), p.I507del (1)p.R75Q HT p.F508del (3), p.E725K (1)

B p.R347H (1), p.R75Q (1), n.i. (4)Br c.1584G4A (2), c.1210-7_1210-6delTT (1),

n.i.(3)NT p.F508del (1)CP c.1584G4A (1), n.i. (3)MI n.i. (1)

CUAVD n.i. (2)OZ n.i. (2)

Normal p.R75Q (1), c.2052_2053insA (1), n.i. (1)p.G85E CF p.F508del (8), p.G542X (2), p.I507del (1),

c.580-1G4T (1), p.G85E (1), c.1477_1478delCA (1)

CBAVD p.G576A (1)HT p.L997F (1),WT (1)

p.G85V CF p.F508del (2), p.G542X (2), p.Y1092X (1),c.265715G4A (1), p.A1006E, c.1210-7_1210-

6delTT (1), n.i. (1)p.Y89C CF n.i. (1)c

p.E92K CF p.F508del (2), p.Q890X (1), p.L206W (1)CBAVD c.1210-7_1210-6delTT (1)

�The recommendations formutationnomenclature (www.hgvs.org/mutnomen/) wereused to nameCFTR gene sequence variations at both the nucleotide level and the pro-tein level. Following the approved HGVS nomenclature format (A of theATG transla-tionstart codonas 11,‘‘c.’’aspre¢x), thevariationsdescribed at the nucleotide level arealso given in parenthesis below,where theAof theATG translation start codon is num-bered as 1133 in accordance with the current CFTR gene numbering based on cDNAsequence (GenBank NM_000492.2) and on the CF mutation database (CFMDB):[c.1584G4A (1716G4A); c.1210-7_1210-6delTT (IVS8-6(5T)); c.2052_2053insA(2184insA); c.580-1G4T (712-1G4T); c.1477_ 1478delCA (1609delCA); c.26571

5G4A (278915G4A)].aThe variants identi¢ed inCFpatients have been reported recently by ourgroup [Alon-so et al.,2007]. Besides being present in CF patients, we have found both p.G85E andp.E92K variants in CBAVD patients [Casals et al., 2000]. Moreover, the p.G85E var-iant was also detected in one newborn with hypertrypsinemia [Gartner et al., 2003].The p.S50P variant was only found in CBAVD patients [Casals et al., 2000], whereasthe p.R75Q variant was described as a neutral change (CFMDB) and reported later inseveral CFTR-related disorders such as obstructive azoospermia [Dork et al., 1997],bronchiectasis [Casals et al., 2004b], chronic pancreatitis [Casals et al., 2004a], hy-pertrypsinemia [Gartner et al., 2003], as well as in the general population [Bombieriet al., 2000].bWT,wild-type; n.i., not identi¢ed.cItalian patient referred to in Padoan et al. [2000].CF, cystic ¢brosis;CBAVD, congenital bilateral absence of vas deferens; H, hypertryp-sinemia; B, bronchiectasis; Br, bronchitis; NT, pneumothorax, CP, chronic pancreati-tis, MI, meconium ileus; CUAVD, congenital unilateral absence of vas deferens; OZ,oligozoospermia.

FIGURE 2. Steady-state maturation patterns of wild-type and variant CFTR proteins. A: HEK 293 cells were transiently transfectedwith the wild type and the di¡erent CFTR variants and analyzed byWestern blotting usingmonoclonal antibody MAB1660, speci¢cagainst the R domain of CFTR, andmonoclonal antibody anti-Golgin 97, directed to the ubiquitous proteinGolgin 97.The positionsof core-glycosylated (bandB) and fully glycosylated (bandC) forms ofCFTR are indicated. Representative image from three indepen-dent experiments. NT, nontransfected cells. B: Immunoblot images fromHEK293 transfectants (like that shown in A), were quanti-¢edbydensitometry. For allCFTRproteins, the level ofmature bandCCFTRwasnormalized to that of bandBCFTR. *Po0.01vs.WTCFTR, n 53.

HUMAN MUTATION 0,1^12,2008 5

cell surface using individual PM capacitance values. On average,wild-type current densities showed a wide distribution, rangingfrom 10 pA/pF to 396 pA/pF, with a mean value of 114722 pA/pF,n 5 24. Similar values were observed for p.R75Q (98755 pA/pF,

n 5 18), and p.Y89C (114754 pA/pF, n 5 6), indicating thatthese CFTR variants were trafficked to the PM. In contrast,variant p.P5L showed a lower current density per surface unit(46716 pA/pF, n 5 20).

FIGURE 3. Immunocytochemical localizationof theCFTRvariants inHEK293 transfectedcells.Thepanels showPMand intracellularcompartments (ER1Golgi) (C) marked with Alexa Fluor 488-conjugated Concanavalin A (ConA) (green). CFTR labeling was per-formedwithMAB1660 plus Alexa Fluor 568-conjugated goat anti-mouse IgG (red).The nuclei were counterstainedwith the nucleardyeTO-PRO3 (blue).Confocal images from representative xy sections taken from1of 3 independent experiments show the subcellu-lar distribution of wild-type CFTR (WT), p.F508del mutant (F508del), and variants p.S50P (S50P), p.E60K (E60K), p.G85E(G85E), p.G85V (G85V), p.E92K (E92K), p.P5L (P5L), p.R75Q (R75Q), andp.Y89C (Y89C). Arrows indicatePM, andarrowheadsintracellular, CFTR andConA colocalization (yellow). In the £uorescent images fromvariants p.P5L, p.R75Q, and p.Y89C, the con-tour plot from a successfully transfected cell, obtained from the corresponding phase contrast image, is shown to aid the identi¢ca-tion of the cell boundary. Although all micrographs hold the same magni¢cation, several images have been zoomed to betterdistinguish subcellular colocalization features betweenConA and the di¡erent CFTRs.

6 HUMANMUTATION 0,1^12,2008

Additional characterization of the processing of p.P5L to thePM involved the analysis of its current density as a function of thetime post-transfection. Figure 4D shows that current densitieswere progressively increasing after transfection, corroborating thatthe channels were able to reach the cell surface. However, variantp.P5L always showed the smallest currents at the studied times.Due to the large variability in current densities, most likely relatedto the transfection procedure [Tseng et al., 1997], differencesbetween wild-type CFTR and the variants were not significant. Onaverage, at both 8–12 hr and 24–36 hr posttransfection, thecurrent density of p.P5L was reduced 1.5 to 2 times compared tothat of wild-type CFTR. This result is in line with a reducednumber of p.P5L channels at the PM (Figs. 2 and 3).

Single-Channel Analysis of N-Terminus CFTR VariantsExpressed at the Plasma Membrane

To further investigate whether variants p.R75Q, and p.Y89Caffected the chloride channel function of CFTR and, particularly,to delineate whether the reduced current density of p.P5L couldbe attributed only to a folding/trafficking defect or also to a

channel malfunction, single-channel experiments were carriedout. Wild-type CFTR channels were recorded in the excisedinside-out configuration by bath application of the catalyticsubunit of PKA. Under symmetrical chloride concentrations,currents elicited by PKA application showed a reversal potentialclose to 0 mV (Supplementary Fig. S3). The current-voltagerelationship was essentially ohmic showing inward rectificationabove 40 mV, typical of wild-type CFTR [Cai et al., 2003]. Single-channel activity was studied at –80 mV, a voltage value at whichthe transitions detected using all-point histograms (SupplementaryFig. S3) showed an estimated chord-conductance of 8.670.6 pS(n 5 11) (Table 2). Transitions detected remained open for periodsusually within the range of 0.1 to 0.7 s, showing a mean open time(to) of 259754 ms (Fig. 5; Table 2). The to values were used tocalculate the NPo of wild-type CFTR, which had a mean value of0.3070.05. Among the N-terminal CFTR variants successfullyexpressed at the PM, p.P5L displayed the most striking phenotype(Fig. 5) and showed a mean NPo of 0.0970.05 (n 5 7), about oneorder of magnitude smaller than wild-type CFTR (Table 2). Asdepicted in Figure 5A, wild-type CFTR single-channel recordscommonly showed more than one channel per patch, a

FIGURE 4. Whole-cell patch-clampanalysis of theN-terminalCFTRmissensevariants. Representativewhole-cell current recordingsofwild-typeCFTR and the eight missense variants transiently expressed inHEK293 cells. Chloride currents were stimulatedwith 5 mMforskolin for the times indicated below (solid blackbars).Themembrane potential was held at �40mV. A: Recordings fromwild-typeCFTR (WT) andp.F508delmutant (F508del). Forskolinwas applied for 3min.B: Recordings fromvariants p.S50P (S50P), p.E60K(E60K), p.G85E (G85E), p.G85V (G85V), and p.E92K (E92K) (superimposed recordings).Wild-type CFTR current is shown forcomparison. Forskolinwas applied for 3min.C:Whole-cell currents evoked by variants p. R75Q (R75Q), p.P5L (P5L), and p.Y89C(Y89C).Wild-type CFTR current (WT) is also shown for comparison. Forskolin was applied for 90s. D:Tra⁄cking of p.R75Q andp.P5L variants to the PM. Current densities of wild-type CFTR (WT), p.R75Q (R75Q) and p.P5L (P5L) variants depended on theposttransfection time. Densities at 8^12 hr after transfection (left column). Densities at 24^36 hr posttransfection (right column).Each determination is shown as a hollow square or circle. Error bars indicate mean7SEM.The large dispersion of the values wasassociated to the transient transfection procedure (see channel activity of the N-Terminus CFTR variants section for details). [Color¢gure can be viewed in the online issue,which is available at www.interscience.wiley.com.]

HUMANMUTATION 0,1^12,2008 7

phenomenon that was not observed in the p.P5L variantrecordings. Remarkably, the chord conductance of the p.P5Lvariant at –80 mV showed a 44% reduction respect to that of wild-type CFTR (Table 2). Thus, this decreased conductance, added toa reduced cell surface presence, could underlie the differences

observed in whole cell membrane currents between p.P5L andwild-type CFTR.

The analysis of transitions in variant p.P5L showed that openingtimes were significantly shorter compared to those from wild-typeCFTR (Fig. 5B; Table 2). On the basis of our observations, wearbitrarily defined open times for wild-type CFTR as short-lasting(o0.1s) and long-lasting (40.1s). After 20 min recording,approximately 43% of the channel openings were long-lasting. Incomparison, for variant p.P5L, only 18% of the observed openingslasted more than 0.1s, thus confirming an effect of the mutationon channel gating.

As explained above, variants p.R75Q and p.Y89C were also ableto reach the PM and supported chloride currents (Fig. 4). Single-channel records showed that variant p.R75Q displayed aphenotype similar to wild-type, in terms of NPo (0.3070.07), to

(2967116 ms), and chord conductance (8.670.8 pS) (n 5 10)(Fig. 5; Table 2). Finally, the p.Y89C CFTR was at some extent,

FIGURE 5. Representative single-channel recordings of wild-type CFTR, and variants p. R75Q and p.P5L. A: Single channel recordsshowing the behavior of wild-type CFTR (WT), p.R75Q (R75Q), and p.P5L (P5L) variants over a prolonged period. B: Histogramsshowing the open time (to) distributions for wild-typeCFTR (WT) and the p.P5L variant (P5L).The insets illustrate in high temporalresolution the behavior of both channels. Dotted lines indicate zero current level.

TABLE 2. Summary of Single-Channel Analysis FromN-Terminal CFTRVariantsy

Variant (number of patches) NPo to (ms) G (pS)

Wild-type (n 511) 0.3070.05 259754 8.670.6p.P5L (n 57) 0.0970.05� 77715� 4.870.3�

p.R75Q (n 510) 0.3070.07 2967116 8.670.8

ySummary of single-channel electrophysiological properties of the N-terminal CFTRvariants studied that showed successful tra⁄cking to the plasmamembrane.�Statistically signi¢cant values (Po0.01) after applying theANOVA test.NPo, product of the numberof channels (N) times openprobability (Po); to, meanopentime;G, single-channel chord conductance at �80mV.

8 HUMANMUTATION 0,1^12,2008

able to function in a similar way to wild-type CFTR but in thismutant channel, transitions to multiple subconductance stateswere frequent (data not shown).

DISCUSSION

The CFTR protein is a complex polytopic membrane channelwhose severe loss of function leads to an impaired transport ofchloride and other electrolytes through epithelial cell membranes,leading to the CF phenotype. Multiple structure-activity studieshave correlated the presence of CFTR gene mutations with theirclinical outcome and have defined the importance of keyfunctional domains at the molecular level, such as the NBDsand the RD [Sheppard and Welsh, 1999; Hanrahan and Wioland,2004]. Conversely, the relevance of the cytosolic N-terminal andC-terminal CFTR segments and their adjacent MSDs has beenrevisited recently [Guggino and Stanton, 2006; Linsdell, 2006,Thelin et al., 2007].

In this study we analyzed the effect of eight N-terminus CFTRmissense disease-related variants on the maturation and electro-physiological attributes of the chloride channel. Four of thevariants (p.P5L, p.S50P, p.E60 K, and p.R75Q) are localized withinthe cytosolic N-terminal tail, and the remaining four (p.G85E,p.G85V, p.Y89C, and p.E92K) are embedded in three positionswithin the first transmembrane segment (TM1) of MSD1. Theabsolute cross-species and evolutionary conservation of all theamino acid residues affected by the N-terminus missense CFTRchanges analyzed highlights their likely importance to CFTRstructure and function. Accordingly, using three differentapproaches (immunoblotting, immunocytochemistry, and electro-physiology) we found that 5 (p.S50P, p.E60K, p.G85E, p.G85V,and p.E92K) out of the 8 variants failed to mature, showing ananalogous behavior than the most common F508del mutation.Thus, they might be classified as class II mutations. Nevertheless,additional pulse-chase experiments assessing CFTR dynamicswould distinguish whether these mutants have increased turnoveror are inefficiently processed.

Conversely, we have observed that three CFTR missensevariants, two (p.P5L and p.R75Q) located in the cytosolicN-terminus and one (p.Y89C) located within the first segmentof MSD1, appeared to mature and reach the PM, albeit withdifferent efficiencies.

MSD1-A¡ectingVariants

Concerning the p.G85E mutation, our results are in agreementwith those reported by Decaestecker et al. [2004] through pulse-chase experiments. Moreover, although both p.G85E and p.G85V,similarly to p.E92K, do not appear to affect TM1 topology [Xionget al., 1997], they are also localized within or adjacent to thebilayer. This might cause subtle alterations in helical packing and/or conformation resulting in changes in CFTR tertiary structure,preventing proper maturation and leading to recognition of chainsby the ER quality control machinery (ERAD). Indeed, it has beenrecently shown that Derlin-1, an ERAD component, efficientlydegrades p.G85E and p.F508del folding mutants [Sun et al.,2006]. Alternatively, nascent mutant CFTR might alter theinteraction with the Hsp90 chaperone system perturbing thecytosolic proteasome activity and accelerating its degradation [Looet al., 1998; Wang et al., 2006].

Variant p.E92K results in an amino acid of opposite charge withpotential consequences in CFTR folding. Charged residues withintransmembrane segments, especially TM1 [Lu et al., 1998], areknown to influence protein topogenesis, folding, and oligomeriza-

tion [Cosson et al., 1991; Hartmann et al., 1989; Parks and Lamb,1991; Szczesna-Skorupa et al., 1988]. Correct membrane insertionand folding of TM1 and TM2 plays an essential role in themembrane insertion and folding of the subsequent TM segments ofCFTR [Chen and Zhang, 1999]. Moreover, CFTR E92 appears toface 401 away from the pore surface, suggesting that it mightcontribute to ionic interactions within the plane of the bilayer[Akabas et al., 1994].

Regarding the p.Y89C variant, based on the structural featuresof the amino acids involved (the bulky hydrophobic tyrosineresidue is substituted by the smaller hydrophilic cysteine residue)and its position within TM1 (between the severe foldingmutations p.G85E/V and p.E92K) it would also be expected amajor structural rearrangement and folding defect of MSD1conferred by this variant. Surprisingly, p.Y89C CFTR was able tomature quite efficiently and to reach the PM. However, its channelactivity was perturbed by the presence of frequent transitions tomultiple subconductance states and short-lasting channel openingscompared to wild-type CFTR, suggesting that this amino acidchange lead to defects in channel gating. This is in agreement withprevious studies reporting that TM1 and TM6 play dominant rolesin forming the CFTR channel pore and determining its functionalproperties [Ge et al., 2004]. Nevertheless, additional work isnecessary to accurately define the electrophysiological properties ofthe p.Y89C variant.

Cytosolic N-TerminalVariants

It has been reported that deletions in the amino-terminalcytosolic tail region of CFTR resulted in defective trafficking ofCFTR out of the ER to the cell surface, suggesting that an intactamino terminus is critical for biosynthesis [Prince et al., 1999].The cytosolic CFTR N-tail is known to physically link the channelto protein components of the membrane traffic machinery(SNAREs) such as SNAP23, syntaxin 1A [Naren et al., 1998;Peters et al., 2001], and syntaxin 8, inhibiting exocytosis, therebyreducing the levels of CFTR to the PM [Bilan et al., 2004].Moreover, a highly conserved acidic stretch of a putative helix inthe N-tail (residues 46–60) also interacts with the R domaincontrolling PKA-dependent channel gating [Chappe et al., 2005;Naren et al., 1999]. Thus, all these important interactions mightbe perturbed by the N-tail CFTR folding mutations p.S50P andp.E60K. In the former, proline residues, such as the introduced inthe p.S50P mutation, are not favored, introducing ‘‘kinks’’ ina-helices because the backbone nitrogen is not available forhydrogen bonding and because of steric constraints caused by theirring structure [Richardson and Richardson, 1989]. Conversely,p.E60K results in an amino acid of opposite charge. Last,mutations within the CFTR N-tail could promote abnormalchannel folding by interfering with the proposed CFTRN-terminus–molecular chaperone cysteine string protein interac-tion [Zhang et al., 2002].

For the remaining N-terminus CFTR cytosolic variants showingmaturation and PM localization, p.R75Q also affected the chargeof the position as a result of amino acid exchange. Nevertheless,p.R75Q CFTR both displayed a maturation pattern andsubcellular localization analogous to the wild-type CFTR, anddid not cause major alterations in intrinsic chloride channelactivity. Thus, the fact that p.R75Q is located at the hingebetween the N-terminal CFTR cytosolic tail and the first segmentof MSD1 may involve some degree of structural interdomainflexibility. Our data seem to corroborate previous results fromgenotype–clinical phenotype correlations [Divac et al., 2004;

HUMANMUTATION 0,1^12,2008 9

Ravnik-Glavac et al., 2000; Ziedalski et al., 2006], leading to thep.R75Q variant’s consideration as a polymorphism [Nikolic et al.,2006].

Finally, on the basis of the presented experimental data, variantp.P5L exhibited a very interesting, complex, and mixed phenotypewith features of folding mutant and defective channel. Indeed,similarly to the above-referred cytosolic N-terminal CFTR variantsp.S50P and p.E60K, folding and/or trafficking/processing defectsseem to be the major factors contributing to the abnormal p.P5LCFTR phenotype, which showed a greatly reduced expression atthe PM by immunodetection. Similar behavior has been describedfor other neighboring CFTR mutations. A recent report analyzingthe natural mutations p.R31C and p.R31L, has shown that thesesubtle N-terminus alterations compromise biogenesis and enhanceinternalization of CFTR, contributing to the loss of surfaceexpression and the associated defect in chloride conductance ofthe channel [Jurkuvenaite et al., 2006]. Our results do notcorroborate a previous report classifying p.P5L variant as a class IImutation [Thelin et al., 2007], as we found a small but significantamount of mature p.P5L CFTR mutant at the PM. The reasons forthis discrepancy are not clear, although it might be due tomethodological divergences in the processing of the cell extractsand/or in the antibody-mediated recognition of the CFTR forms.

A further electrophysiological analysis of the mature p.P5LCFTR fraction revealed also altered channel activity. Asmentioned above, the CFTR N-terminus is important not onlyfor channel trafficking, but it also regulates its activity in the PM.A recent functional model for CFTR channel gating proposes thatits N-tail regulates channel gating by facilitating entering into orremaining within a long open state. Hence, N-tail mutants may bedefective in gating transitions without affecting ATP binding [Fuet al., 2001]. In p.P5L variant, disruption of the steric constraintsimposed by the ring structure of the proline residue next to theputative a-helix segment responsible for interacting with the Rdomain could affect this interdomain communication leading tothe observed altered gating parameters of the mutant channel.

Genotype^Phenotype Correlations

All selected N-terminus CFTR missense variants exceptp.Y89C, described in an Italian patient [Padoan et al., 2000],have been identified in Spanish individuals with CF or differentCFTR-related disorders (see Table 1 and references therein).Thus, taking into account the second allele, it seems to be asignificant correlation between the chloride channel function frommost of the CFTR variants under study and the severity of thecorresponding clinical phenotypes. The main exception is variantp.S50P, which, despite exhibiting absence of mature CFTR protein,was merely found in two CBAVD patients [Casals et al., 2000].Lack of genotype–phenotype correlation has been reported fordifferent CFTR mutations and could involve additional regulatoryelements/factors contributing to the overall outcome of thepatient’s phenotype [Romey et al., 1999].

In summary, using a combination of immunochemical andelectrophysiological methodologies, we have established themolecular functional characterization of natural missense CFTRvariants located in the N-terminus of the channel. Overall, wehave verified the severe outcome of the majority of theseN-terminal variants, affecting biogenesis and/or channel gating,which highlights the functional importance of the N-tail and TM1segment in CFTR physiology. Moreover, our gathered data mightprove helpful to contrast correlations inferred between thegenotypes and the clinical phenotypes observed in patients bearing

the studied N terminus CFTR missense variants. Thus, thesestudies might be clinically valuable both for the prognosis ofdifferent CFTR-related diseases resulting from the assessed CFTRvariants and for the rational design of personalized therapiesaccording to the underlying molecular and functional CFTRdefect.

ACKNOWLEDGMENTS

D.S. and I.M. were recipients of a fellowship from ‘‘AssociacioCatalana de Fibrosi Quıstica’’. A.L. and S.L. are Sistema Nacionalde Salud (SNS) researchers supported by FIS 04-00173 and FIS03-00088, respectively. J.M.A. is sponsored by the ResearchersStabilization Program from the SNS. The authors are members ofthe Red de Centros del Instituto de Salud Carlos III (Ref. C03/07).

REFERENCES

Akabas MH, Kaufmann C, Cook TA, Archdeacon P. 1994. Amino acid

residues lining the chloride channel of the cystic fibrosis transmembrane

conductance regulator. J Biol Chem 269:14865–14868.

Alonso MJ, Heine-Suner D, Calvo M, Rosell J, Gimenez J, Ramos MD,

Telleria JJ, Palacio A, Estivill X, Casals T. 2007. Spectrum of mutations in

the CFTR gene in cystic fibrosis patients of Spanish ancestry. Ann Hum

Genet 71:194–201.

Aznarez I, Chan EM, Zielenski J, Blencowe BJ, Tsui LC. 2003.

Characterization of disease-associated mutations affecting an exonic

splicing enhancer and two cryptic splice sites in exon 13 of the cystic

fibrosis transmembrane conductance regulator gene. Hum Mol Genet

12:2031–2040.

Bilan F, Thoreau V, Nacfer M, Derand R, Norez C, Cantereau A, Garcia M,

Becq F, Kitzis A. 2004. Syntaxin 8 impairs trafficking of cystic fibrosis

transmembrane conductance regulator (CFTR) and inhibits its channel

activity. J Cell Sci 117:1923–1935.

Bombieri C, Giorgi S, Carles S, de Cid R, Belpinati F, Tandoi C, Pallares-

Ruiz N, Lazaro C, Ciminelli BM, Romey MC, Casals T, Pompei F,

Gandini G, Claustres M, Estivill X, Pignatti PF, Modiano G. 2000. A new

approach for identifying non-pathogenic mutations. An analysis of the

cystic fibrosis transmembrane regulator gene in normal individuals. Hum

Genet 106:172–178.

Cai Z, Scott-Ward TS, Sheppard DN. 2003. Voltage-dependent gating of

the cystic fibrosis transmembrane conductance regulator Cl- channel.

J Gen Physiol 122:605–620.

Casals T, Bassas L, Egozcue S, Ramos MD, Gimenez J, Segura A, Garcia F,

Carrera M, Larriba S, Sarquella J, Estivill X. 2000. Heterogeneity for

mutations in the CFTR gene and clinical correlations in patients with

congenital absence of the vas deferens. Hum Reprod 15:1476–1483.

Casals T, Aparisi L, Martinez-Costa C, Gimenez J, Ramos MD, Mora J, Diaz

J, Boadas J, Estivill X, Farre A. 2004a. Different CFTR mutational

spectrum in alcoholic and idiopathic chronic pancreatitis? Pancreas

28:374–379.

Casals T, De-Gracia J, Gallego M, Dorca J, Rodriguez-Sanchon B, Ramos

MD, Gimenez J, Cistero-Bahima A, Olveira C, Estivill X. 2004b.

Bronchiectasis in adult patients: an expression of heterozygosity for

CFTR gene mutations? Clin Genet 65:490–495.

Chang XB, Cui L, Hou YX, Jensen TJ, Aleksandrov AA, Mengos A,

Riordan JR. 1999. Removal of multiple arginine-framed trafficking

signals overcomes misprocessing of delta F508 CFTR present in most

patients with cystic fibrosis. Mol Cell 4:137–142.

Chappe V, Irvine T, Liao J, Evagelidis A, Hanrahan JW. 2005.

Phosphorylation of CFTR by PKA promotes binding of the regulatory

domain. EMBO J 24:2730–2740.

Chen M, Zhang JT. 1999. Topogenesis of cystic fibrosis transmembrane

conductance regulator (CFTR): regulation by the amino terminal

transmembrane sequences. Biochemistry 38:5471–5477.

Cheng SH, Gregory RJ, Marshall J, Paul S, Souza DW, White GA,

O’Riordan CR, Smith AE. 1990. Defective intracellular transport and

10 HUMANMUTATION 0,1^12,2008

processing of CFTR is the molecular basis of most cystic fibrosis. Cell

16:827–834.

Cosson P, Lankford SP, Bonifacino JS, Klausner RD. 1991. Membrane

protein association by potential intramembrane charge pairs. Nature

351:414–416.

Decaestecker K, Decaestecker E, Castellani C, Jaspers M, Cuppens H, De

Boeck K. 2004. Genotype/phenotype correlation of the G85E mutation

in a large cohort of cystic fibrosis patients. Eur Respir J 23:679–684.

Divac A, Nikolic A, Mitic-Milikic M, Nagorni-Obradovic L, Petrovic-

Stanojevic N, Dopudja-Pantic V, Nadaskic R, Savic A, Radojkovic D.

2004. High frequency of the R75Q CFTR variation in patients with

chronic obstructive pulmonary disease. J Cyst Fibros 3:189–191.

Dork T, Dworniczak B, Aulehla-Scholz C, Wieczorek D, Bohm I, Mayerova

A, Seydewitz HH, Nieschlag E, Meschede D, Horst J, Pander HJ,

Sperling H, Ratjen F, Passarge E, Schmidtke J, Stuhrmann M. 1997.

Distinct spectrum of CFTR gene mutations in congenital absence of vas

deferens. Hum Genet 100:365–377.

Fanen P, Labarthe R, Garnier F, Benharouga M, Goossens M, Edelman A.

1997. Cystic fibrosis phenotype associated with pancreatic insufficiency

does not always reflect the cAMP-dependent chloride conductive

pathway defect. Analysis of C225R-CFTR and R1066C-CFTR. J Biol

Chem 272:30563–30566.

Fu J, Ji HL, Naren AP, Kirk KL. 2001. A cluster of negative charges at the

amino terminal tail of CFTR regulates ATP-dependent channel gating.

J Physiol 536:459–470.

Gartner S, Cobos N, Maya A, Casals T, Seculi JL, Asensio O, Bosque M,

Estivill X, Puliol M, Prats R and study group for neonatal screening of

cystic fibrosis in Cataluna. 2003. Neonatal screening for cystic fibrosis in

Cataluna, Spain. Pediatr Pulmonol S25:221.

Ge N, Muise CN, Gong X, Linsdell P. 2004. Direct comparison of the

functional roles played by different transmembrane regions in the cystic

fibrosis transmembrane conductance regulator chloride channel pore.

J Biol Chem 279:55283–55289.

Guggino WB, Stanton BA. 2006. New insights into cystic fibrosis:

molecular switches that regulate CFTR. Nat Rev Mol Cell Biol

7:426–436.

Hanrahan JW, Wioland MA. 2004. Revisiting cystic fibrosis transmem-

brane conductance regulator structure and function. Proc Am Thorac

Soc 1:17–21.

Hartmann E, Rapoport TA, Lodish HF. 1989. Predicting the orientation of

eukaryotic membrane-spanning proteins. Proc Natl Acad Sci USA

86:5786–5790.

Hegedus T, Aleksandrov A, Cui L, Gentzsch M, Chang XB, Riordan JR.

2006. F508del CFTR with two altered RXR motifs escapes from ER

quality control but its channel activity is thermally sensitive. Biochim

Biophys Acta 1758:565–572.

Hyde SC, Emsley P, Hartshorn MJ, Mimmack MM, Gileadi U, Pearce SR,

Gallagher MP, Gill DR, Hubbard RE, Higgins CF. 1990. Structural model

of ATP-binding proteins associated with cystic fibrosis, multidrug

resistance and bacterial transport. Nature 346:362–365.

Jurkuvenaite A, Varga K, Nowotarski K, Kirk KL, Sorscher EJ, Li Y, Clancy

JP, Bebok Z, Collawn JF. 2006. Mutations in the amino terminus of the

cystic fibrosis transmembrane conductance regulator enhance endocy-

tosis. J Biol Chem 281:3329–3334.

Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK,

Chakravarti A, Buchwald M, Tsui LC. 1989. Identification of the cystic

fibrosis gene: genetic analysis. Science 245:1073–1080.

Linsdell P. 2006. Mechanism of chloride permeation in the cystic fibrosis

transmembrane conductance regulator chloride channel. Exp Physiol

91:123–129.

Loo MA, Jensen TJ, Cui L, Hou Y, Chang XB, Riordan JR. 1998.

Perturbation of Hsp90 interaction with nascent CFTR prevents its

maturation and accelerates its degradation by the proteasome. EMBO J

17:6879–6887.

Lu Y, Xiong X, Helm A, Kimani K, Bragin A, Skach WR. 1998. Co- and

posttranslational translocation mechanisms direct cystic fibrosis trans-

membrane conductance regulator N terminus transmembrane assembly.

J Biol Chem 273:568–576.

Ma T, Thiagarajah JR, Yang H, Sonawane ND, Folli C, Galietta LJ,

Verkman AS. 2002. Thiazolidinone CFTR inhibitor identified by high-

throughput screening blocks cholera toxin-induced intestinal fluid

secretion. J Clin Invest 110:1651–1658.

Mendes F, Roxo Rosa M, Dragomir A, Farinha CM, Roomans GM, Amaral

MD, Penque D. 2003. Unusually common cystic fibrosis mutation in

Portugal encodes a misprocessed protein. Biochem Biophys Res Commun

311:665–671.

Naren AP, Quick MW, Collawn JF, Nelson DJ, Kirk KL. 1998. Syntaxin 1A

inhibits CFTR chloride channels by means of domain-specific protein-

protein interactions. Proc Natl Acad Sci USA 95:10972–10977.

Naren AP, Cormet-Boyaka E, Fu J, Villain M, Blalock JE, Quick MW, Kirk

KL. 1999. CFTR chloride channel regulation by an interdomain

interaction. Science 286:544–548.

Naren AP, Di A, Cormet-Boyaka E, Boyaka PN, McGhee JR,

Zhou W, Akagawa K, Fujiwara T, Thome U, Engelhardt JF,

Nelson DJ, Kirk KL. 2000. Syntaxin 1A is expressed in airway epithelial

cells, where it modulates CFTR Cl(-) currents. J Clin Invest 105:

377–386.

Nikolic A, Divac A, Stankovic M, Dinic J, Tomic B, Ljujic M. 2006.

Analysis of common CFTR polymorphisms 5 T, M470 V and R75Q in

healthy Serbian population. Genetika 42:996–998.

Padoan R, Costantini D, Russo MC, Ambrosioni A, Fiori S, Prandoni S,

Cantu-Rajnoldi A, Seia M, Giunta A. 2000. A novel missense mutation

(Y89C) in exon 3 of the CFTR (ABCC7) gene in a young male. Hum

Mutat 15:486.

Pagani F, Stuani C, Tzetis M, Kanavakis E, Efthymiadou A, Doudounakis

S, Casals T, Baralle FE. 2003. New type of disease causing mutations: the

example of the composite exonic regulatory elements of splicing in

CFTR exon 12. Hum Mol Genet 12:1111–1120.

Parks GD, Lamb RA. 1991. Topology of eukaryotic type II membrane

proteins: importance of N-terminal positively charged residues flanking

the hydrophobic domain. Cell 64:777–787.

Peters KW, Qi J, Johnson JP, Watkins SC, Frizzell RA. 2001. Role of snare

proteins in CFTR and ENaC trafficking. Pflugers Arch 443:S65–S69.

Prince LS, Peter K, Hatton SR, Zaliauskiene L, Cotlin LF, Clancy JP,

Marchase RB, Collawn JF. 1999. Efficient endocytosis of the cystic

fibrosis transmembrane conductance regulator requires a tyrosine-based

signal. J Biol Chem 274:3602–3609.

Ravnik-Glavac M, Dean M, Glavac D. 2000. Study of mutant and

polyvariant mutant CFTR genes in patients with congenital absence of

the vas deferens. Pflugers Arch 439:R53–R55.

Richardson JS, Richardson DC. 1989. Principles and patterns of protein

conformation. In: Fasman GD, editor. Prediction of protein structure and

the principles of protein conformation. New York: Plenum Press.

p. 1–98.

Riordan JR, Rommens JM, Kerem B, Alon N, Rozmahel R, Grzelczak Z,

Zielenski J, Lok S, Plavsic N, Chou JL, Drumm ML, Iannuzzi MC,

Collins FS, Tsui LC. 1989. Identification of the cystic fibrosis gene:

cloning and characterization of complementary DNA. Science 245:

1066–1073.

Romey MC, Guittard C, Chazalette JP, Frossard P, Dawson KP, Patton MA,

Casals T, Bazarbachi T, Girodon E, Rault G, Bozon D, Seguret F,

Demaille J, Claustres M. 1999. Complex allele [-102T4A1

S549R(T4G)] is associated with milder forms of cystic fibrosis than

allele S549R(T4G) alone. Hum Genet 105:145–150.

Rommens JM, Iannuzzi MC, Kerem B, Drumm ML, Melmer G,

Dean M, Rozmahel R, Cole JL, Kennedy D, Hidaka N, Zsiga M,

Buchwald M, Riordan JR, Tsui LC, Collins FS. 1989. Identification of

the cystic fibrosis gene: chromosome walking and jumping. Science

245:1059–1065.

Roxo-Rosa M, Xu Z, Schmidt A, Neto M, Cai Z, Soares CM,

Sheppard DN, Amaral MD. 2006. Revertant mutants G550E and 4RK

rescue cystic fibrosis mutants in the first nucleotide binding domain of

CFTR by different mechanisms. Proc Natl Acad Sci USA 103:

17891–17896.

Sheppard DN, Rich DP, Ostedgaard LS, Gregory RJ, Smith AE, Welsh MJ.

1993. Mutations in CFTR associated with mild-disease-form Cl-

channels with altered pore properties. Nature 362:160–164.

HUMANMUTATION 0,1^12,2008 11

Sheppard DN, Welsh MJ. 1999. Structure and function of the CFTR

chloride channel. Physiol Rev 79:S23–S45.

Sun F, Zhang R, Gong X, Geng X, Drain PF, Frizzell RA. 2006. Derlin-1

promotes the efficient degradation of CFTR and CFTR folding mutants.

J Biol Chem 281:36856–36863.

Szczesna-Skorupa E, Browne N, Mead D, Kemper B. 1988. Positive charges

at the NH2 terminus convert the membrane-anchor signal peptide of

cytochrome P-450 to a secretory signal peptide. Proc Natl Acad Sci USA

85:738–742.

Thelin WR, Chen Y, Gentzsch M, Kreda SM, Sallee JL, Scarlett CO,

Borchers CH, Jacobson K, Stutts MJ, Milgram SL. 2007. Direct

interaction with filamins modulates the stability and plasma membrane

expression of CFTR. J Clin Invest 117:364–374.

Tseng WC, Haselton FR, Giorgio TD. 1997. Transfection by cationic

liposomes using simultaneous single cell measurements of plasmid

delivery and transgene expression. J Biol Chem 272:25641–25647.

Vankeerberghen A, Cuppens H, Cassiman JJ. 2002. The cystic fibrosis

transmembrane conductance regulator: an intriguing protein with

pleiotropic functions. J Cyst Fibros 1:13–29.

Wang X, Venable J, LaPointe P, Hutt DM, Koulov AV, Coppinger J, Gurkan

C, Kellner W, Matteson J, Plutner H, Riordan JR, Kelly JW, Yates JR 3rd,

Balch WE. 2006. Hsp90 cochaperone Aha1 downregulation rescues

misfolding of CFTR in cystic fibrosis. Cell 127:803–815.

Welsh MJ, Ramsey BW, Accurso F, Cutting GR. 2001. Cystic fibrosis. In:

Scriver CR, Beaudet AL, Sly WS, Valle D, editors. The metabolic and

molecular basis of inherited disease. New York: McGraw-Hill, Inc. p.

5121–5188.

Xiong X, Bragin A, Widdicombe JH, Cohn J, Skach WR. 1997. Structural

cues involved in endoplasmic reticulum degradation of G85E and G91R

mutant cystic fibrosis transmembrane conductance regulator. J Clin

Invest 100:1079–1088.

Zhang H, Peters KW, Sun F, Marino CR, Lang J, Burgoyne RD, Frizzell RA.

2002. Cysteine string protein interacts with and modulates the

maturation of the cystic fibrosis transmembrane conductance regulator.

J Biol Chem 277:28948–28958.

Ziedalski TM, Kao PN, Henig NR, Jacobs SS, Ruoss SJ. 2006. Prospective

analysis of cystic fibrosis transmembrane regulator mutations in adults

with bronchiectasis or pulmonary nontuberculous mycobacterial infec-

tion. Chest 130:995–1002.

Zielenski J. 2000. Genotype and phenotype in cystic fibrosis. Respiration

67:117–133.

12 HUMAN MUTATION 0,1^12,2008