Production and characterisation of monoclonal and polyclonal ...

4 downloads 241 Views 640KB Size Report
nologies Ltd, Paisley, Scotland) supplemented with 10% foetal calf ..... (a) to show reactivity on subglandular epithelial cells; GL, glandular lumen; S, stroma.
2433

Journal of Cell Science 108, 2433-2444 (1995) Printed in Great Britain © The Company of Biologists Limited 1995

Production and characterisation of monoclonal and polyclonal antibodies to different regions of the cystic fibrosis transmembrane conductance regulator (CFTR): detection of immunologically related proteins Jenny Walker1, Judy Watson1, Christopher Holmes2, Aleksander Edelman3 and George Banting1,* 1Department of Biochemistry, and Biotechnology and Biological Sciences Research Council Molecular Recognition Centre, University of Bristol, Bristol BS8 1TD, UK 2Department of Obstetrics and Gynaecology, University of Bristol, Bristol BS8 1TD, UK 3Inserm U323, Faculte de Medecine Necker, Enfants Malades, 156, rue de Vaugirard, 75730 Paris, Cedex 15, France

*Author for correspondence

SUMMARY We have raised mouse monoclonal antibodies to eight synthetic peptides corresponding to different regions of the human cystic fibrosis transmembrane conductance regulator (CFTR) and rabbit polyclonal antisera to βgalactosidase fusion proteins which encompass three different regions of CFTR. Immunoblot, immunoprecipitation, immunofluorescence and immunocytochemical

INTRODUCTION Cystic fibrosis (CF) is a common lethal genetic disease in Western Europe. The principal manifestations of CF include increased concentrations of chloride ions in exocrine gland secretions, pancreatic insufficiency, chronic lung disease, intestinal blockage, malabsorption of fat, male infertility (due to bilateral lack of a vas deferens) and reduced female fertility (Boat et al., 1989). CF is caused by mutations in the gene encoding the cystic fibrosis transmembrane conductance regulator (CFTR) (Rommens et al., 1989). The principal mutation identified in the gene, accounting for approximately 70% of the individuals with the disease, is a three base pair deletion resulting in the loss of a phenylalanine residue at position 508 of the protein sequence (Kerem et al., 1989). The remaining 30% of mutations occur at diverse sites within CFTR (Tsiu, 1992). Conceptual translation of the cDNA sequence of CFTR predicts a ~170 kDa protein with homology to a family of membrane associated ATP-dependent transport proteins, the ATP-binding cassette (ABC) transporter family (see Higgins, 1992, for recent review). At present there is little direct biochemical evidence concerning the structure of CFTR, but models have been predicted based on conceptual translation of the cDNA sequence (Riordan et al., 1989). The key features of the proposed structure of CFTR include twelve hydrophobic α-helical transmembrane domains, two nucleotide triphosphate binding domains, a unique highly charged cytoplasmic domain and two external potential glycosylation sites. In light

experiments demonstrate that, in addition to recognising CFTR, these antibodies recognise one or more immunologically related proteins with a similar molecular mass, calcium responsiveness and tissue distribution to CFTR. Key words: CFTR, cystic fibrosis, ionomycin, MDR, Cl− channel, immunolocalisation

of the primary symptoms of CF, the predicted structure of CFTR led to speculation that it might encode a Cl− ion channel (Riordan et al., 1989). Data from reconstitution and transfection experiments indicate that, as predicted, CFTR acts as a cAMP activated Cl− channel (Drumm et al., 1990; Rich et al., 1990; Tabcharani et al., 1991; Bear et al., 1991, 1992; Berger et al 1991; Anderson et al., 1991a,b; Dalemans et al., 1991; Rommens et al., 1991; Rich et al., 1991; Kartner et al., 1991; Tilly et al., 1992). However, a variety of other experiments indicate that it additionally performs several other functions within the cell (Barasch et al., 1991; Bradbury et al., 1992; Reisen et al., 1994; see Welsh et al., 1992, for recent review of CFTR function). A more accurate structural model of CFTR would be of great benefit in terms of elucidating its exact functions. Antibodies to different regions of CFTR would clearly help in the generation of such a model, since they would provide information concerning: (i) the topology of the protein in the membrane; and (ii) those residues exposed on the surface of the folded protein. Others have followed various procedures and raised monoclonal and polyclonal antibodies to different epitopes within CFTR (e.g. Gregory et al., 1990; Crawford et al., 1991; Cohn et al., 1991, 1992; Marino et al., 1991; Denning et al., 1992a,b; Kartner et al., 1991; Engelhardt et al., 1991; Zeitlin et al., 1992). These antibodies have been used in a variety of experiments ranging from immunoprecipitation and immunoblot analyses to immunofluorescence assays, immuno-electron microscopy and immunohistochemical

2434 J. Walker and others tissue distribution studies (see Welsh et al., 1992; Tizzano and Buchwald, 1993; and Wine, 1993, for relevant recent reviews). Many of the published antibodies appear to recognise CFTR; however, the specificity with which they do so has not always been exhaustively analysed. We have raised novel monoclonal and polyclonal antibodies to different regions of CFTR. Immunoblot, immunoprecipitation, immunofluorescence and immunohistochemical studies indicate that these antibodies recognise one or more immunologically related proteins with a similar molecular mass, calcium responsiveness and tissue distribution to CFTR. The implications of these results are discussed. MATERIALS AND METHODS Cell culture T84 (colon carcinoma) (Murakami and Masui, 1980), HT29 (colon carcinoma; Fogh and Trempe, 1975), Caco-2 (colon carcinoma; Fogh et al., 1977), PANC-1 (Leiber et al., 1975) and Heb7a (HeLa derived; Wallace et al., 1975) cells were cultured in 95% air 5% CO2 in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Life Technologies Ltd, Paisley, Scotland) supplemented with 10% foetal calf serum (FCS; Gibco), and 60 µg/ml penicillin and 100 µg/ml streptomycin (Evans Medical Ltd, Langhurst, Horsham, England). Sp2/0Ag14 cells (mouse myeloma) (Shulman et al., 1978) were cultured in 95% air 5% CO2 in RPMI 1640 (Gibco) supplemented with 10% FCS (Gibco), and 60 µg/ml penicillin and 100 µg/ml streptomycin. Hybridoma fusions were cultured in 95% air 5% CO2 in RPMI 1640 (Gibco) supplemented with 20% FCS (Gibco), 60 µg/ml penicillin and 100 µg/ml streptomycin, plus hypoxanthine, methotrexate and thymidine as previously described (Goodfellow et al., 1988). Dilution cloning of hybridomas was performed in the same medium. Established monoclonal hybridoma colonies were grown in medium in which the concentration of FCS was only 10%. Peptide synthesis and conjugation Peptides were synthesised on a Milligen 9050 peptide synthesiser and their purity determined by HPLC analysis within the SERC Molecular Recognition Centre (University of Bristol). Peptide 1 (see Fig. 1) was synthesised with an N-terminal cysteine residue and peptides 2-8 with C-terminal cysteine residues for coupling to carrier proteins. Each peptide was coupled to bovine serum albumin (BSA) via the bifunctional cross-linker SMCC (Pierce) and to thyroglobulin via the bifunctional cross-linker SM-MBS (Pierce) using previously published procedures (Green et al., 1982). Synthesis of pUEX expression constructs The following synthetic oligonucleotides were used to amplify specific regions of CFTR cDNA (using as template the plasmid construct pCOF-1, generously provided by Johanna Rommens, Department of Genetics, University of Toronto) by standard PCR techniques (Saiki, 1990). CF3, CATCGGGATCCACTTCACTTCTAATGATG; CF10, CTTGTCTGCAGAATTTCTTCACTTATTTC; CF3 and CF10 are designed to amplify a 1079 bp fragment encoding amino acids 465 to 824 of CFTR. CF5, ATGTGGGATCCAGACCAATTGAGGAAA; CF4, TGGGGCTGCAGTTCTAGTTGGCATGCTTT; CF5 and CF4 are designed to amplify a 1522 bp fragment encoding amino acids 20 to 527 of CFTR. CF9, AGCGGGGATCCTGTAAAGTGATGGCTAAC; CF8, CTTTTCTGCAGCCTTGTATCTTGCACCTC;

CF9 and CF8 are designed to amplify a 2666 bp fragment encoding amino acids 592 to 1479 of CFTR. All PCR products were purified on Chroma-spin-100 columns (Clontech) according to the manufacturer’s instructions, prior to restriction enzyme digestion and ligation into the BamHI and PstI sites of the multiple cloning site of pUEX1 (Bressan and Stanley, 1987), using previously described procedures (Sambrook et al., 1989). The oligonucleotide primers were designed so that, using these restriction enzymes, the CFTR fragments were cloned in-frame in pUEX1. Immunisation procedures Balb/c female mice were immunised with the thyroglobulin-peptide conjugate and spleen cells from immunised mice were used for fusion to Sp2/0-Ag14 mouse myeloma cells (Shulman et al., 1978) according to previously published procedures (Horn and Banting, 1994). Rabbits were immunised 3 times with SDS-PAGE purified β-galactosidase fusion proteins (Luzio et al., 1990) intramuscularly and sub-cutaneously at 14 day intervals prior to intravenous injections with SDSPAGE purified β-galactosidase fusion protein at monthly intervals. Serum samples were obtained 10 days after each intravenous injection. Screening of hybridoma supernatants Hybridoma supernatants were initially screened by ELISA (see below). Positive hybridoma supernatants were rescreened by immunoblot analysis of T84 membrane lysate and by immunofluorescence analysis of methanol fixed T84 cells (see below). A random selection of those that detected a protein of ~170 kDa in immunoblot analysis of T84 membrane lysate and/or were positive in immunofluorescence analysis of methanol fixed T84 cells were subjected to cloning by limiting dilution. Tissue culture supernatants from clones obtained by limiting dilution were re-screened using the same procedures. Only data from hybridomas which have been subjected to at least one round of limiting dilution cloning are presented. Enzyme-linked immunosorbent assays (ELISAs) Hybridoma supernatants were initially screened by ELISA against the corresponding BSA-peptide conjugate as described elsewhere (Banting, 1995). Assay plates were read using a LabSystems Multiscanplus plate reader. Supernatants were considered positive if they gave a reading equal to or greater than twice that obtained with supernatant from an irrelevant hybridoma. Immunofluorescence analysis Immunofluorescence analysis of methanol fixed cells was performed as previously described (Reaves and Banting, 1992) using either fluorescein isothiocyanate (FITC) or Texas Red-conjugated rabbit antimouse Ig (DAKO) to detect binding of the primary antibody. Preparation of membrane lysate Medium was removed from cells by aspiration. Cells were washed 5× with ice-cold PBS prior to the addition of 10 ml of ice-cold hypotonic lysis buffer (10 mM Tris-HCl, pH 7.5, 10 mM EDTA, 10 µg/ml aprotonin, 1 mM benzamidine, 5 µg/ml antipain, 5 µg/ml leupeptin, 5 µg/ml pepstatin A, 10 mM freshly prepared PMSF). Cells were then scraped from the surface of the flask (using a Costar 3010 cell lifter) and the cell suspension was passed through a 19 gauge needle attached to a 10 ml syringe 5-6 times to fragment the cells. This mixture was then centrifuged in a Sorvall RC3B at 1,000 rpm (300 g), 4°C for 5 minutes. The supernatant was then transferred to a centrifuge tube suitable for the Sorvall A814 rotor and centrifuged at 40,000 rpm, 4°C for 30 minutes in a Sorvall OTD55B ultracentrifuge. The supernatant was removed and 800 µl 2% Triton X-100 (in PBS), or 5% decanoylN-methylglucamide (MEGA-10; Sigma) in PBS, was added to the pellet, which was resuspended using a 21 gauge needle attached to a 1 ml syringe prior to transfer to a 1.5 ml Eppendorf microfuge tube. This lysate was used for immunoblot analysis and for incubation with

Cross-reactivity of antibodies to CFTR 2435 Immunohistochemistry Adult human tissues were obtained either at post mortem or from surgical specimens, and snap-frozen in liquid N2-cooled isopentane (2-methylbutane). Acetone-fixed cryostat sections were prepared and stained by an indirect immunoperoxidase staining technique exactly as described previously (Holmes et al., 1990). Negative controls include sections incubated both with an irrelevant primary antibody and in the absence of primary antibody.

antibodies in immunoprecipitation analysis. Protein concentration in the membrane lysate was measured by the Bradford test (Bradford, 1976), and 20 µg of protein was generally loaded per lane of an SDSpolyacrylamide gel. Immunoblot analysis Prior to SDS-PAGE (Laemmli, 1970) on an 8% gel, an equal volume of SDS-PAGE buffer (200 mM Tris-HCl, pH 6.8, 25% glycerol, 5% SDS, 0.1% Bromophenol Blue, 20 mM DTT) was added to each sample and the mixture incubated at 37°C for 5 minutes. Following electrophoresis, separated proteins were transferred to nitrocellulose membranes (Scleicher and Schuell) as previously described (Brake et al., 1990). Membranes were briefly stained in Ponceau S (Sigma) in order to gauge the efficency of transfer, then placed in Blotto (3% Marvel, 0.02% Tween-20 in PBS) and gently agitated at ambient temperature for at least 30 minutes to block non-specific protein binding sites. Membranes were then either cut into strips (corresponding to the lanes in the original gel) or placed in a multiblot apparatus (BioRad) for incubation with primary antibody. Neat hybridoma tissue culture supernatant or rabbit polyclonal antiserum diluted 1:500 was used for these incubations; 30-60 minutes at ambient temperature with gentle agitation. Non-bound antibody was removed by three sequential 10 minute washes at ambient temperature in 100 ml ‘Blotto’ per 20 cm2 membrane. Membranes were then incubated for 30-60 minutes at ambient temperature with gentle agitation in horseradish peroxidase (HRP)-conjugated second antibody - either rabbit anti-mouse Ig (Sigma) or swine anti-rabbit Ig (Sigma) as appropriate - diluted 1:1,000 in Blotto. Non-bound antibody was removed by three sequential 10 minute washes at ambient temperature in 100 ml Blotto per 20 cm2 membrane, followed by two sequential washes in PBS alone. Bound antibody was visualised using Nitroblue Tetrazolium (NBT) and 5-bromo-4-chloro-3-indolyl phosphate (BCIP) as substrate as previously described (Banting et al., 1985). Pre-stained Rainbow markers (Amersham) were loaded on all gels to be used in immunoblot experiments.

RT-PCR Total RNA was isolated and mRNA purified from tissue culture cells, then first strand cDNA was synthesised using Pharmacia kits (279270-01, 27-9258-01 and 27-9261-01). Subsequent PCR reactions were carried out according to standard procedures (Kawasaki 1990; Saiki, 1990) using the following pairs of oligonucleotide primers: CF7, AGACGGGATCCAAAGTGCGGCAGTACGAT; CF8, CTTTTCTGCAGCCTTGTATCTTGCACCTC; CF7 and CF8 both prime within exon 24 of CFTR and generate a fragment of 192 bp from CFTR genomic or cDNA. CFX, AGACGGGATCCAGGATAGAAGCAATGCTG; CF8 and CFX prime within exons 23 and 24 of CFTR to generate a fragment of 255 bp from CFTR cDNA. MIC2/11, TACCAGAAAAAGAGATCTTGCTTCAAAG; MIC2/13, GATCCGCCTGGGCTGTTTCTGCCGACAAT; MIC2/11 and MIC2/13 prime within exons 7 and 8 of MIC2 (Smith et al., 1993) to generate a fragment of 144 bp from MIC2 cDNA.

RESULTS Antibody production Monoclonal antibodies were raised to thyroglobulin-peptide conjugates bearing peptides corresponding to eight different regions of human CFTR (CF1-8) (following the procedures described in Materials and Methods). The sequences of the peptides, and their corresponding positions within the predicted structure of CFTR are indicated in Fig. 1. Peptides corresponding to the predicted N and C termini of mature CFTR (CF1 and CF2) as well as peptides corresponding to two predicted extracellular ‘loops’ (CF3 and CF4) and two predicted intracellular loops (CF7 and CF8) were chosen on the basis that they were hypothesised to be exposed at the surface of the protein, and would elicit antibodies which could

Immunoprecipitation For immunoprecipitation the final membrane lysate pellet was dissolved in 5% decanoyl-N-methylglucamide (Mega-10; Sigma). A 1 in 25 dilution of polyclonal antiserum in membrane lysate solution was mixed for 1 hour at room temperature. Immune complexes were collected on pepstatin A beads (which had been prewashed in membrane lysis buffer) by mixing at 4°C for 1 hour. The beads were then washed four times before elution of the immune complexes by incubation in SDS-PAGE sample buffer and 50 mM DTT at 37°C for 10 minutes. Eluted material was electrophoresed on an 8% SDS-polyacrylamide gel, blotted onto nitrocellulose membrane and tested for reactivity with monoclonal antibodies to CFTR.

3

Extracellular

4

Transmembrane Domain Intracellular

7 C 2 5/6

8 N 1

AAAA AAAA AAAA AAAA

Fig. 1. Cartoon indicating the positions of peptides CF1-8 within the sequence of CFTR. ( ) R domain; ( ) nucleotide binding domain; ( ) the transmembrane domain; and ( ) glycosylation site. Peptide sequences are: 1 (CF1), MQRSPLFKASVVSKLF; 2 (CF2), LKEETEEEVQDTRL; 3 (CF3), GRIIASYDPDNKEER; 4 (CF4), LWLLGNTPLQDKGNST; 5 (CF5), TIKENIIFGVSYDEY; 6 (CF6), TIKENIIGVSYDEY; 7 (CF7), RMMMKYRDQRAGKIS; 8 (CF8), QTSQQLKQLESEGRSP.

AA

2436 J. Walker and others be used to address the membrane topology of CFTR. Two other peptides were synthesised (CF5 and CF6). Both encompass amino acids 501-515 of the predicted mature CFTR sequence, but CF6 lacks phenylalanine 508 and thus corresponds to the sequence of CFTR found in the majority of CF individuals (Tsui, 1992). These peptides were used in the hope that they might generate antibodies which could discriminate between wild type and ∆F508 CFTR. Hybridomas producing antibodies which recognise each of the corresponding BSA-peptide conjugates (by ELISA) were generated. A random selection of these was re-screened by immunoblot analysis of T84 membrane lysate and by immunofluorescence analysis of methanol fixed T84 cells. Anti-peptide antibodies which recognised a protein of ~170 kDa in T84 membrane lysate and/or were positive in immunofluorescence analysis of methanol fixed T84 cells were identified for all except those in the anti-CF1 category. Hybridomas producing antibodies to peptides CF3, CF6, CF7 and CF8, which also recognised a protein of ~170 kDa in T84 membrane lysate and were positive in immunofluorescence analysis of methanol fixed T84, were subjected to cloning by limiting dilution. Supernatants from hybridoma clones were re-screened (as above) and shown to contain antibodies which were positive in both the ELISA and T84 assays. The antibodies produced by these clones are referred to as MabCF3, MabCF6, MabCF7 and MabCF8 hereafter. In addition to these monoclonal antibodies, polyclonal antibodies were raised in rabbits. Specific regions of CFTR cDNA were amplified by PCR and cloned into an appropriate site in the plasmid based prokaryotic expression vector pUEX1 (Bressan and Stanley, 1987) as described in Materials and Methods. This vector allows the expression of a correctly inserted cDNA as the carboxy-terminal region of a β-galactosidase fusion protein. Recombinant β-galactosidase fusion proteins were used to immunise rabbits. The antisera from immunised rabbits were designated CF5/4, CF9/8 and CF3/10 according to the construct used for immunisation (see Materials and Methods). Immunoblot analysis of HT29 membrane lysate HT29 cells have previously been shown to express CFTR (Denning et al., 1992a). We therefore chose to screen immunoblots of HT29 membrane lysate with selected monoclonal and polyclonal antibodies, anticipating that each antibody would detect a protein of ~170 kDa if it recognises

CFTR. The results are presented in Fig. 2. Each of the antibodies tested (with the exception of polyclonal antiserum CF9/8, which fails to recognise any protein) detects a protein of the same apparent molecular mass, ~170 kDa. Similar results were obtained using whole cell lysate from T84 cells (e.g. see Fig. 4). Immunofluorescence analysis of T84 cells T84 cells were processed for immunofluorescence analysis (as described in Materials and Methods), using selected monoclonal or polyclonal antibodies as the primary antibody. Each gave a diffuse pattern of staining across the cell, with a concentration of fluorescence intensity around the periphery (Fig. 3). This pattern is consistent with a predominantly plasma membrane (or immediately sub-plasma membrane) localisation. T84 cells have previously been shown to express CFTR (Denning et al., 1992a), and CFTR has previously been localised to the plasma membrane and to vesicles adjacent to the plasma membrane (Crawford et al., 1991). Peptide competition We chose to perform peptide competition experiments in order to check the specificity of the monoclonal antibodyantigen interactions we had observed. Immunofluorescence analysis of methanol-fixed T84 cells, using as primary antibody either MabCF8 alone, or MabCF8 following preincubation for 30 minutes with different BSA-peptide conjugates, demonstrated that only the immunising peptide blocks binding of antibody (Fig. 3). Similar results were obtained using MabCF2 and MabCF8, i.e. only the immunising peptide blocks binding of antibody (data not shown). In addition, an immunoblot of T84 membrane lysate was probed with either MabCF2, MabCF3 or MabCF8 alone, or with one of these antibodies following their preincubation for 30 minutes with different BSA-peptide conjugates (Fig. 4). In all cases, only the immunising peptide blocks binding of antibody to the ~170 kDa protein. Similar results were obtained with MabCF6 (data not shown). Do the monoclonal and polyclonal antibodies recognise the same molecule ? The immunoblot and immunofluorescence data presented in Figs 2-4 are consistent with the assumption that the antibodies concerned recognise CFTR. However, we felt that this point

Fig. 2. Immunoblots of HT29 membrane lysate probed with MabCF3 (lane 1), MabCF8 (lane 2), MabCF6 (lane 3), MabCF7 (lane 4), polyclonal CF3/10 (lane 6), polyclonal 5/4 (lane 8) or polyclonal 9/8 (lane 10) as primary antibody. Preimmune sera from rabbits immunised with the CF3/10 (lane 5), CF5/4 (lane 7) or CF9/8 (lane 9) were used as controls. The relative mobilities of molecular mass standards, loaded in adjacent lanes, are indicated.

Cross-reactivity of antibodies to CFTR 2437

a

b

c

d

e

f

would be strengthened if protein immunoprecipitated from T84 membrane lysate by any one of the polyclonal antisera was detected by the monoclonal antibodies in an immunoblot. The results of such an experiment are presented in Fig. 5. Polyclonal antiserum 3/10 immunoprecipitates a protein of ~170 kDa which is recognised in immunoblot anlayis by MabCF8 (lane 2). MabCF8 fails to detect anything in the lane loaded with material immunoprecipitated by preimmune serum from the rabbit which generated the 3/10 antiserum (lane 3). Another polyclonal antibody (169), previously published as being a polyclonal antibody to CFTR (Crawford et al., 1991) was included in this experiment. MabCF8 detects a protein of ~170 kDa in the material immunoprecipitated by 169 from T84 membrane lysate (lane 1). MabCF3 was used in immunoblot analysis of the same material and gave results identical to those obtained with MabCF8 (data not shown). Similar experiments were performed with polyclonal antiserum CF9/8 as the immunoprecipitating antibody, and MabCF3 and MabCF8 as the immuno-detecting antibodies; both monoclonal antibodies recognised the ~170 kDa protein immunoprecipitated by polyclonal antiserum CF9/8 (data not shown).

Fig. 3. Immunofluorescence analysis of methanol fixed T84 cells using MabCF8 (a), MabCF8 + CF8 peptide (b), MabCF8 + CF2 peptide (c), MabCF8 + CF3 peptide (d), polyclonal 3/10 (e) or preimmune serum from the rabbit which subsequently generated polyclonal 3/10 (f) as primary antibody. Bound antibody was detected with a species-specific FITCconjugated second antibody in each case.

Is expression of the protein recognised by the monoclonal and polyclonal antibodies calcium regulated? It has been shown previously that an increase in intracellular Ca2+ concentration, induced by treatment with the ionophores A23187 or ionomycin, reduces the level of CFTR expression in HT29, T84 and freshly isolated normal human bronchial epithelial cells (Bargon et al., 1992). If the antibodies we have raised recognise CFTR, then ionomycin treatment of HT29 cells should induce a decrease in the level of expression of the ~170 kDa protein recognised in immunoblot analysis of HT29 membrane lysate by those antibodies. HT29 cells were therefore incubated in the presence or absence of ionomycin for various times prior to processing for immunoblot analysis with MabCF3 as probe (Fig. 6). The level of expression of the ~170 kDa protein recognised by MabCF3 is reduced following ionomycin treatment (compare lanes 1,2 and 3). Similar results were obtained with MabCF8 (data not shown). Immunohistochemistry Since the immunoblot, immunofluorescence and Ca2+ modu-

2438 J. Walker and others

Fig. 4. Immunoblots of T84 membrane lysate probed with monoclonal antibodies in the presence of competing peptide. Antibodies were incubated in the presence or absence of competing BSA-peptide conjugate as indicated prior to incubation with the nitrocellulose membrane bearing T84 membrane lysate. Detection of bound antibody was as described in Materials and Methods. The relative mobilities of molecular mass standards, loaded in adjacent lanes, are indicated.

lation data were consistent with the hypothesis that MabCF3 and MabCF8 recognise CFTR, we chose to conduct preliminary immunolocalisation studies on cryostat sections of human tissues. We focussed particularly on MabCF8 and some of the observed patterns of reactivity are shown in Fig. 7. MabCF8 additionally recognised epithelial cells lining ducts of the pancreas and showed weak reactivity in liver (where staining was more prominent on the bile ductular rather than the hepatocyte epithelium), in kidney, where some (but not all) of the tubules were stained, and in testis, where relatively weak staining occurred throughout the germinal epithelium of seminiferous tubules (data not shown). In prostate, the antibody showed intense reactivity throughout a discrete cell population underlying the unreactive glandular epithelium (Fig. 7a, c). A more complex pattern of reactivity was observed in the endometrium (Fig. 7d-g); although MabCF8 consistently

Fig. 5. Immunoblot analysis of material immunoprecipitated from T84 membrane lysate. Polyclonal antiserum 169 (Crawford et al., 1992) (lane 1) or 3/10 (lane 2), and preimmune serum from the rabbit which subsequently generated polyclonal 3/10 (lane 3), were used to immunoprecipitate material from T84 membrane lysate. The immunoprecipitated material was subjected to SDS-PAGE, transferred to nitrocellulose and immunoblotted using MabCF8 as primary antibody. Pre-stained molecular mass standards were loaded in lane 4.

stained the myometrium (Fig. 7d), reactivity with glandular and stromal elements within the tissue varied both within and between specimens. The specimen illustrated in Fig. 7d-g, for example, contains both unreactive (Fig. 7d) and reactive (Fig. 7f) glands, and in Fig. 7g, MabCF8 positive and negative (arrowheads in Fig. 7g) cells are present in the surrounding stroma. The antibody also showed extensive reactivity in colon, where staining was present throughout the colonic crypt epithelium (Fig. 7h). Extensive reactivity was also observed with tracheal epithelium (Fig. 7j), and less pronounced staining in the underlying mucus glands (Fig. 7l). These patterns of reactivity are similar to those previously described for CFTR (e.g. see Crawford et al., 1991; Cohn et al., 1991; Marino et al., 1991; Trezise and Buchwald, 1991; Trezise et al., 1992; Tizzano et al., 1993), but there are significant differences; most notably the fact that MabCF8 decorates the whole of the colonic crypt epithelium, an observation which conflicts with those of Trezise and colleagues who clearly showed that the multidrug resistance (MDR) and CFTR

Fig. 6. Immunoblot analysis of lysate from ionomycin treated HT29 cells. MabCF3 was used as primary antibody in immunoblot analysis of HT29 membrane lysate (100 µg/lane) prepared from cells which had been incubated in the presence (lanes 2 and 3) or absence (lane 1) of 2 µg/ml ionomycin for 18 (lane 2) or 24 hours (lane 3). The relative mobilities of molecular mass standards, loaded in adjacent lanes, are indicated.

Cross-reactivity of antibodies to CFTR 2439

Fig. 7. Immunoperoxidase staining of normal human tissues with MabCF8. (a) Prostate (×63); arrowheads indicate areas of specific reactivity: GL, glandular lumen; S, stroma. (b) Prostate (×63); adjacent section negative control (irrelevant primary antibody). (c) Prostate (×630); detail from (a) to show reactivity on subglandular epithelial cells; GL, glandular lumen; S, stroma. (d) Endometrium (×50); the myometrium (M) is reactive, but adjacent glands (G) and stroma are negative. (e) Endometrium (×50); semi-adjacent section negative control. (f) Endometrium (×125); same specimen as that shown in (d); MabCF8 stains endometrial gland epithelial cells (EP) and stromal cells (S). (g) Endometrium (×400); detail from (f), arrowheads indicate MabCF8 negative stromal cells. (h) Colon (×40); C, crypts. (i) Colon (×50); negative control. (j) Trachea (×100); EP, tracheal epithelium. (k) Trachea; negative control (×100). (l) Tracheal mucus glands (MG) (×160). (m) Tracheal mucus glands (×160); negative control.

2440 J. Walker and others ionomycin for various times prior to processing for immunoblot analysis with MabCF3 as probe. The level of expression of the ~170 kDa protein recognised by MabCF3 is reduced following ionomycin treatment (data not shown). Similar results were obtained with MabCF8 (data not shown). DISCUSSION Fig. 8. Agarose gel analysis of PCR products from mRNA obtained from HT29 (lanes 1-3) and Heb7A (lanes 4-6) cell lines. Oligonucleotide primers designed to amplify a 255 bp fragment spanning exons 23 and 24 of CFTR were used to amplify the material loaded in lanes 1 and 4. A band of this size is observed in lane 1 (HT29) but not lane 4 (Heb7a), indicating the presence of CFTR mRNA in HT29 but not Heb7a cells. Contamination of cDNA with genomic DNA was investigated using a pair of oligonucleotide primers designed to amplify a 192 bp fragment within exon 24 of CFTR. The products of these reactions were loaded in lanes 2 and 5. A band of appropriate size is more intense in lane 2 (HT29) than lane 5 (Heb7a) but present in both lanes, indicating some contamination of the Heb7a cDNA with genomic DNA. The integrity of the template mRNA was tested using a pair of oligonucleotide primers designed to amplify a 144 bp fragment spanning exons 7 and 8 of the MIC2 gene (Smith et al., 1990). MIC2 encodes the E2 protein described by Gelin et al. (1989) and is expressed in all human tissues (Banting et al., 1985). A band of 144 bp is present in lanes 3 (HT29) and 6 (Heb7a), indicating that intact mRNA was present in both original samples. DNA fragments of known molecular mass were loaded in an adjacent lane.

genes have complementary patterns of epithelial expression (Trezise et al., 1992). Recognition of CFTR negative cells in tissue culture Since the pattern of tissue section immunoreactivity did not accurately reflect that previously described for CFTR, we chose to screen cell lines reported to lack CFTR expression (CFTR negative cell lines) by immunoblot and immunofluorescence analyses using the antibodies we had produced. The HeLaderived cell line Heb7a was chosen for these experiments, since others have shown, by reverse transcription PCR (RT-PCR), functional studies and immunoprecipitation analysis, that HeLa cells fail to express CFTR mRNA (Anderson et al., 1991b; Gregory et al., 1990). RT-PCR was used to confirm that the Heb7a cell line chosen for these experiments failed to produce any mRNA encoding CFTR. This was the case (Fig. 8). However, the monoclonal and polyclonal antibodies we had produced all recognised methanol fixed Heb7a cells in immunofluorescence analysis (Fig. 9), giving a pattern of staining similar to that observed when T84 cells had been used (Fig. 3). The monoclonal and polyclonal antibodies also recognised a protein of ~170 kDa in immunoblot analysis of Heb7a membrane lysate (Fig. 10). This protein was indistinguishable in size from that detected in HT29 or CaCo-2 membrane lysates (compare lanes 1, 2 and 3). Similar results were obtained using another human cell line (PANC-1) (Leiber et al., 1975) shown by RT-PCR to be CFTR negative (data not shown). Is expression of the protein recognised by the monoclonal and polyclonal antibodies in CFTR negative cells also calcium regulated? Heb7a cells were incubated in the presence or absence of

We have raised monoclonal antibodies to synthetic peptides corresponding to eight regions of CFTR (Fig. 1). Antibodies to all but one of those peptides detect a protein of ~170 kDa in immunoblot analysis of HT29 and T84 membrane lysates (Fig. 2), and recognise T84 and HT29 cells in immunofluorescence analysis (Fig. 3). We have also raised polyclonal antisera to β-galactosidase fusion proteins containing sequences corresponding to three different regions of CFTR. The polyclonal antisera detect a protein of ~170 kDa in immunoblot analysis of T84 and HT29 membrane lysates (Fig. 2), recognise T84 and HT29 cells in immunofluorescence analysis (Fig. 3) and immunoprecipitate a protein of ~170 kDa from HT29 membrane lysate, which is recognised by the monoclonal antibodies in immunoblot analysis (Fig. 5). A previously published anti-CFTR polyclonal antiserum (169; Crawford et al., 1992) also immunoprecipitates a protein of ~170 kDa, which is recognised by the monoclonal antibodies in immunoblot analysis (Fig. 5). The level of expression of the protein recogonised by these antibodies in HT29 cells is modulated by intracellular Ca2+ levels (Fig. 6), a phenomenon previously ascribed to CFTR (Bargon et al., 1992). All of these data indicate that the antibodies we have raised recognise CFTR. One of the monoclonal antibodies was therefore used in immunohistochemical screening of different human tissues (Fig. 7). It recognises epithelial cells in pancreas, gut, liver, kidney, placenta, prostate, seminiferous tubules and trachea. The gross overall pattern of protein expression detected by this antibody is similar to that previously described using either other anti-CFTR antibodies or in situ hybridisation techniques (Crawford et al., 1991; Cohn et al., 1991; Marino et al., 1991; Trezise and Buchwald, 1991; Trapnell et al., 1991; Trezise et al., 1992; Tizzano et al., 1993). However, significant differences are noticeable, e.g. (i) the entire length of the villi are stained in colonic epithelia (Fig. 7h), in contrast to the data presented by Trezise et al. (1992), which show that, in rat, CFTR is expressed at the base of the villi and multiple drug resistance protein (MDR) at the top; (ii) there is clear expression of the protein recognised by MabCF8 in adult tracheal epithelia (Fig. 7j,l), whereas it has previously proved difficult to demonstrate CFTR expression in these cells by immunocytochemical means (Crawford et al., 1991) and assays of CFTR mRNA levels have suggested that there is low level expression of the protein in the respiratory tract (Trapnell et al., 1991), (iii) there appears to be expression of the protein recognised by MabCF8 in certain non-epithelial cells in some tissues (Fig. 7d); and (iv) others have shown, by in situ hybridisation techniques, that in the rat uterus CFTR expression is limited to the epithelial lining of the endometrium, with no expression in the underlying stroma or myometrium (Trezise and Buchwald, 1991); in human tissues, MabCF8 recognises the myometrium and some stromal cells in addition to epithelial cells of the endometrium (Fig. 7d-g). These differences

Cross-reactivity of antibodies to CFTR 2441

a

b

c

d

e

f

g

h

Fig. 9. Immunofluorescence analysis of methanol fixed Heb7a cells using polyclonal antibody CF5/4 (a), polyclonal antibody 3/10 (c), polyclonal antibody 9/8 (e), MabCF3 (g) or MabCF8 (h). Preimmune sera from the rabbits immunised with the CF5/4, CF3/10 and CF9/8 constructs were incubated with cells presented in b, d and f. Bound antibody was detected with a species-specific FITC-conjugated second antibody in each case.

2442 J. Walker and others might provide a means of alleviating the symptoms of cystic fibrosis (as an alternative to directly addressing the defect in CFTR). We are grateful to Dr G. Bloomberg, Biological Sciences Research Council (BBSRC) Molecular Recognition Centre, University of Bristol, for peptide synthesis and purification, to Prof. C. F. Higgins for the generous provision of antibody 169, and to the Cystic Fibrosis Trust for financial support.

REFERENCES Fig. 10. Immunoblot analysis of membrane lysates using monoclonal or polyclonal antibodies. Heb7a (lane 1 and 4), CaCo-2 (lane 2 and 5) and HT29 (lane 3 and 6) membrane lysates were probed using MabCF3 (lanes 1-3) or MabCF8 (lanes 4-6) as primary antibody. Heb7a membrane lysates were also probed using polyclonal antibodies 3/10 (lane 7) or 5/4 (lane 8) as primary antibody. The relative mobilities of molecular mass standards, loaded in adjacent lanes, are indicated.

prompted us to ask whether the antibodies cross-reacted with other proteins within cells. CFTR is a member of the ABC transporter class of multiple membrane spanning proteins (Higgins and Hyde, 1991). This family is large, with in the order of 100 recognised prokaryotic and eukaryotic members (see Higgins, 1992, for recent review). It is not improbable that, since CFTR is an ABC transporter, anti-CFTR antibodies will recognise other members of this family. Indeed, a recently described yeast metal-resistance protein (YCF1; Szczypka et al., 1994) has striking overall sequence similarity to MDR and CFTR, including a region with strong sequence homology to the R domain of CFTR (this domain had previously been considered unique to CFTR) and a monoclonal antibody raised against a synthetic peptide corresponding to part of the CFTR sequence recognises Hlyb, an Escherichia coli member of the ABC transporter family with limited linear sequence homology to CFTR (B. Holland, personal communication*). It seems highly likely that some anti-CFTR antibodies will recognise other eukaryotic members of the ABC transporter family if immuno-cross-reactivity can occur between prokaryotic and eukaryotic members of that family. Immunoblot (Fig. 10) and immunofluorescence assays (Fig. 9) clearly show that the antibodies we have raised do recognise a protein in cells which fail to express CFTR (Fig. 8). This protein (or possibly proteins) is of a similar size to CFTR (Fig. 10) and is closely related to CFTR immunologically (as demonstrated by the fact that it is recognised by diverse monoclonal and polyclonal antibodies raised against different regions of CFTR; Fig. 9). The data we present indicate the importance of demonstrating the specificity of antibodies to CFTR. The antibodies we have generated will provide a means of identifying related members of the ABC transporter family; pharmacological manipulation of the expression/function of any one of which *Holland, B. A monoclonal antibody to the F508 region of CFTR recognises the equivalent ATPase domain of the haemolysin translocator HlyB. International Workshop on anti-CFTR antibodies, Paris, 1993, sponsored by the Association de Lutte contre la Mucovoscosidose (AFLM).

Anderson, M. P., Gregory, R. J., Thompson, S., Souza, D. W., Paul, S., Mulligan, R. C., Smith, A. E. and Welsh, M. J. (1991a). Demonstration that CFTR is a chloride channel by alteration of anion specificity. Science 253, 202-205. Anderson, M. P., Rich, D. P., Gregory, R. J., Smith, A. E. and Welsh, M. J. (1991b). Generation of cAMP-activated chloride currents by expression of CFTR. Science 251, 679-682. Banting, G. S., Pym, B. and Goodfellow, P. N. (1985). Biochemical analysis of an antigen produced by both human sex chromosomes. EMBO J. 4, 19671972. Banting, G. (1995). Antibody probes. In Gene Probes - A Practical Approach, volume I (ed. D. Hames and S. Higgins). IRL Press Practical Approach Series, IRL Press, Oxford, UK (in press). Barasch, J., Kiss, B., Prince, A., Saiman, L., Gruenert, D. and Al-Awqati, Q. (1991). Acidification of intracellular organelles is defective in cystic fibrosis. Nature 352, 70-73. Bargon J., Trapnell, B. C., Yoshimura, K., Dalemans, W., Pavirani, A., Lecocq, J.-P. and Crystal, R. G. (1992). Expression of the cystic fibrosis transmembrane conductance regulator by protein kinase C. J. Biol. Chem. 267, 16056-16060. Bear, C. E., Duguay, F., Naismith, A. L., Kartner, N., Hanrahan, J. W. and Riordan, J. R. (1991). Cl− channel activity in Xenopus oocytes expressing the cystic fibrosis gene. J. Biol. Chem. 266, 19142-19145. Bear, C. E., Li, C., Kartner, N., Bridges, R. J., Jensen, T. J., Ramjeesingh, M. and Riordan, J. R. (1992). Purification and functional reconstitution of the cystic fibrosis transmembrane conductance regulator (CFTR). Cell 68, 809-818. Berger, H. A., Anderson, M. P., Gregory, R. J., Thompson, S., Howard, P. W., Maurer, R. A., Mulligan, R. C., Smith, A. E. and Welsh, M. J. (1991). Identification and regulation of the CFTR generated chloride channel. J. Clin. Invest. 88, 1422-1431. Boat, T. F., Welsh, M. J. and Beaudet, A. L. (1989). In The Metabolic Basis of Inherited Disease (ed. C. Scriver, A. Beaudet, W. Sly and D. Valle), pp. 2649-2680. McGraw-Hill, New York. Bradbury, N., Jilling, T., Berta, T. Sorscher, E. J. Bridges, R. J. and Kirk, K. L. (1992). Regulation of plasma membrane recycling by CFTR. Science 256, 530-532. Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248-254. Brake, B., Braghetta, P., Banting, G., Bressan, G., Luzio, J. P. and Stanley, K. K. (1990). A new recombinant DNA strategy for the molecular cloning of rare membrane proteins. Biochem. J. 267, 631-637. Bressan. G. M. and Stanley, K. K. (1987). pUEX, a bacterial expression vector related to pEX with universal host specificity. Nucl. Acids Res. 15, 10056. Cohn, J. A., Melhus, O., Page, L. J., Dittrich, K. L. and Vigna, S. R. (1991). CFTR: Development of high-affinity antibodies and localization in sweat gland. Biochem. Biophys. Res. Commun. 181, 36-43. Cohn, J. A., Nairn, A. C., Marino, C. R., Melhus, O. and Kole, J. (1992). Characterization of the cystic fibrosis transmembrane conductance regulator in a colonocyte cell line. Proc. Nat. Acad. Sci. USA 89, 2340-2344. Crawford, C., Maloney, P. C., Zeitlin, P. L., Guggino, W. B., Hyde, S. C., Turley, H., Gatter, K. C., Harris, A. and Higgins, C. F. (1991). Immunocytochemical localization of the cystic fibrosis gene product CFTR. Proc. Nat. Acad. Sci. USA 88, 9262-9266. Dalemans, W., Barbry, P., Champigny, G., Jallat, S., Dott, K., Dreyer, D., Crystal, R. G., Pavirani, A., Lecocq, J. and Lazdunski, M. (1991). Altered chloride ion channel kinetics associated with ∆F508 cystic fibrosis mutation. Nature 354, 524-528.

Cross-reactivity of antibodies to CFTR 2443 Denning, G. M., Ostedgaard, L. S., Cheng, S. H., Smith, A. E. and Welsh, M. J. (1992a). Localization of the cystic fibrosis transmembrane conductance regulator in chloride secretory epithelia. J. Clin. Invest. 89, 339349. Denning, G. M., Ostedgaard, L. S. and Welsh, M. J. (1992b). Abnormal localization of cystic fibrosis transmembrane conductance regulator in primary cultures of cystic fibrosis airway epithelia. J. Cell Biol. 118, 551559. Drumm, M. L., Wilkinson, D. J., Smit, L. S., Worrell, R. T., Strong, T. V., Frizzell, R. A., Dawson, D. C. and Collins, F. S. (1991). Chloride conductance expressed by ∆F508 and other mutant CFTRs in Xenopus oocytes. Science 254, 1797-1799. Engelhardt, J. F., Allen, E. D. and Wilson, J. M. (1991). Reconstitution of tracheal grafts with a genetically modified epithelium. Proc. Nat. Acad. Sci. USA 88, 11192-11196. Fogh, J. and Trempe, G. (1975). New human tumor cell lines. In Human Tumour Cells in Vitro (ed. J. Fogh), pp. 115-119. Plenum Press, New York, USA. Fogh, J., Wright, W. C. and Loveless, J. D. (1977). Absence of HeLa cell contamination in 169 cell lines derived from human tumors. J. Nat. Cancer Inst. 58, 209-214. Gelin, C. Aubrit, F., Phalipon, A., Raynal, B., Cole, S., Kaczorek, M., Bernard, A. (1989). The E2 antigen, a 32kd glycoprotein involved in T-cell adhesion processes, is the MIC2 gene product. EMBO J. 8, 3253-3259. Goodfellow, P. N., Pritchard, C. A. and Banting, G. S. (1988). Techniques for mammalian genome transfer. In Human Genetic Diseases, Volume II (ed. K. Davies), pp. 1-18. IRL Press Practical Approach Series, IRL Press, Oxford, UK. Green, N., Alexander, H., Olsen, A., Alexander, S., Shinnick, T. M., Sutcliffe, J. G. and Lerner, R. A. (1982). Immunogenic structure of the influenza virus hemagglutinin. Cell 28, 477-487. Gregory, R. J., Cheng, S. H., Rich, D. P., Marshall, J., Paul, S., Hehir, K., Ostedgaard, L., Klinger, K. W., Welsh, M. J. and Smith, A. E. (1990). Expression and characterization of the cystic fibrosis transmembrane conductance regulator. Nature 347, 382-386. Higgins, C. F. and Hyde, S. (1991). Channeling our thoughts. Nature 352, 194-195. Higgins, C. F. (1992). ABC transporters: from microorganisms to man. Annu. Rev Cell Biol. 8, 67-113. Holmes, C. H., Simpson, K. L., Wainwright, S. D., Tate, C. G., Houlihan, J. M., Sawyer, I. H., Rogers, I. P., Spring, F. A., Anstee, D. J. and Tanner, M. J. (1990). Preferential expression of the complement regulatory protein decay accelerating factor at the fetomaternal interface during pregnancy. J. Immunol. 144, 3099-3105. Horn, M. and Banting, G. (1994). Okadaic acid treatment leads to a fragmentation of the TGN and an increase in expression of TGN38 at the cell surface. Biochem. J. 301, 69-73. Kartner, N., Hanrahan, J. W., Jensen, T. J., Naismith, A. L., Sun, S., Ackerley, C. A., Reyes, E. F., Tsui, L.-C., Rommens, J. M., Bear, C. E. and Riordan, J. R. (1991). Expression of the cystic fibrosis gene in nonepithelial invertebrate cells produces a regulate anion conductance. Cell 64, 681-691. Kerem, B., Rommens, J. M., Buchanan, J. A., Markiewicz, D., Cox, T. K., Chakravarti, A., Buchwald, M. and Tsiu, L.-C. (1989). Identification of the cystic fibrosis gene: genetic analysis. Science 245, 1073-1080. Kawasaki, E. S. (1990). In PCR Protocols (ed. M. A. Innis, D. H. Gelfand, J. J. Sninsky and T. J. White), pp. 21-27. Academic Press, London. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. Leiber, M., Mazzetta, J., Nelson-Rees, W., Kaplan, M. and Todaro, G. (1975). Establishment of a continuous tumor cell line (Panc-1) from a human carcinoma of the exocrine pancreas. Int. J. Cancer 15, 741-747. Luzio, J. P., Brake, B., Banting, G., Howell, K., Braghetta, P. and Stanley, K. K. (1990). A new immunological strategy for molecular cloning of organelle specific membrane proteins: full length sequence of TGN38, an integral membrane protein of the trans-Golgi network. Biochem. J. 270, 97102. Marino, C. R., Matovcik, L. M., Gorelick, F. S. and Cohn, J. A. (1991). Localization of the cytsic fibrosis transmembrane conductance regulator in pancreas. J. Clin. Invest. 88, 712-716. Murakami, H. and Masui, H. (1980). Hormonal control of human colon carcinoma cell growth in serum-free medium. Proc. Nat. Acad. Sci. USA 77, 3464-3468. Reaves, B. J. and Banting, G. (1992). Perturbation of the morphology of the

trans-Golgi network following brefeldin A treatment: redistribution of a TGN-specific integral membrane protein, TGN38. J. Cell Biol. 116, 85-94. Reisen, I. L., Prat, A. G., Abraham, E. H., Amara, J. F., Gregory, R. J., Ausiello, D. A. and Cantiello, H. F. (1994). The cystic fibrosis transmembrane conductance regulator is a dual ATP and chloride channel. J. Biol. Chem. 269, 20584-20591. Rich, D. P., Anderson, M. P., Gregory, R. J., Cheng, S. H., Paul, S., Jefferson, D. M., McCann, J. D., Klinger, K. W., Smith, A. E. and Welsh, M. J. (1990). Expression of cystic fibrosis transmembrane conductance regulator corrects defective chloride channel regulation in cystic fibrosis airway epithelia cells. Nature 347, 358-363. Rich, D. P., Gregory, R. J., Anderson, M. P., Manavalan, P., Smith, A. E. and Welsh, M. J. (1991). Effect of deleting the R domain on CFTRgenerated chloride channels. Science 253, 202-205. Riordan, J. R., Rommens, J. M., Kerem, B., Alon, N., Rozmahel, R., Grzelczack, Z., Zielenski, J., Lok, S., Plavsic, N., Chou, J.-L., Drumm, M. L., Ianuzzi, M. C., Collins, F. S. and Tsui, L.-C. (1989). Identification of the cystic fibrosis gene: cloning and characterization of the complementary DNA. Science 245, 1066-1073. Rommens, J. M., Iannuzzi, M. C., Kerem, B., Drumm, M. I., Melmer, G., Dean, M., Rozmahel, R., Cole, J. L., Kennedy, D., Hidaka, N., Zsiga, M., Buchwald, M., Riordan, J. R., Tsiu, L.-C. and Collins, F. S. (1989). Identification of the cystic fibrosis gene: chromosome walking and jumping. Science 245, 1059-1065. Rommens, J. M., Dho, S., Bear, C. E., Kartner, N., Kennedy, D., Riordan, J. R., Tsiu, L.-C. and Foskett, J. K. (1991). cAMP-inducible chloride conductance in mouse fibroblast lines stably expressing the human cystic fibrosis transmembrane conductance regulator. Proc. Nat. Acad. Sci. USA 88, 7500-7504. Saiki, R. (1990). In PCR Protocols (ed. M. A. Innis, D. H. Gelfand, J. J. Sninsky and T. J. White), pp. 3-12. Academic Press, London. Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edition. Cold Spring Harbour Laboratory Press, Cold Spring Harbour, New York. Shulman, M., Wilde, C. D. and Kohler, G. (1978). A better cell line for making hybridomas secreting specific antibodies. Nature 276, 269-270. Smith, M. J., Goodfellow, P. J. and Goodfellow, P. N. (1993). The genomic organisation of the human pseudoautosomal gene MIC2 and the detection of a related locus. Hum. Mol. Genet. 2, 417-422. Szczypka, M. S., Werramie, J. A., Moye-Rowley, W. S. and Thiele, D. J. (1994). A yeast metal resistance protein similar to human cystic fibrosis transmembrane conductance regulator (CFTR) and multidrug resistanceassociated protein. J. Biol. Chem. 269, 22853-22857. Tabcharani, J. A., Chang, X.-B., Riordan, J. R. and Hanrahan, J. W. (1991). Phosphorylation-regulated Cl− channel in CHO cells stably expressing the cystic fibrosis gene. Nature 352, 628-631. Tilly, B. C., Winter, M., Ostedgaard, L. S., O’Riordan, C., Smith, A. E. and Welsh, M. J. (1992). cAMP-dependent protein kinase activation of CFTR chloride channels in planar lipid bilayers. J. Biol. Chem. 267, 94709473. Tizzano, E. F. and Buchwald, M. (1992). Cystic fibrosis: beyond the gene to therapy. J. Pediat. 120, 337-349. Tizzano, E. F. and Buchwald, M. (1993). Recent advances in cystic fibrosis research. J. Pediat. 122, 985-988. Tizzano, E. F., Chitayat, D. and Buchwald, M. (1993). Cell-specific localization of CFTR mRNA shows developmentally regulated expression in human fetal tissues. Hum. Mol. Genet. 2, 219-224. Trapnell, B. C., Chu, C. S., Paakko, P. K., Banks, T. C., Yoshimura, K., Ferrans, V. J., Chernick, M. S. and Crystal, R. G. (1991). Expression of the cystic fibrosis transmembrane conductance regulator gene in the respiratory tract of normal individuals and individuals with cystic fibrosis. Proc. Nat. Acad. Sci. USA 88, 6565-6569. Trezise, A. E. O. and Buchwald, M. (1991). In vivo cell-specific expression of the cystic fibrosis transmembrane conductance regulator. Nature 353, 434437. Trezise, A. E. O., Romano, P. R., Gill, D. R., Hyde, S. C., Sepulveda, F. V., Buchwald, M. and Higgins, C. F. (1992). The multidrug resistance and cystic fibrosis genes have complimentary patterns of epithelial expression. EMBO J. 11, 4291-4303. Tsiu, L.-C. (1992). The spectrum of cystic fibrosis mutations. Trends Genet. 8, 392-398. Wallace, D. C., Bunn, C. L. and Eisenstadt, J. M. (1975). Cytoplasmic transfer of chloramphenicol resistance in human tissue culture cells. J. Cell Biol. 67, 174-188.

2444 J. Walker and others Welsh, M. J., Anderson, M. P., Rich, D. P., Berger, H. A., Denning, G. M., Ostedgaard, L. S., Sheppard, D. N., Cheng, S. H., Gregory, R. J. and Smith, A. E. (1992). Cystic fibrosis transmembrane conductance regulator: A chloride channel with novel regulation. Neuron 8, 821-829. Wine, J. J. (1993). Ion channels and transmembrane transporters. Curr. Biol. 3, 118-120.

Zeitlin, P. L., Crawford, I., Lu, L., Woel, S., Cohen, M. E., Donowitz, M., Montrose, M. H., Hamosh, A., Cutting, G. R., Gruenert, D., Huganir, R., Maloney, P. and Guggino, W. B. (1992). CFTR protein expression in primary and cultured epithelia. Proc. Nat. Acad. Sci. USA 89, 344-347. (Received 16 September 1994 - Accepted 9 February 1995)